ES2448815T3 - Structure of ribosomes and protein synthesis inhibitors - Google Patents

Abstract

A method for identifying a molecule that binds to a large ribosomal subunit, wherein the method comprises the following steps: (a) providing a molecular model comprising one or more target regions selected from the group consisting of at least a portion of: (i) an anisomycin binding site, (ii) an azithromycin binding site, (iii) a blasticidine binding site, (iv) a carbomycin binding site, (v) an erythromycin binding site, (vi) a linezólid binding site, (vii) a sparsomycin binding site, (viii) a spiramycin binding site, (ix) a tylosin binding site, and (x) a virginiamycin binding site M, where the molecular model is made (1) from the atomic coordinates corresponding to the great ribosomal subunit of Haloarcula marismortui found in Disk 1 under the file names ANISOMYCIN.PDB, azithromycin.pdb, BLASTICIDIN.PDB , CARBOMYCIN.PDB, erythromycin. pdb, linezolid.pdb, SPARSOMYCIN.PDB, SPIRAMYCIN.PDB, TYLOSIN.PDB, or VIRGINAMYCIN. PDB, or found on Disk 1, under FOLDERB, under the file names ANISOMYC.PDB, BLASTICI.PDB, CARBOMYC. PDB, SPARSOMY.PDB, SPIRAMYC.PDB, TYLOSIN.PDB, VIRGINIA.PDB, or found on Disk 1, under FOLDERC, under the file names AZITHROM.PDB, or LINEZOLI.PDB, or (2) as of of atomic coordinates derived by molecular modeling by using coordinates corresponding to the large ribosomal subunit of Haloarcula marismortui found in Disk 1 under the file names ANISOMYCIN.PDB, azithromycin. pdb, BLASTICIDIN.PDB, CARBOMYCIN.PDB, erythromycin.pdb, linezolid.pdb, SPARSOMYCIN.PDB, SPIRAMYCIN.PDB, TYLOSIN.PDB, or VIRGINAMYCIN.PDOLD, or which are under the FB names, or 1, under F FILE ANISOMYC.PDB, BLASTICI.PDB, 20 CARBOMYC.PDB, SPARSOMY.PDB, SPIRAMYC.PDB, TYLOSIN.PDB, VIRGINIA.PDB, or found on Disk 1, under FOLDERC, under the file names AZITHROM. PDB, or LINEZOLI.PDB; and (b) the use of the molecular model to identify a candidate molecule that can bind to the molecular model

Description

Structure of ribosomes and protein synthesis inhibitors

The present invention relates generally to the field of protein biosynthesis and modulators, for example, inhibitors, of protein biosynthesis. More particularly, the invention relates to: methods and compositions for elucidating the three-dimensional structure of the large ribosomal unit, either alone or in combination with an inhibitor of protein synthesis; to the three-dimensional structure of the large ribosomal subunit, either alone or in combination with an inhibitor of protein synthesis; to the use of such structures in the design and testing of new protein synthesis inhibitors; and to the new protein synthesis inhibitors.

BACKGROUND

I. Ribosomes: structure, function and composition

Ribosomes are ribonucleoproteins that are present in prokaryotes and eukaryotes. They comprise about two thirds of RNA and one third of protein. Ribosomes are the cellular organs responsible for protein synthesis. During gene expression, the ribosomes translate the genetic information encoded in a messenger RNA to the protein (Garrett et al. (2000) "The Ribosome: Structure, Function, Antibiotics and Cellular Interactions", American Society for Microbiology, Washington, D.C.).

Ribosomes comprise two subunits of non-equivalent ribonucleoproteins. The largest subunit (also known as the "large ribosomal subunit") is about twice as large in size as the smaller subunit (also known as the "small ribosomal subunit"). The small ribosomal subunit binds to messenger RNA (mRNA) and mediates in the interactions between mRNA and transfer RNA (tRNA) anticodons on which translation fidelity depends. The large ribosomal subunit catalyzes the formation of peptide bonds - the peptidyltransferase reaction of protein synthesis - and includes (at least) two different tRNA binding sites: the A site that hosts the incoming aminoacyl-tRNA, which will contribute amino acid to the growing peptide chain, and the P site that hosts the peptidyl-tRNA complex, that is, the tRNA bound to all amino acids that have been added so far to the peptide chain. The large ribosomal subunit also includes one or more binding sites for protein G factors that aid the initiation, elongation and termination phases of protein synthesis. The large and small ribosomal subunits behave independently during the initiation phase of protein synthesis; however, they are assembled as complete ribosomes when the elongation is about to begin.

The molecular weight of the prokaryotic ribosome is approximately 2.6 x 106 daltons. In prokaryotes, the small ribosomal subunit contains a 16S ribosomal RNA (rRNA) that has a molecular weight of approximately 5.0 x 105 daltons. The large ribosomal subunit contains a 23S rRNA that has a molecular weight of about 1.0 x 106 daltons and a 5S rRNA that has a molecular weight of about 4.0 x 105 daltons. The small prokaryotic subunit contains about 20 different proteins and its large subunit contains about 35 proteins. The large and small ribosomal subunits together constitute a 70S ribosome in the prokaryotes.

Eukaryotic ribosomes are generally larger than their prokaryotic counterparts. In eukaryotes, large and small subunits together form an 80S ribosome. The small subunit of a eukaryotic ribosome includes a single 18S rRNA, while the large subunit includes a 5S rRNA, a 5.8S rRNA and a 28S rRNA. The 5.8S rRNA is structurally related to the 5 'end of the prokaryotic 23S rRNA, and the 28S rRNA is structurally related to the rest of the 3S prokaryotic 23S rRNA (Moore (1998) Annu. Rev. Biophys.

27: 35-58). Eukaryotic ribosomal proteins are qualitatively similar to prokaryotic ribosomal proteins; however, eukaryotic proteins are larger and more numerous (Moore (1998) supra).

II. Structural conservation of the large ribosomal subunit

While the chemical composition of the large ribosomal subunits varies from species to species, the sequences of their components provide unequivocal evidence that they are similar in three-dimensional structure, function in a similar manner and are evolutionarily related. The evolutionary implications of the available rRNA sequence data are synthesized in the Woese et al. Articles in part II of "Ribosomal RNA. Structure, Evolution, Processing and Function in Protein Biosynthesis", (Zimmermann and Dahlberg, eds., CRC Press, Boca Raton, FL, 1996). The article by Garret and Rodriguez-Fonseca in part IV of the same volume comments on the unusually high level of sequence conservation observed in the peptidyltransferase region of the large ribosomal subunit. The ribosomes of archea species such as Haloarcula marismortu resemble those obtained from eubacterial species such as E. coli in size and complexity. However, the proteins in the ribosomes of H. marismortui are more closely related to the ribosomal proteins found in eukaryotes (Wool et al. (1995) Biochem. Cell Biol. 73: 933-947).

III. Determination of the structure of ribosomes

Much of what is known regarding the structure of ribosomes is derived from physical and chemical methods that produce relatively low resolution information. Electron microscopy (MS) has contributed to an understanding of the structure of ribosomes since the ribosome was discovered. In the 1970s, low resolution MS revealed the quaternary shape and organization of the ribosome. In the late 1980s, the positions of the surface epitopes of all proteins in the small subunit of E. coli, as well as many in the large subunit had been mapped using immunoelectronic microscopy techniques (Oakes et al. (1986) , Structure. Function and Genetics of Ribosomes, (Hardesty, B. and Kramer, G., eds.) Springer-Verlag, New York, NY, pp. 47-67; Stoeffler et al. (1986), Structure, Function and Genetics of Ribosomes, (Hardesty, B. and Kramer, G., eds.) Springer-Verlag, New York, NY, pp. 28-46). In recent years, the progress made in the cryo-MS of single particles and the reconstruction of images have led to three-dimensional reconstructions of the E. coli 70S ribosome and its complexes with tRNAs and elongation factors at resolutions between 15 A and 25 A (Stark et al. (1995) Structure 3: 815-821; Stark et al. (1997) Nature 3898: 403-406; Agrawal et al. (1996) Science 271: 1000-1002; Stark et al. ( 1997) Cell 28: 19-28). Additionally, three-dimensional MS images of the ribosome have been produced at sufficiently high resolutions so that many of the proteins and nucleic acids that help protein synthesis can be visualized attached to the ribosome. An approximate model of the RNA structure in the large subunit has been constructed to fit an electron microscopy map of 7.5 A of the 50S subunit resolution of E. coli and the available biochemical data (Mueller et al. (2000) J. Mol. Biol. 298: 35-59).

While images provided by MS have been useful, it has long been recognized that a complete understanding of the structure of ribosomes would derive only from X-ray crystallography. In 1979, Yonath and Wittman obtained the first potentially useful crystals. of ribosomes and ribosomal subunits (Yonath et al. (1980) Biochem. Internat. 1: 428-435). In the mid-1980s, scientists prepared ribosome crystals for X-ray crystallography (Maskowski et al. (1987) J. Mol. Biol. 193: 818-822). The first crystals of the 50S ribosomal subunit of H. marismortui were obtained in 1987. In 1991, improvements were reported in the resolution of the diffraction data that could be obtained from the crystals of the 50S ribosomal subunit of

H. marismortui (van Bohlen, K. (1991) J. Mol. Biol. 222: 11).

In 1995, low-resolution electron density maps were reported for large and small ribosomal subunits from halophilic and thermophilic sources (Schlunzen et al. (1995) Biochem. Cell Biol. 73: 739-749). However, these low-resolution electron density maps proved to be false (Ban et al. (1998) Cell

93: 1105-1115).

The first electronic density map of the ribosome that showed recognizable characteristics as double RNA was a 9-ray X-ray crystallographic map of the large subunit of Haloarcula marismortui (Ban et al. (1998) supra). The extension of the phase of that map to 5 A of resolution made it possible to locate several sequences of proteins and nucleic acids, whose structures had been determined independently (Ban et al. (1999) Nature 400: 841-847).

At approximately the same time, using similar crystallographic strategies, a 7.8 A resolution map of the entire Thermus thermophilus ribosome was generated showing the positions of the tRNA molecules attached to their A, P, and E sites (exit site of the protein) (Cate et al. (1999) Science 285: 2095-2104), and a 5.5 A resolution map of the 30S subunit of T. thermophilus was obtained which allowed the adjustment of the dissolved protein structures and the interpretation of some of its RNA characteristics (Clemons, Jr. et al. (1999) Nature

400: 833-840). Subsequently, a 4.5 A resolution map of the 30S subunit of T. thermophilus was published, based in part on calculated phases of a model corresponding to 28% of the mass of the subunit that had been obtained using a map 6 A experimental resolution (Tocilj et al. (1999) Proc. Natl. Acad. Sci. USA 96: 14252-14257).

IV. Location of the peptidyltransferase site in the large ribosomal subunit

It has been known for approximately 35 years that the activity of the peptidyltransferase responsible for the formation of the peptide binding that occurs during protein synthesis directed by the messenger RNA is intrinsic to the large ribosomal subunit (Traut et al. (1964) J. Mol. Biol. 10: 63; Rychlik (1966) Biochim. Biophys. Acta 114: 425; Monro (1967) J. Mol. Biol. 26: 147-151; Maden et al. (1968) J. Mol. Biol. 35: 333-345) and it has been understood for a long time that the ribosome contains proteins as well as RNA. In some species of bacteria, for example, the large ribosomal subunit contains approximately 35 different proteins and two RNAs (Noller (1984) Ann. Rev. Biochem. 53: 119-162; Wittmann-Liebold et al. (1990) The Ribosome: Structure, Function, and Evolution, (WE Hill et al., Eds.) American Society for Microbiology, Washington,

D.C. (1990), pp. 598-616). These findings raised three related issues. What of the almost 40 macromolecular components of the large ribosomal subunit contributes to your peptidyltransferase site? Where is that site located in the large subunit? How does it work?

By 1980, the list of components that could be part of the ribosome peptidyltransferase had been reduced to approximately half a dozen proteins and 23S rRNA (see Cooperman (1980) Ribosomes: Structure, Function and Genetics, (G. Chambliss et al., eds.) University Park Press, Baltimore, MD (1980), 531-554), and following the discovery of catalytic RNAs (Guerrier-Takada et al. (1983) Cell 35: 849-857; Kruger et al. (1982) Cell 31: 147-157), the hypothesis that the 23S rRNA could be its only constituent, which had been proposed years before, began to be preferred. In 1984, Noller and colleagues published affinity tagging results that showed that U2619 and U2620 (in E. coli: U2584, U2585) were adjacent to the CCA end of the tRNA attached to the P site (Barta et al. (1984) Proc. Nat Acad. Sci. USA 81: 3607-3611; Vester et al. (1988) EMBO J. 7: 3577-3587). These nucleotides appear to be part of a highly conserved internal loop at the center of domain V of the 23S rRNA. The hypothesis that this loop is intimately involved in the activity of peptidyltransferase was sustained by the observation that mutations in that loop make the cells resistant to many peptidyltransferase inhibitors, and the evidence that implies it in this activity has continued to grow. (see, Noller (1991) Ann. Rev. Biochem. 60: 191-227; Garrett et al. (1996) Ribosomal RNA: Structure, Evolution, Processing and Function in Protein Biosynthesis, (RA Zimmerman and AE Dahlberg, eds.) CRC Press, Boca Raton, FL (1996), pp. 327-355).

The definitive proof that the central loop in the V domain is the only component of the ribosome involved in peptidyltransferase activity has remained elusive. Studies have shown that it was possible to prepare particles that retained peptidyltransferase activity by increasingly vigorous deproteinizations of large ribosomal subunits, however, it was not possible to produce active particles that were completely free of proteins. However, combined with previous reconstitution results (Franceschi et al. (1990) J. Biol. Chem. 265: 6676-6682), this work reduced the number of proteins that could be involved to only two: L2 and L3 (see , Green et al. (1997) Annu. Rev. Biochem. 66: 679-716). More recently, Watanabe et al. Reported success in provoking peptidyltransferase activity from protein-free 23S rRNA, synthesized in vitro (Nitta et al. (1998) RNA 4: 257-267), however, their observations appear not to have resisted further scrutiny. Therefore, the question still remained: is the ribosome a ribozyme or not?

During the following years, the location of the peptidyltransferase site in the ribosome has been studied almost exclusively by electron microscopy. In the mid-1980s, evidence began to accumulate that there is a tunnel that runs through the large ribosomal subunit from the center of its subunit interface side to its rear (Milligan et al. (1986) Nature 319: 693-695; Yonath et al. (1987) Science 236: 813816), and there have been important reasons to believe that polypeptides pass through it as they are synthesized (Bernabeu et al. (1982) Proc. Nat. Acad Sci. USA 79: 3111-3115; Ryabova et al. (1988) FEBS Letters

226: 255-260; Beckmann et al. (1997) Science 278: 2123-2126). More recent research with cryo-MS (Frank et al. (1995) Nature 376: 441-444; Frank et al. (1995) Biochem. Cell Biol. 73: 757-765; Stark et al. (1995) supra) confirmed the existence of the tunnel and demonstrated that the CCA ends of the RNAs attached to the ribosome attached to sites A and P are located at the end of the interface of the tunnel subunit. Consequently, the peptidyltransferase site has to be located in the same position, which is at the bottom of a deep groove in the center of the surface of the subunit interface of the large subunit, immediately below its central bulge .

The substrates of the reaction catalyzed at the large subunit peptidyltransferase site are an aminoacyl-tRNA (aa-tRNA) and a peptidyl-tRNA. The first binds at the A site of the ribosome and the last one at its P site. The aa-tRNA amino group attacks carbonyl carbon by acylating the 3'-hydroxyl group of peptidyl-tRNA, and a tetrahedric intermediate in carbon of carbonyl. The tetrahedric intermediate resolves to give a peptide extended by an amino acid esterified to tRNA bound to site A and a deacylated tRNA at site P.

This reaction scheme is supported by the observations of Yarus and colleagues who synthesized an analog of the tetrahedric intermediate by attaching an oligonucleotide having the CCdA sequence to puromycin through a phosphoramide group (Welch et al. (1995) Biochemistry 34: 385-390 ). The CCA sequence, which is the 3 'terminal sequence of all tRNAs, binds to the large subunit itself, consistent with biochemical data that shows that interactions between tRNAs and the large subunit depend largely on their sequences. CCA (Moazed et al. (1991) Proc. Natl. Acad. Sci. USA 88: 3725-3728). Puromycin is an aa-tRNA analog that interacts with the ribosomal A site, and the phosphoramide group of the compound simulates the tetrahedric carbon intermediate. This transition state analogue, CCdA-phosphatopuromycin (CCdA-p-pure), binds closely to the ribosome, and inhibits its peptidyltransferase activity (Welch et al. (1995) supra).

V. Determination of the structure of macromolecules using X-ray crystallography

To better describe the efforts made to determine the structure of the ribosomes, a general synthesis of X-ray crystallography is provided below.

Each atom in a crystal disperses the X-rays in all directions, but the crystalline diffraction is observed only when a crystal is oriented relative to the X-ray beam such that the atomic dispersion interferes constructively. The orientations that lead to diffraction can be computed if the wavelength of the X-rays used and the symmetry and dimensions of the crystal cell unit are known (Blundell et al. (1976) Protein Crystallography (Molecular Biology Series), Academic Press, London). The result is that if a detector is placed behind a crystal that is being irradiated with monochromatic X-rays of an appropriate wavelength, the registered diffraction model will consist of points, each point representing one of the orientations that gives rise to the interference constructive

Each point in such a model, as recorded, is characterized by (i) an intensity (its blackness); (ii) a location, which encodes information about the orientation of the diffraction; and (iii) a phase. If all these things are known with respect to each point in a crystal diffraction model, the distribution of electrons in the crystal cell unit can be computed by the Fourier transform (Blundell et al. (1976) supra), and from that distribution or electronic density map, atomic positions can be determined.

Unfortunately, the essential phase information to compute electronic distributions cannot be measured directly from diffraction models. One of the methods routinely used to determine the phases of macromolecules, such as proteins and nucleic acids, is called multiple isomorphic replacement (MIR) that involves the introduction of new X-ray dispersers into the unit of crystal cell Typically, these additions are heavy atoms, which make a significant contribution to the diffraction model. It is important that the additions are sufficiently low in number, so that their positions can be located and leave the structure of the molecule or crystalline cell unchanged, that is, the crystals should be isomorphic. Isomorphic replacement is usually done by diffusing different heavy metal complexes in the channels of the preformed protein crystals. Macromolecules expose side chains (such as SH groups) in these solvent channels that are capable of bonding heavy metals. It is also possible to replace light endogenous metals in metalloproteins with heavier ones, for example, zinc by mercury, or calcium by samarium. Alternatively, the isomorphic derivative can be obtained by covalently attaching a heavy metal to the macromolecule in solution and then subjecting it to crystallization conditions.

Heavy metal atoms that are routinely used for isomorphic replacement include, but are not limited to mercury, uranium, platinum, gold, lead and selenium. Specific examples include mercury chloride, ethyl mercury phosphate and osmium pentamine, iridium pentamine. As such heavy metals contain many more electrons than the light atoms (H, N, C, O, and S) of the protein, heavy metals disperse the x-rays more strongly. All diffracted beams would therefore increase in intensity after the replacement of the heavy metal if the interference were positive. In fact, however, some interference is negative; consequently, after the replacement of the heavy metal, some points increase in intensity, others decrease, and many do not show a detectable difference.

Phase differences between diffracted points can be determined from changes in intensity after heavy metal replacement. First, intensity differences are used to deduce the positions of heavy atoms in the cell unit of the crystal. The Fourier summation of these intensity differences gives maps of vectors between heavy atoms, the so-called Patterson maps. From these vector maps the atomic ordering of heavy atoms is deduced. From the positions of the heavy metals in the cell unit, the amplitudes and phases of their contribution to the diffracted beams of the protein crystals containing heavy metals are calculated.

This knowledge is then used to find the phase of protein contribution in the absence of heavy metal atoms. As the phase and the amplitude of the heavy metals and the amplitude of the protein alone are known, as well as the amplitude of the protein plus the heavy metals (i.e., the heavy metal-protein complex), a phase and three amplitudes From here you can calculate the interference of X-rays dispersed by heavy metals and protein to determine if the interference is constructive or destructive. The measurement of positive or negative interference, with knowledge of the heavy metal phase, gives an estimate of the protein phase. Since two different phase angles are determined and are equally good solutions, a second full heavy metal can be used that also gives two possible phase angles. Only one of these will have the same value as one of the two previous phase angles; this therefore represents the correct phase angle. In practice, more than two different heavy metal complexes are usually performed to give a reasonably good estimate of the phase for all reflections. Each individual phase estimate contains experimental errors that arise from errors in the measured amplitudes. In addition, for many reflections the intensity differences are too small to be measured after a particular isomorphic replacement, and others can be tested.

The amplitudes and phases of the diffraction data of the protein crystals are used to calculate an electronic density map of the crystal repeating unit. This map is then interpreted to accommodate the residues of the molecule of interest. That interpretation becomes more complex due to several limitations in the data. First, the map itself contains errors, mainly due to errors in phase angles. In addition, the quality of the map depends on the resolution of the diffraction data, which in turn depends on how well the crystals are arranged. This directly influences the quality of the map that can be produced. The resolution is measured in angstrom units (A); The smaller this number, the higher the resolution and, therefore, the greater the amount of detail that can be seen.

The construction of the initial model is a trial and error process. First, it has to be decided how the chain of polypeptides or nucleic acids makes their way through the electron density map. The resulting chain trace constitutes a hypothesis whereby the density of the side chains is matched with the known sequence of the polypeptide or nucleic acid. When a reasonable chain trace has finally been obtained, an initial model is constructed that adjusts the atoms of the molecule in electron density. Computer graphics are used both for tracking the chain and for building the model to present the data and manipulate the models.

The initial model will contain some errors. If the crystals diffract at a sufficiently high resolution (for example, better than 3.5 A), most or substantially all errors can be removed by crystallographic refinement of the model using computer algorithms. In this process, the model is changed to minimize the difference between the amplitudes of diffraction observed experimentally and those calculated for a hypothetical crystal that contains the model (instead of the actual molecule). This difference is expressed as an R factor (no residual agreement) that is 0.0 for the exact agreement and approximately 0.59 for the total lack of agreement.

In general, the R factor for a well-determined macromolecular structure is preferably between 0.15 and 0.35 (such as less than about 0.24-0.28). The residual difference is a consequence of errors and imperfections in the data. These are derived from several sources, including slight variations in the conformation of protein molecules, as well as inaccurate corrections for the presence of solvent and for differences in the orientation of the microcrystals from which the crystal is constructed. This means that the final model represents an average of molecules that are slightly different in conformation and orientation.

In refined structures at high resolution, there are usually no major errors in the orientation of individual residues, and the estimated errors in the atomic positions are usually about 0.1-0.2 A, as long as the protein sequence is known or of nucleic acid. Hydrogen bonds, both within the molecule of interest and to bind ligands, can be identified with a high degree of confidence.

Typically, X-ray structures can be determined as long as the resolution is better than 3.5 A. Electronic density maps are interpreted by placing known amino acid and / or nucleic acid sequences in regions of electron density.

SAW. The need for greater resolution for the 50S ribosomal subunit

Although the art provides 50S ribosomal subunit crystals, and 9 A and 5 A X-ray crystallographic maps of 50S ribosome structure resolution, the crystals and X-ray diffraction data of the prior art are not sufficient to establish the three-dimensional structures of the 31 proteins and the 3,043 nucleotides of the 50S ribosomal subunit. Thus, crystals and prior art maps are unsuitable for design based on the structure of active agents, such as herbicides, drugs, insecticides and animal poisons.

More detailed, more detailed crystallographic X-ray maps are necessary to determine the location and three-dimensional structure of the proteins and nucleotides in ribosomes and ribosomal subunits, particularly for the 50S ribosomal subunit. An exact molecular structure of the 50S ribosomal subunit will not only make possible further investigation and understanding of the mechanism of protein synthesis, but also the development of effective therapeutic agents and drugs that modulate (for example, induce or inhibit) the synthesis of protein

SUMMARY OF THE INVENTION

The present invention is based, in part, on the determination of a high-resolution atomic structure of a ribosomal subunit, more particularly, a large subunit of a ribosome. In addition, the invention is based, in part, on the determination of high-resolution atomic structures of some protein synthesis inhibitors, namely antibiotics, when they interact with the large subunit of the ribosome.

The invention provides the structure of the large subunit of a ribosome isolated from the organism Haloarcula marismortui. However, in view of the high level of sequence and structural homology between the ribosomes of organisms in different realms, the structural information disclosed herein can be used to produce, using routine techniques, high resolution structural models of large ribosomal units. for any agency of interest.

Although there is significant homology between ribosomes of different organisms, for example, between ribosomes of humans and some human pathogens, there are still differences that can be exploited therapeutically. For example, many clinically and commercially significant protein synthesis inhibitors, for example, antibiotics such as streptomycin, tetracycline, chloramphenicol and erythromycin, selectively target bacterial ribosomes and interrupt the synthesis of bacterial proteins, but at the same time they are not directed or otherwise significantly affect human ribosomal function. As a result, over the years, antibiotics have proven valuable in the treatment of microbial infections in humans. However, there is still a need for new protein synthesis inhibitors, particularly due to the development of pathogen strains that are resistant to known antibiotics. The information provided herein provides an idea of the design of new protein synthesis inhibitors.

The invention uses computer systems that contain atomic coordinates that define at least a portion of the three-dimensional structure of a ribosome, more specifically, a large ribosomal subunit. The invention provides methods for using atomic coordinates to identify new molecules that can selectively bind ribosomes, and that preferably act as selective inhibitors of protein synthesis. In addition, the invention uses computer systems that contain atomic coordinates that define at least a portion of some antibiotics when they interact with their binding sites in the large ribosomal subunit. In addition, the invention provides methods for using the atomic coordinates of antibiotics to identify new molecules that selectively bind to ribosomes, and that preferably act as selective inhibitors of protein synthesis. Each of these aspects of the invention is discussed in more detail below.

The invention can use a computer system comprising: (a) a memory having stored in the same data indicative of atomic coordinates derived from an electronic density map having a resolution of at least about 4.5 A, more preferably of at least about 3.0 A, and more preferably even about 2.4 A and defining a ribofunctional locus of a large subunit of a ribosome; and (b) a processor in electrical communication with memory, the processor comprising a program to generate a representative three-dimensional model of the ribofunctional locus.

The computer system may further comprise a device, for example, a computer monitor, or a terminal to provide a visual representation of the molecular model. The processor may also comprise one or more programs to facilitate the rational design of the drug.

The computer system may also comprise at least a portion of the atomic coordinates deposited in the Protein Data Bank under the access number PDB ID: 1FFK (large ribosomal subunit), 1FFZ (large ribosomal subunit complexed with CCdA-p- pure), 1FG0 (large ribosomal subunit complexed with a mini-helix analogue of aminoacyl-rRNA), 1JJ2 (large ribosomal structure), 1K73 (large ribosomal subunit complexed with anisomycin), 1KC8 (large ribosomal subunit complexed with blasticidine), 1K8A ( large ribosomal subunit complexed with carbomycin), 1KD1 (large ribosomal subunit complexed with spiramycin); or 1 K9M (large ribosomal subunit complexed with tylosin) or recorded on a compact disc, Disk N.D 1 of 1.

Atomic coordinates further define at least a portion of an inhibitor of protein synthesis, for example, an antibiotic, more specifically an antibiotic selected from the group consisting of anisomycin, blasticidine, carbomycin A, sparsomycin, spiramycin, tylosin, virginiamycin M, Azithromycin, linezolide, chloramphenicol and erythromycin, complexed with a ribofunctional locus. More specifically, the invention provides atomic coordinates of the large ribosomal subunit along with the atomic coordinates of antibiotics that interact with the large ribosomal subunit. These atomic coordinates are recorded on a compact disc, Disk N.D 1, and correspond to: large ribosomal subunit complexed with anisomycin (file name: anisomysin.pdb or ANISOMYC.PDB); large ribosomal subunit complexed with blasticidine (file name: blasticidin.pdb or BLASTICI.PDB); large ribosomal subunit complexed with carbomycin (file name: carbomycin.pdb or CARBOMYC.PDB); large ribosomal subunit complexed with tylosin (file name: tylosin.pdb or TYLOSIN.PDB); large ribosomal subunit complexed with sparsomycin (file name: sparsomycin.pdb or SPARSOMY.PDB); large ribosomal subunit complexed with virginiamycin M (file name: virginiamycin.pdb or VIRGINIA.PDB); large ribosomal subunit complexed with spiramycin (file name: spiramycin.pdb or SPIRAMYC.PDB); large ribosomal subunit complexed with azithromycin (file name: AZITHROM.PDB or azithromycin.pdb); or large ribosomal subunit complexed with linezolide (file name: LINEZOLI.PDB or linezolid.pdb); or large ribosomal subunit complexed with erythromycin (file name: erythromycin.pdb).

In a preferred embodiment, the ribofunctional locus comprises at least a portion of an active site in the ribosomal subunit, for example, at least a portion of one or more of: a peptidyltransferase site (a portion of which can be defined by a plurality of waste presented in Table 5A or 5B); a site A (a portion of which can be defined by a plurality of residues presented in Table 6A or 6B); a P site (a portion of which can be defined by a plurality of residues presented in Table 7A or 7B); an exit tunnel of polypeptides (a portion of which can be defined by a plurality of residues presented in Table 8A or 8B, Table 9 or Table 10); or a domain that binds an antibiotic (a portion of which may be defined by a plurality of residues presented in Table 11A or 11B, Table 12A or 12B, Table 13A or 13B, Table 14A or 14B, Table 15A or 15B, Table 16A or 16B, Table 17A or 17B, Table 18A or 18B, Table 19A or 19B, or Table 20A or 20B). A plurality of residues should be considered to include at least 3 residues, preferably at least 5 residues, and more preferably at least 10 residues. The ribofunctional locus can be defined by atoms of ribosomal RNA, one or more ribosomal proteins, or a combination of ribosomal RNA and one or more ribosomal proteins.

In another preferred embodiment, the atomic coordinates are produced by molecular modeling. Using the atomic coordinates provided herein, the person skilled in the art can generate models of any ribosome of interest using conventional techniques, for example, conventional homology modeling, and or molecular replacement techniques. In another embodiment, the atomic coordinates are produced by homology modeling using at least a portion of the atomic coordinates deposited in the Protein Data Bank under the access number PDB ID: 1FFK, 1FFZ, 1FG0, 1JJ2, 1K73, 1KC8, 1K8A, 1KD1 or 1K9M, or any of the atomic coordinates included in the compact disk, Disk ND 1. In another embodiment, the atomic coordinates are produced by molecular replacement using at least a portion of the atomic coordinates deposited in the Data Bank of Proteins under the PDB ID access number: 1FFK, 1FFZ, 1FG0, 1JJ2, 1K73, 1KC8, 1K8A, 1KD1 or 1K9M, or any of the atomic coordinates recorded on the compact disc, Disk ND 1.

In a preferred embodiment, the atomic coordinates define residues that are conserved between ribosomes or ribosomal subunits of pathogens, for example, prokaryotic organisms, and, optionally, but more preferably, are also absent in ribosomes or ribosomal subunits of a host organism, for example , a human being. In another preferred embodiment, the atomic coordinates can define residues that are conserved between ribosomes or ribosomal subunits of prokaryotic organisms, for example, bacteria, and, optionally, but more preferably, are also absent in eukaryotic ribosomal subunits, for example, a mammal , more preferably, a human being. This information can be used, for example, through the use of one or more molecular models, to identify targets for rational drug design that can be exploited to develop new molecules, for example, protein synthesis inhibitors, which disrupt the synthesis of proteins in a pathogen, for example, a bacterium, but does not substantially interrupt or otherwise affect the synthesis of proteins in a host organism, for example, a human being.

In another aspect, the invention provides a variety of methods for designing, testing and refining new molecules through the rational design of drugs. For example, the invention provides a method that comprises the steps of:

(a) provide a model, for example, a molecular model, which has a ribofunctional locus of a large subunit of a ribosome, where the model is defined by the spatial ordering of atoms derived from an electron density map that has a resolution at least about 4.5 A, more preferably at least about 3.0 A, and more preferably even about 2.4 A; and (b) using the model to identify a candidate molecule that has a surface complementary to the ribofunctional locus. Preferably, the candidate molecule is stereochemically adjusted, and more preferably, is attached to the ribofunctional locus of the large subunit of the ribosome.

In a preferred embodiment, the method comprises one or more additional steps of: producing the candidate molecule identified in a method like this; determine whether the candidate molecule, when produced, modulates (for example, induces or reduces) ribosomal activity; identifying a modified molecule; produce the modified molecule; determine if the modified molecule, when produced, modulates ribosomal activity; and produce the modified molecule to be used either alone or in combination with a pharmaceutically acceptable vehicle. The candidate molecule and / or the modified molecule may be an antibiotic or an antibiotic analog, for example, a macrolide antibiotic or a macrolide analog.

In a preferred embodiment, the ribofunctional locus used in a method such as this comprises at least a portion of an active site in the ribosomal subunit. In another preferred embodiment, the ribofunctional locus is defined by at least a portion of one or more of: a peptidyltransferase site (a portion of which may be defined by a plurality of residues presented in Table 5A or Table 5B); a site A (a portion of which can be defined by a plurality of residues presented in Table 6A or Table 6B); a P site (a portion of which can be defined by a plurality of residues presented in Table 7A or Table 7B); an exit tunnel of the polypeptide (a portion of which can be defined by a plurality of residues presented in Table 8A, Table 8B, Table 9, Table 10); or a domain that binds an antibiotic (a portion of which may be defined by a plurality of residues presented in Table 11A, Table 11B, Table 12A, Table 12B, Table 13A, Table 13B, Table 14A, Table 14B, Table 15A, Table 15B, Table 16A, Table 16B, Table 17A, Table 17B, Table 18A, Table 18B, Table 19A, Table 19B, Table 20A or Table 20B). The ribofunctional locus can be defined by atoms of ribosomal RNA, one or more ribosomal proteins, or a combination of ribosomal RNA and one or more ribosomal proteins.

In another preferred embodiment, the atomic coordinates are used to produce a molecular model in an electronic form. Atomic coordinates are preferably produced by molecular modeling. In another embodiment, the atomic coordinates are produced by homology modeling using at least a portion of the atomic coordinates deposited in the Protein Data Bank under the PDB ID access number: 1FFK, 1FFZ, 1FG0, 1JJ2, 1K73, 1KC8, 1K8A, 1KD or 1 K9M, or the atomic coordinates recorded on a compact disc, Disk ND 1. In another embodiment, the atomic coordinates are produced by molecular replacement using at least a portion of the atomic coordinates deposited in the Protein Data Bank under the PDB ID access number: 1FFK, 1FFZ, 1FG0, 1JJ2, 1K73, 1KC8, 1K8A, 1KD1 or 1K9M or any of the atomic coordinates included in the compact disc, Disk ND 1.

In a preferred embodiment, atomic coordinates may define residues that are conserved between ribosomes or ribosomal subunits of pathogens, for example, prokaryotic organisms, and, optionally, but more preferably, are also absent in ribosomes or ribosomal subunits of an organism. host, for example, a human being. In another preferred embodiment, the atomic coordinates may define residues that are conserved between ribosomes or ribosomal subunits of prokaryotic organisms, for example, bacteria, and, optionally but more preferably, are also absent from ribosomes or ribosomal eukaryotic subunits, for example, a Mammal, more preferably a human being. This information can be used, for example, by the use of one or more molecular models, to identify targets for rational drug design that can be exploited to develop new molecules, for example, protein synthesis inhibitors, which disrupt the protein synthesis in a pathogen, for example, a bacterium, but does not substantially interrupt or otherwise substantially affect protein synthesis in a host organism, for example, a human being.

In a preferred embodiment, the atomic coordinates further define at least a portion of a protein synthesis inhibitor, for example, an antibiotic, more specifically an antibiotic selected from the group consisting of anisomycin, blasticidine, carbomycin A, sparsomycin, spiramycin , tylosin, virginiamycin M, azithromycin, linezolide and erythromycin, complexed with a ribofunctional locus. In a preferred embodiment, the ribofunctional locus comprises at least a portion of an active site of a ribosome, for example, at least a portion of one or more of (i) a peptidyltransferase site, (ii) a site A, (iii ) a P site, (iv) an exit tunnel of polypeptides.

More specifically, the invention provides atomic coordinates of antibiotics that interact with the large ribosomal subunit. These atomic coordinates are recorded on a compact disc, Disk N.D 1, and correspond

a: the large ribosomal subunit complexed with anisomycin (file name: anisomysin.pdb or ANISOMYC.PDB); the large ribosomal subunit complexed with blasticidine (file name: blasticidin.pdb

or BLASTICI.PDB); the large ribosomal subunit complexed with carbomycin A (file name: carbomycin.pdb or CARBOMYC.PDB); the large ribosomal subunit complexed with tylosin (file name: tylosin.pdb or TYLOSIN.PDB); the large ribosomal subunit complexed with sparsomycin (file name: sparsomycin.pdb or SPARSOMY.PDB); the large ribosomal subunit complexed with virginiamycin M (file name: virginiamycin.pdb or VIRGINIA.PDB); the large ribosomal subunit complexed with spiramycin (file name: spiramycin.pdb or SPIRAMYC.PDB); the large ribosomal subunit complexed with azithromycin (file name: AZITHROM.PDB or azithromycin.pdb); or the large ribosomal subunit complexed with linezolide (file name: LINEZOLI.PDB or linezolid.pdb); or the large ribosomal subunit complexed with erythromycin (file name: erythromycin.pdb).

In another aspect, the invention provides a method for identifying a principal candidate for a new protein synthesis inhibitor. The method comprises the steps of (a) providing a molecular model of at least one portion of a protein synthesis inhibitor when the protein synthesis inhibitor is interacting with a ribofunctional locus of a large subunit of a ribosome; and (b) use the model to identify the main candidate. In a preferred embodiment, the main candidate is able to interact, and even more preferably, join a ribofunctional locus.

In another preferred embodiment, the method comprises the additional step of producing the main compound. After the synthesis, the main compound can be tested for biological activity, for example, modulation of ribosomal activity in an in vitro assay or inhibition of the growth of microorganisms of interest. Based on the results of such studies, it is possible to determine the structure-activity relationships, which can then be used to disseminate other modifications of the main compound to improve a particular characteristic of interest. The modified main compounds can then be produced and evaluated with respect to biological activity, as before. Once a compound of interest has been designed, synthesized and tested for activity, it can then be produced in commercially feasible quantities for use as a pharmaceutical product.

In a preferred embodiment, the starting molecule (ie, the protein synthesis inhibitor), the main compound and the penultimate compound is an antibiotic or an antibiotic analog. In another embodiment, the main compound and also the penultimate compound is a hybrid antibiotic (ie, it comprises a portion of a first antibiotic and a portion of a different second antibiotic).

In another preferred embodiment, the model used in the practice of the invention is a model of an antibiotic selected from the group consisting of anisomycin, blasticidine, carbomycin A, sparsomycin, spiramycin, tylosin, virginiamycin M, azithromycin, linezolide, or erythromycin. The model preferably comprises a portion of the atomic coordinates recorded on the compact disc, Disk ND 1 of 1 under the file name: anisomycin.pdb, blasticidin.pdb, carbomycin.pdb, sparsomycin.pdb, spiramycin.pdb, tylosin.pdb , virginiamycin.pdb, ANISOMYC.PDB, BLASTICI.PDB, CARBOMYC.PDB, SPARSOMY.PDB, SPIRAMYC.PDB, TYLOSIN.PDB, VIRGINIA.PDB, AZITHROM.PDB, LINEZOLI.PDBD, azithd, line, azd, line erythromycin.pdb. In another preferred embodiment, the penultimate compound is an analog of an antibiotic selected from the group consisting of anisomycin, blasticidine, carbomycin A, sparsomycin, spiramycin, tylosin, virginiamycin M, azithromycin, linezolide or erythromycin.

In a preferred embodiment, the ribofunctional locus comprises at least a portion of an active site of a ribosome, for example, at least a portion of one or more of (i) a peptidyltransferase site, (ii) a site A, (iii ) a P site, and (iv) a polypeptide exit tunnel. In another embodiment, the molecular model useful in the practice of the invention is an electronic form, in which case the molecular model is preferably generated by molecular modeling.

In another aspect, the invention provides new protein synthesis inhibitors that disrupt the function of an objective ribosome. These inhibitors can be designed and tested quickly as disclosed herein.

One type of protein synthesis inhibitor of the invention comprises: a first binding domain having a surface, for example, a surface accessible to a solvent, that mimics or duplicates a surface of a first known molecule, for example, a first antibiotic, which binds to a first contact site, for example, a first ribofunctional locus, in or on a large ribosomal subunit; and a second binding domain having a surface, for example, a surface accessible to a solvent, that mimics or duplicates a surface of a second known molecule, for example, a second antibiotic, which binds with a second contact site, for example, a second ribofunctional locus, in or on the ribosomal subunit. The first domain is linked to the second domain in such a way as to allow the first domain and the second domain to join simultaneously with their respective contact sites in or on the ribosomal subunit so as to interrupt the synthesis of proteins in a ribosomal subunit. In a preferred embodiment, the protein synthesis inhibitor has a molecular weight of less than about 1,500 and an IC50 of less than about 50 µM, more preferably less than about 10 µM.

Another type of protein synthesis inhibitor is a synthetic molecule, created by genetic engineering, comprising: (i) a binding domain that has a surface, for example, a surface accessible to a solvent, that mimics or duplicates a surface accessible to a solvent of a known molecule, for example, a first known antibiotic, which binds with a contact site, for example, a ribofunctional locus in or on a ribosomal subunit; and (ii) a new effector domain linked to the binding domain, which, after the union of the binding domain with the contact site, occupies a space inside or adjacent to the ribosomal subunit, thereby interrupting protein synthesis in the ribosomal subunit. In a preferred embodiment, the protein synthesis inhibitor has a molecular weight of less than about 1,500 and an IC50 of less than about 50 µM, more preferably less than about 10 µM.

The method of the invention can provide new protein synthesis inhibitors, for example, a molecule capable of contacting at least three residues, but less than thirteen contact residues in Table 11A that together define an anisomycin binding pocket of a large ribosomal subunit, a molecule capable of contacting at least three residues, but less than ten contact residues in Table 12A, which together define a blasticidine binding pocket of a large ribosomal subunit, a molecule capable of contacting at least three residues , but less than sixteen contact residues in Table 13A that together define a carbomycin A junction pocket of a large ribosomal subunit, a molecule capable of contacting at least three residues, but less than twenty contact residues in Table 14A that together they define a tylosin binding pocket of a large ribosomal subunit, a molecule capable of contacting at least three residues, but less than nine re contact wastes in Table 15A that together define a sparsomycin binding pocket of a large ribosomal subunit, a molecule capable of contacting at least three residues, but less than thirteen contact residues in Table 16A that together define a binding pocket of virginiamycin M of a large ribosomal subunit, a molecule capable of contacting at least three residues, but less than fifteen contact residues in Table 17A which together define a spiramycin binding pocket of a large ribosomal subunit, a molecule capable of contacting at least three residues, but less than thirteen contact residues in Table 18A that together define an erythromycin binding pocket of a large ribosomal subunit, a molecule capable of contacting at least three but less than eleven contact residues in Table 19A which together define an azithromycin binding pocket of a large ribosomal subunit, or a molecule capable of contacting at least three residues p It was less than fifteen contact residues in Table 20A that together define a linezolide junction pocket of a large ribosomal subunit. Contact residues are those residues in the large ribosomal subunit that are in contact with van der Waals with the antibiotic of interest.

In another aspect, the method provides a protein synthesis inhibitor comprising, for example, a molecule capable of contacting a plurality of residues in Table 11A that together define an anisomycin binding pocket of a large ribosomal subunit, but which they do not have one or more atoms present in the anisomycin molecule, whose atomic coordinates are recorded on Disk ND 1 under the file name ANISOMYC.PDB; a molecule capable of contacting a plurality of residues in Table 12A that together define a blasticidine junction pocket of a large ribosomal subunit, but that do not have one or more atoms present in the blasticidine molecule, whose atomic coordinates are engraved on the Disk ND 1 under the filename BLASTICI.PDB; a molecule capable of contacting a plurality of residues in Table 13A that together define a carbomycin A junction pocket of a large ribosomal subunit, but that do not have one or more atoms present in carbomycin A, whose atomic coordinates are engraved on the Disk ND 1 under the filename CARBOMYC.PDB; a molecule capable of contacting a plurality of residues in Table 14A that together define a tylosin binding pocket of a large ribosomal subunit, but which do not have one or more atoms present in tylosin, whose atomic coordinates are recorded on Disk ND 1 under the filename TYLOSIN.PDB; a molecule capable of contacting a plurality of residues in Table 15A that together define a sparsomycin binding pocket of a large ribosomal subunit, but that do not have one or more atoms present in the sparsomycin, whose atomic coordinates are recorded on the ND Disk 1 under the filename SPARSOMY.PDB; a molecule capable of contacting a plurality of residues in Table 16A that together define a virginiamycin M junction pocket of a large ribosomal subunit, but that do not have one or more atoms present in virginiamycin M, whose atomic coordinates are engraved on the Disk ND 1 under the filename VIRGINIA.PDB; a molecule capable of contacting a plurality of residues in Table 17A that together define a spiramycin binding pocket of a large ribosomal subunit, but that do not have one or more atoms present in the spiramycin, whose atomic coordinates are recorded on the ND Disk 1 under the filename SPIRAMYC.PDB; a molecule capable of contacting a plurality of residues in Table 18A that together define an erythromycin binding pocket of a large ribosomal subunit, but which do not have one or more atoms present in the erythromycin, whose atomic coordinates are recorded on the ND Disk 1 under the filename erythromycin.pdb; a molecule capable of contacting a plurality of residues in Table 19A that together define an azithromycin binding pocket of a large ribosomal subunit, but which do not have one or more atoms present in azithromycin, whose atomic coordinates are recorded on Disk ND 1 under the filename azithromycin.pdb; or a molecule capable of contacting a plurality of residues in Table 20A that together define a linezolide junction pocket of a large ribosomal subunit, but that do not have one or more atoms present in the linezolide, whose atomic coordinates are engraved on the Disc ND 1 under the file name linezolid.pdb.

The foregoing aspects and embodiments of the invention can be better understood by reference to the following figures, to the detailed description and to the claims. Other advantages are evident from the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

The patent or application file contains at least one drawing made in color. Copies of this patent or patent application publication with color drawing (s) will be provided by the Office upon request and after payment of the necessary fee.

Color presentations similar to some of the following figures can be found, for example, in Ban et al. (2000) Science 289: 905-920; or Nissen et al. (2000) Science 289: 920-930.

Color figures are also presented in gray scales with the same figure number followed by an apostrophe.

The objects and features of the invention can be better understood by reference to the drawings described below:

Figures 1 (A) - (E) show the electronic density of an electronic density map of 2.4 A resolution. Specifically, Figure 1 (A) shows a stereo view of a junction between domains II, III and IV of 23S rRNA. Figure 1 (B) shows the extended region of the L2 protein that interacts with the surrounding RNA. Figure 1 (C) shows in detail the region L2 with a bound Mg2B ion. Figure 1 (D) shows in detail L2 with amino acid side chains. Figure 1 (E) shows helices 94-97 of domain 6.

Figure 2 shows the large ribosomal subunit of H. marismortui in the crown view. The subunit is shown in the crown view, with its stem L7 / L12 on the right, its stem L1 on the left, and its central bulge (CP) above. In this view, the surface of the subunit that interacts with the small ribosomal subunit faces the reader. RNA is shown in gray in a presentation that fills the space. The skeletons of the visible proteins are shown in gold. An analogue of the transition state attached to the peptidyltransferase site of the subunit is indicated in green. The particle is approximately 250 A across.

Figures 3 (A) - (B) show the secondary structure of the 23S rRNA of H. marismortui. The secondary structure of this 23S rRNA is shown in a standardized format. Figure 3 (A) shows the 5 'half of the large rRNA subunit. Figure 3 (B) shows the 3 'half of the large subunit of rRNA. This diagram shows all base pairings seen in the large subunit crystal structure that are stabilized by at least two hydrogen bonds. The pairings shown in red were predicted and observed. Those shown in green were predicted, but not observed. The interactions shown in blue are observed, but were not predicted. The bases shown in black do not appear to be involved in mating interactions. Sequences that cannot be visualized on the 2.4A resolution electronic density map are illustrated in gray with the secondary structures predicted for them.

Figures 4 (A) - (L) show the tertiary structures of the RNA domains in the large ribosomal subunit of H. marismortui, its RNA as a whole, and schemes of its RNAs. Specifically, Figures 4 (A) and 4 (B) show the RNA structure of the entire subunit. The domains are color coded as shown in the scheme of Figure 5 (C). Figure 4 (A) shows the particle in the crown view. Figure 4 (B) shows the image in Figure 4 (A) rotated 1800 around an axis that runs vertically in the image plane. Figures 4 (C) and 4 (D) show a schematic diagram of 23S rRNA and the secondary structure of 5S rRNA. Figure 4 (C) shows a schematic diagram of the secondary structure of 23S rRNA of Figure 3 with numbered helices according to Leffers et al. ((1987) J. Mol. Biol. 195: 43-61), and the domains of the molecule are indicated by color tones. Figure 4 (D) shows the secondary structure of 5S rRNA from H. marismortui. Thick lines joining bases represent the Watson-Crick mating. The bases joined by a bottom locker "or" indicate no Watson-Crick mating. The bases joined by thin lines interact through a single hydrogen bond. The bases shown in black are not paired. The bases shown in red are phylogenetically predicted pairings that have now been confirmed (Symanski et al. (1998) Nucl. Acids Res. 26: 156-159). The pairs shown in blue are observed, but were not predicted, and the pairs shown in green were predicted but not observed. Figures 4 (E) to 4 (L) show stereo views of the RNA domains in the 23S rRNA and / or 5S rRNA. Each domain has a color code from its 5 'end to its 3' end to facilitate the observer by following its path in three dimensions. The surfaces where the most important inter-domain interactions occur are shown in mono to the right of the stereo views. Figure 4 (E) shows domain I; Figure 4 (F) shows domain II; Figure 4 (G) shows domain III; Figure 4 (H) shows domain IV; Figure 4 (I) shows domain V, seen in crown; Figure 4 (J) shows domain V, rear view; Figure 4 (K) shows domain VI; and Figure 4 (L) shows the 5S rRNA.

Figures 5 (A) - (C) show conservations and expansions in the 23S rRNA of H. marismortui. The generality of RNA in these images is gray. Sequences found to be> 95% conserved in the three phylogenetic kingdoms are shown in red. Sequences where expansion in the basic structure 23S is allowed are shown in green (Gutell et al. (2000) supra). Specifically, Figure 5 (A) shows the rotated particle with respect to the crown view so that the cleft of its active site can be seen. Figure 5 (B) shows the crown view. Figure 5 (C) shows the rear view of the particle, that is, the crown view rotated 1800 around its vertical axis.

Figures 6 (A) - (I) show structures of some large subunit ribosomal proteins that have non-globular extensions. Only protein skeletons are shown. The globular domains of these proteins are shown in green and their non-globular extensions are illustrated in red. The positions of the zinc ions in L44e and L37e are also indicated. Figure 6 (A) shows L2; Figure 6 (B) shows L3; Figure 6 (C) shows L39; Figure 6 (D) shows L4; Figure 6 (E) shows L15; Figure 6 (F) shows L21e; Figure 6 (G) shows L44e; Figure 6 (H) shows L37e; Figure 6 (I) shows L19;

Figures 7 (A) - (C) show proteins that appear on the surface of the large ribosomal subunit. The RNA of the subunit is shown in gray, as in Figure 2, and the skeletons of the proteins are shown in gold. Specifically, Figure 7 (A) shows the subunit in the crown view of the subunit. Figure 7 (B) shows the rear side of the subunit in the orientation of the crown view. Figure 7 (C) shows the bottom view; The end of the peptide tunnel appears in the center of this image. The proteins visible in each image are identified in the small images in the lower left corner of the Figure.

Figures 8 (A) - (F) show the distribution of proteins and protein-RNA interactions in the large ribosomal subunit. Specifically, Figure 8 (A) shows the structures of the proteins in the near end of the peptide tunnel and how they relate to the RNA sequences with which they interact. The L22 protein extends a long fork extension � within the 23S rRNA. L24 has a similar extension but the whole protein is on the surface of the particle. L39 is the only protein in the subunit that has no tertiary structure, while L37e has both terminal extensions NH2 and COOH. L19 is unique because it has two globular domains on the surface of the subunit connected by an extended sequence that extends through the RNA. The end of L39 (green) actually enters the tunnel, while L37e (red) is completely surrounded by RNA. Figure 8 (B) shows the non-globular extensions of L2 and L3 that reach through the mass of 23S rRNA towards the peptidyltransferase site, which is marked by a CCdA-ppuromycin molecule. Figure 8 (C) shows L22 interacting with portions of six of the domains of 23S rRNA. Figure 8 (D) shows a 23S rRNA scheme showing the locations of the sequences that make at least one van der Waals contact with the protein (red). Figure 8 (E) shows a stereo view of the proteins of the large ribosomal subunit in which all RNA was removed. Proteins are red as an aid to visualization only. Figure 8 (F) shows a cross section of the subunit in the tunnel exit area. The L22 protein is shown as mofos in red, and the hairpin loop � where mutations confer resistance to erythromycin is shown in orange. Atoms on the surface are shown in gray, protein atoms are shown in green, and atoms at the interface of the cut are shown in blue.

Figures 9 (A) - (C) show chemical structures of substrates and analogs of ribosome peptidyltransferase. Specifically, Figure 9 (A) shows the tetrahedric carbon intermediate produced during the formation of the peptide bond; Tetrahedric carbon is indicated by an arrow. Figure 9 (B) shows the transition state analog formed by coupling the 3'OH of CCdA to the amino group of the puromycin Ometiltyrosine residue via a phosphate, CCdA-p-pure group (Welch et al. (1995) supra ). Figure 9 (C) shows an amino-N-acylated mini-helix constructed to be directed to site A. The oligonucleotide sequence 5 'phosphate CCGGCGGGCUGGUUCAAACCGGCCCGCCGGACC 3' (SEQ ID NO: 1) puromycin should form 12 base pairs. The construct was based on a mini-helix that is a suitable substrate for amino-acylation by Tyr-tRNA synthetase. The 3'OH of its C terminal is coupled to the 5'OH of the N6-dimethyl A part of the puromycin by a phosphodiester junction.

Figures 10 (A) - (C) show experimentally phased electron density maps of the analog complexes of the substrate at a resolution of 3.2 A, with overlapping models (oxygen, red; phosphorus, purple; nitrogen, blue; and carbon, green for rRNA and yellow for the substrate). Specifically, Figure 10 (A) shows an electronic density map of Fo (complex) -Fo (original) difference with a superimposed CCdA-ppuro skeleton model. Figure 10 (B) shows a 2Fo (complex) -Fo (original) electronic density map of the pure-CCdAp in the region of the active site with the overlapping ribosome and inhibitor structures showing the proximity of A2486 N3 (2451) to phosphate oxygen, not forming a bridge in this complex. Figure 10 (C) shows an electronic density map with differences Fo (complex) - Fo (original) of the trunk analogue of the tRNA acceptor with an overlapped CCpuro skeleton model. There is only density for ribose and C74 phosphate and none for the rest of the RNA hairpin.

Figures 11 (A) and (B) show a combined model of the CCA portion of the mini-helix attached to site A and CCdAp-pure linked to sites A and P, with color coding as in Figure 2. Specifically, Figure 11 (A) shows the base pairing interactions between site P C74 and C75 and the P loop of the 23S rRNA on the left and the A C75 site with the A loop of the 23S rRNA on the right. The catalytic A2486 is close to phosphate (P) oxygen which is the analog of the tetrahedric intermediate oxianion. Figure 11 (B) shows A2637 (in blue) that lies between the two CCAs and A2486 (green) whose N3 approximates a phosphate oxygen that does not form a bridge. The N1 atoms of the A76 bases of the A and P site tRNAs perform almost identical interactions with a 2'OH ribose in both loops, A and P, respectively, and an axis of about 2 times relates these residues.

Figure 12 shows a space filling model of the 23S and 5S rRNA, the proteins and the CCdA-ppuro inhibitor seen down from the slit of the active site in a rotated "crown view". The bases are white and the skeletons of the sugar phosphate are yellow. The inhibitor is shown in red and the numbered proteins are shown in blue. The L1 and L11 proteins positioned at a lower resolution are in the blue skeleton. The central protuberance is marked CP.

Figure 13 (A) shows a stereo view diagram of the three-dimensional distribution of the residues comprising the A and P loops and the peptidyltransferase loop. Figure 13 (B) shows a stereo view of the central loop in domain V from the tunnel direction. The residues are color-coded based on the mutations that confer resistance to the antibiotic. Figure 13 (C) shows the active site of domain V with its central loop shown as the secondary structure.

Figures 14 (A) and (B) show the closest approach of the polypeptides to the active peptidyltransferase site marked by a representation of balls and sticks of the Yarus inhibitor, CCdA-p-pure. Specifically, Figure 14 (A) shows a spiral representation of the RNA skeleton of the V domain in red and gray bases and a representation of the skeleton in bands of the thirteen proteins that interact with it. Figure 14 (B) shows a close view of the active site where the RNA was removed. The phosphate of the Yarus analogue and the proteins whose extensions are closer to the inhibitor are shown in bands with their side chains closer in a representation of the atom. The distances in A between the closest protein atoms and the phosphorus analog of tetrahedric carbon (pink) are shown, as well as a modeled peptide (pink).

Figure 15 shows nucleotides conserved in the region of the peptidyltransferase that joins the representation of the filling of CCdA-p-pure spaces A of the region of the active site with the Yarus inhibitor seen down the slit of the active site. All atoms that belong to the nucleotides of the 23S rRNA that are 95% conserved in the three kingdoms (Gutell et al. (2000) supra) are colored in red and all other nucleotides in white; The inhibitor is colored blue.

Figures 16 (A) - (C) show the catalytic apparatus of the active site of peptidyltransferase. Specifically, Figure 16 (A) shows a stereo view of a portion of the 2.4 A electronic density map of experimental resolution (Ban et al. (2000) Science 289: 905-920) of the large subunit in the region of the stereo catalytic site. The structure of the RNA involved in interactions with A2486 is superimposed. Residues G2102 (2061) and G2482 (2447) are bound by hydrogen to N6 of A2486 (2451) and G2482 that interacts with a neighboring phosphate group. Figure 16 (B) shows a representation of the skeleton with hydrogen bonds with dashed lines showing G2482, G2102, A2486 and the incorporated phosphate that is proposed to result in a charge forwarding through G2482 to N3 of A2486. Figure 16 (C) shows the rarest normal and imine tautomeric forms of G2482 and A2486 that are proposed to stabilize by the incorporated 2485 phosphate residue.

Figures 17 (A) - (C) show the proposed mechanism of peptide synthesis synthesized by the ribosome. Specifically, Figure 17 (A) shows the N3 of A2486 subtracting a proton from the NH2 group since the latter attacks the carbonyl carbon of the peptidyl-rRNA. Figure 17 (B) shows a protonated N3 that stabilizes the tetrahedric carbon intermediate by hydrogen bonding to the oxyanion. Figure 17 (C) shows the proton transferred from N3 to 3'OH of the peptidyl rRNA as the newly formed peptide is deacylated.

Figures 18 (A) and (B) show space fill representations of the 50S ribosomal subunit with the 3 tRNA molecules, in the same relative orientation as they are in the 70S ribosome structure by Noller and colleagues coupled to the CCA junction at site A and site P. Specifically, Figure 18 (A), shown on the left side, shows the entire subunit in a crown view rotated with the rRNA in yellow, the proteins in pink and the tRNAs in orange . Figure 18 (B), shown on the right side, shows a close-up view showing the proteins numbered in pink and the rRNA in blue. A representation of the skeleton tape of sites A, P and E is shown in yellow, red and white, respectively.

Figures 19 (A) - (F) show the exit tunnel of the polypeptide. Specifically, Figure 19 (A) shows the subunit cut in the middle, approximately bisecting its central protuberance and peptide tunnel along its entire length. The two halves have been opened as pages of a book. All ribosome atoms are shown in a CPK representation, with all RNA atoms that are not in contact with the solvent shown in white and all protein atoms that are not in contact with solvent shown in green. The surface atoms of proteins and RNA are color-coded with carbon in yellow, oxygen in red, and nitrogen in blue. A possible path for a polypeptide that passes through the tunnel is shown as a white ribbon. The peptidyltransferase (PT) site is also shown. Figure 19 (B) shows a detail of the polypeptide exit tunnel with the distribution of polar and non-polar groups, with colored atoms as in Figure 19 (A), the constriction in the tunnel formed by proteins L22 and L4 ( green spots near the PT), and the relatively wide tunnel exit. A modeled polypeptide is shown in white. Figure 19 (C) shows the surface of the tunnel with RNA skeleton atoms color-coded by domain: domain I (white), II (light blue), III (gold), IV (green), V (orange), 5S (pink) and the proteins are in blue. The center of the peptidyltransferase (PTC) is shown. Figure 19 (D) is a representation of space filling of the large subunit surface at the exit of the tunnel showing the arrangement of proteins, some of which may have a role in protein secretion. The RNA is white (bases) and yellow (skeleton) and the numbered proteins are in blue. A modeled polypeptide leaves the tunnel in red. Figure 19 (E) shows a close-up view of the middle of the exit tunnel showing the relationship between the center of the peptidyltransferase (PTC) and the L4 (yellow) and L22 (blue) proteins. The Yarus inhibitor and a modeled peptide are in purple and the 23S rRNA is in red and white. Figure 19 (F) shows a schematic of the secondary structure of the 23S rRNA that identifies the sequences that contact the tunnel in red.

Figure 20 is an illustrative representation showing the spatial relationship between the antibiotic anisomycin bound to a large ribosomal subunit.

Figure 21 is an illustrative representation showing the spatial relationship between the antibiotic blasticidin bound to a large ribosomal subunit.

Figure 22 is an illustrative representation showing the spatial relationship between carbomycin and tylosin antibiotics bound to a large ribosomal subunit.

Figure 23 is an illustrative representation showing the spatial relationship between the sparsomycin antibiotic bound to a large ribosomal subunit.

Figure 24 is an illustrative representation showing the spatial relationship between the antibiotics virginiamycin M (streptogramin A) and carbomycin A bound to a large ribosomal subunit.

Figure 25 is an illustrative representation showing the spatial relationship between the antibiotic spiramycin bound to a large ribosomal subunit.

Figure 26 is an illustrative representation showing the spatial relationship between the antibiotic azithromycin bound to a large ribosomal subunit.

Figure 27 is an illustrative representation showing the spatial relationship between the antibiotic linezolide bound to a large ribosomal subunit.

Figure 28 is an illustrative representation showing the spatial relationship between the erythromycin antibiotic bound to a large ribosomal subunit.

Figure 29 is an illustrative representation showing the spatial relationship of some antibiotics, namely, anisomycin, blasticidine, carbomycin A, and virginiamycin M, attached to a large ribosomal subunit. The locations of bound antibiotics are shown in relation to ribosomal site A, site P and the polypeptide exit tunnel.

Figures 30 (A) - (C) are illustrative representations showing a peptidyltransferase site disposed within a large ribosomal subunit. Figure 30 (A) shows a bound tylosin molecule, and identifies a pocket of union of disacarids and two cavities denoted "cavity 1" and "cavity 2". Figures 30 (B) and (C) are provided on the left side to orient the reader to the locations of the peptidyltransferase (PT) site and the polypeptide exit tunnel in the large ribosomal subunit.

Figure 31 is a schematic representation of a computer system useful for the molecular modeling of a ribosomal subunit and / or for making a rational design of a drug.

Figure 32 is a schematic representation of some potential drug target sites in a large ribosomal subunit.

Figures 33 (A) - (D) are illustrative representations showing residues within the wall of the exit tunnel of polypeptides that are conserved (red) or not conserved (blue) between E. coli and rats. The ribosomal subunit has been cut down from the polypeptide exit tunnel with one half of the polypeptide exit tunnel shown in Figure 33 (A), and the other half of the polypeptide exit tunnel shown in Figure 33 (B) . Figure 33 (C) is provided to guide the reader to show the location of the portion of the ribosomal subunit shown in Figure 33 (A) in relation to the ribosomal subunit as a whole. Figure 33 (D) is provided to guide the reader to show the location of the portion of the ribosomal subunit shown in Figure 33 (B) in relation to the large ribosomal subunit as a whole.

Figure 34 is a schematic representation showing the synthesis of two hybrid antibiotics, sparsocloranfenicol hybrids A (Compound 1) and B (Compound 2), of the individual antibiotics sparsomycin and chloramphenicol.

Figure 35 is a schematic representation showing the synthesis of the sparsoanisomycin hybrid antibiotic (Compound 3) of the individual sparsomycin and anisomysin antibiotics.

Figure 36 is a schematic representation showing the synthesis of a sparsomycin fragment (Compound 9).

Figure 37 is a schematic representation showing the synthesis of a cysteine derivative (Compound 7) and a chloramphenicol fragment (Compound 13).

Figure 38 is a schematic representation showing the synthesis of the sparsocloranfenicol hybrid A (Compound 1).

Figure 39 is a schematic representation showing the synthesis of the sparsocloranfenicol hybrid B (Compound 2).

Figure 40 is a schematic representation showing the synthesis of an anisomycin fragment (Compound 24).

Figure 41 is a schematic representation showing the synthesis of the sparsoanisomycin hybrid (Compound 3).

DETAILED DESCRIPTION OF THE INVENTION

I. Definitions

As used herein, the term "active site" refers to regions in a ribosome or a ribosomal subunit that are directly involved in the synthesis of proteins, for example, the peptidyltransferase site, site A, site P , the exit tunnel of polypeptides, the binding site of the elongation factor, and other similar sites.

As used herein, the terms "agent", "ligand" and "principal candidate" are used as synonyms and refer to any atom, molecule, or chemical group that binds or interacts with a ribosome, a ribosomal subunit or a fragment of ribosome. Thus, ligands include, but are not limited to, a single heavy atom, an antibiotic or an analogue or derivative thereof, an tRNA, a peptidyl tRNA, an aminoacyl tRNA, or a signal recognition particle ("SRP") .

As used herein, "archaebacteria" refers to the kingdom of moneras that includes methane producers, sulfur-dependent species, and many species that tolerate very salty or hot environments.

As used herein, the term "site A" refers to the locus occupied by an aminoacyl-tRNA molecule immediately prior to its participation in the peptide bond forming reaction.

As used herein, the term "asymmetric unit" refers to a minimum set of atomic coordinates that when operated on it with the symmetry operations of a crystal will regenerate the entire crystal.

As used herein, "at least a portion of" or "at least a portion of the three-dimensional structure of" is understood to mean a portion of the three-dimensional structure of a ribosome or a ribosomal subunit, including the distribution of charges and the hydrophilicity / hydrophobicity characteristics, formed by at least three, more preferably at least three to ten, and more preferably even at least ten amino acids and / or nucleotide residues of the ribosome or ribosomal subunit. The residues that form a portion like this can be, for example, residues that form a contiguous portion of the primary sequence of a ribosomal RNA or ribosomal protein, residues that form a contiguous portion of the three-dimensional structure of the ribosome or ribosomal subunit, or a combination of these. As used herein, the residues that form "a portion of the three-dimensional structure of" a ribosome or a ribosomal subunit, form a three-dimensional form in which each atom or functional group that forms the portion of the form is separated from the nearest atom or functional group that forms the portion of the form for no more than 40 A, preferably for no more than 20 A, more preferably for no more than 5-10 A, and more preferably even for no more than 1- 5 A.

As used herein, the term "atomic coordinates" or "structure coordinates" refers to mathematical coordinates (represented as "X," "Y" and "Z" values) that describe the positions of atoms in a crystal. of a ribosome or ribosomal subunit. The diffraction data obtained from the crystals are used to calculate an electronic density map of the crystal repeating unit. Electronic density maps are used to establish the positions of individual atoms in a single ribosomal subunit. Those skilled in the art understand that a set of structure coordinates determined by X-ray crystallography is not exempt from standard error. For the purposes of this invention, the structures of two ribosomes, ribosomal subunits or portions thereof are considered to be the same if they meet one of the following two tests. In a first test, the structures are considered to be the same if a set of structure coordinates for a ribosome or a ribosomal subunit of any source has a deviation from the mean square value of atoms that are not hydrogen of less than about 2.0 A , or more preferably less than about 0.75 A, when they overlap at the positions of non-hydrogen atoms of the atomic coordinates deposited in the Protein Data Bank (PDB) of the Research Collaborative for Structural Bioinformatics (RCSB) (Berman et al. (2000) Nucleic Acids Research 28, 235-242; http://www.rcsb.org/pdb/) with PDB access numbers ID: 1FFK; PDB ID: 1FFZ; PDB ID: 1FG0; PDB ID: 1JJ2; PDB ID: 1K73; PDB ID: 1KCB; PDB ID: 1K8A; PDB ID: IKD1; or PDB ID: 1K9M, or contained in Disk 1 of 1, the disclosure of each of the foregoing is incorporated herein by reference in its entirety. In a second test, the structures are considered to be equal if the deviation of the mean square value between a set of atoms in a test structure and a corresponding set of atoms in a reference structure is less than 2.0 A. For the purpose of this test, the set of atoms in the reference structure comprises at least five of the series of 23S rRNA residues indicated below as 631-633, 835-841, 844-846, 882885, 1836-1839, 2095- 2105, 2474-2478, 2485-2490, 2528-2530, 2532-2543, 2607-2612, 2614-2623, 2642-2648 of the structure deposited in the PDB under the PDB ID access number: 1JJ2 or contained as the name of file 1jj2.rtf on Disk 1 of 1. The residues in the test structure corresponding to those indicated above are identified by sequence alignment using the Lasergene v program. 5.0 (DNA Star, Inc., Madison, WI) with default values. Specifically, the computer program is used to align those residues indicated above in the 23 S RNAR sequence of Haloarcula marismortui with those in the rRNA of the test organism. Once aligned, the corresponding residues in the rRNA of the test organism are identified. The atomic coordinates of the atoms of the skeleton (P, C5 ', 05', C4 ', C3', 03 ') of atoms in the test structure are superimposed on the atoms of the corresponding skeleton (P, C5', 05 ', C4 ', C3', 03 ') of the reference structure using the MIDAS Plus program (Ferrin et al. (1988) J. Mol. Graphics 6: 13-27 and 36-37). The test and reference structures are considered equal if the deviation of the mean square value between the two sets of atoms after the overlap is less than 2.0 A, as determined by MIDAS Plus.

In the list of atomic coordinates deposited in the RCSB Protein Data Bank or included herein as files recorded on compact discs, the term "atomic coordinate" or structure coordinates refers to the measured position of an atom in the structure in the Protein Data Bank (PDB) format, including X, Y, Z and B, for each. The term "type of atom" refers to the element whose coordinates are measured. The first letter in the column defines the element. The term "X", "Y", "Z" refers to the crystallographically defined atomic position of the measured element with respect to the chosen crystallographic origin. The term "B" refers to a thermal factor that measures the average variation of a position of an atom with respect to its average position.

Reference is made to the sets of atomic coordinates and related tables included in this memory and presented on a compact disc (two compact discs in total including an original compact disc, and a duplicate copy of the original compact discs). Disk N.D 1 contains thirty-nine files. Disk No. 1 contains the files identified as DB1FFK.DOC and PDB1FFK.ENT that represent coordinate files that define the large ribosomal subunit; PDB 1 FFZ.DOC and PDB1FFZ.ENT that represent coordinate files that define the large-CCdA-p-pure ribosomal subunit complex; and PDB1FGO.DOC and PDB1FGO.ENT representing coordinate files that define the large ribosomal subunit complex - aa-tRNA analogue; 1 JJ2.RTF and 1JJ2.TXT representing coordinate files that define the large, fully refined ribosomal subunit; anisomycin.pdb, blasticidin.pdb, carbomycin.pdb, sparsomycin.pdb, spiramycin.pdb, tylosin.pdb and virginiamycin.pdb representing coordinate files that define the large ribosomal subunit linked to anisomycin, blasticidine, carbomycin spicomycin, sparsomycin, sparsomycin , tylosin and virginiamycin, respectively; Three folders: FOLDERA contains the file identified as 1JJ2.PDB (which represents a more refined coordinate file that defines the large ribosomal subunit), FOLDERB contains the files identified as ANISOMYC.PDB, BLASTICI.PDB, CARBOMYC.PDB, SPARSOMY.PDB , SPIRAMYC.PDB, TYLOSIN.PDB, and VIRGINIA.PDB (representing files of refined coordinates that define the large ribosomal subunit linked to anisomycin, blasticidine, carbomycin, sparsomycin, spiramycin, tylosin and virginiamycin, respectively), FOLDERC contains the files identified as AZITHROM.PDB, and LINEZOLI.PDB (representing the coordinate files that define the large ribosomal subunit linked to azithromycin and linezolide, respectively); the file identified as erythromycin.pdb (representing a file of the coordinates that define the large ribosomal subunit linked to erythromycin) and azithromycin.pdb and linezolido.pdp (representing files of the refined coordinates that define the large ribosomal subunit attached to azithromycin and linezolido, respectively).

As will be apparent to those skilled in the art, the atomic structures presented herein are independent of their orientation, and that the atomic coordinates identified herein simply represent a possible orientation of a particular large ribosomal subunit. It is evident, therefore, that the atomic coordinates identified herein can be rotated mathematically, translated, augmented, or a combination thereof, without changing the relative positions of the atoms or the characteristics of the respective structure. Such mathematical manipulations are included herein.

As used herein, the terms "atomic coordinates derived from" and "atoms derived from" refer to atomic coordinates or to atoms derived, either directly or indirectly, from an electronic density map. It is understood that the atomic coordinates or atoms derived "directly" from an electronic density map refer to atomic coordinates or atoms that are identified from, and / or adjusted to, an electronic density map, using crystallographic and / or Conventional molecular modeling and therefore can be considered as atomic coordinates or primary atoms. It is understood that the atomic coordinates or atoms derived "indirectly" from an electronic density map refers to atomic coordinates or atoms that are derived from, and therefore are derived or transformations of the atomic coordinates or primary atoms and therefore can be considered as atomic coordinates or secondary atoms. Atomic coordinates or secondary atoms can be generated from atomic coordinates or from primary atoms using conventional molecular modeling techniques. By way of non-limiting example, the atomic coordinates for the large ribosomal subunit of H. marismortui, as described below are considered primary coordinates, while the atomic coordinates of a large ribosomal mammalian subunit that can be derived from the atomic coordinates of H. marismortui by molecular modeling, including, for example, homology modeling and / or molecular replacement, are considered to be secondary coordinates. Both types of atomic coordinates and atoms are considered to be comprised by the invention.

As used herein the terms "join", "join", "joined", "union" or "joined", when used with reference to the association of atoms, molecules or chemical groups, refers to any physical contact or association of two or more atoms, molecules or chemical groups (for example, the binding of a ligand with a ribosomal subunit refers to the physical contact between the ligand and the ribosomal subunit). Such contacts and associations include types of covalent and non-covalent interactions.

As used herein, the terms "complex" or "complexed" refer to the assembly of two or more molecules to give a larger order structure, such as a 50S ribosomal subunit attached to a ligand.

As used herein, the term "computational chemistry" refers to calculations of the physical and chemical properties of the molecules.

As used herein, the term "conjugate system" refers to more than two double bonds that are spatially positioned so that their electrons are completely delocalized with the entire system. The aromatic residues contain conjugated double bond systems.

As used herein, the terms "covalent bond" or "union of Valencia" refers to a chemical bond between two atoms in a molecule created by sharing electrons, usually in pairs, by attached atoms.

As used herein, the term "crystal" refers to any ordered three-dimensional set of molecules that diffracts X-rays.

As used herein, the term "crystallographic origin" refers to a reference point in the cell unit with respect to the operation of crystallographic symmetry.

As used herein, the term "binding domain of the elongation factor" refers to the region of the ribosome that interacts directly with elongation factors, including, for example, elongation factors, EF-Tu and EF-G .

As used herein, the expression "site E" refers to the locus occupied by a deactivated tRNA molecule when it leaves the ribosome after its participation in the formation of the peptide junction.

As used herein, the expression "derivation of the heavy atom" refers to the method of producing a chemically modified form, also known as a "derivative of a heavy atom", of a ribosome crystal and the ribosomal subunit and its complexes. . In practice, a crystal is embedded in a solution containing salts of heavy atoms, or organometallic compounds, for example, mercury chlorides, ethyl mercury phosphate, osmium pentamine or iridium pentamine, which can diffuse through the crystal and join the ribosome or ribosomal subunit. The location or locations of the bonded heavy metal atom (s) can be determined by X-ray diffraction analysis of the embedded crystal. This information, in turn, is used to generate the phase information used to construct the three-dimensional structure of the complex (Blundell et al. (1976) supra).

As used herein, the term "homologous" is understood to mean any one or a combination of

(i) any protein isolated or isolated from a ribosome or a ribosomal subunit (i.e., a ribosomal protein), (ii) any nucleic acid sequence isolated or isolated from a ribosome or a ribosomal subunit (i.e., a ribosomal RNA) , (iii) any protein that has at least 25% sequence identity with respect to an isolated ribosomal protein of E. coli or Rattus norvegicus as determined using the "BLAST" version 2.1.1 computer program implementing all default parameters, or (iv) any nucleic acid that has at least 30% sequence identity with respect to a ribosomal RNA isolated from E. coli or Rattus norvegicus as determined using the "BLAST" version 2.1 computer program. 1 implementing all default parameters. "BLAST" version number 2.1.1 is available and accessible via the world wide web at http: //www/ncbi.nlm.nih.gov/BLAST/ or can be operated locally as a fully executable program on a separate computer.

As used herein, the term "homology modeling" refers to the practice of deriving models for three-dimensional structures of macromolecules from existing three-dimensional structures for their counterparts. Homology models are obtained using computer programs that make it possible to alter the identity of residues in positions, where the sequence of the molecule of interest is not the same as that of the molecule of known structure.

As used herein, the term "hydrogen bonding" refers to two electronegative atoms (or OR

or N), which share a hydrogen that is covalently bound to a single atom, while interacting with the other.

As used herein, the expression "hydrophobic interaction" refers to interactions made by two hydrophobic residues.

As used herein, the terms "IC50" or "inhibitory concentration 50" are understood to mean the concentration of a molecule that inhibits 50% of the activity of a biological process of interest, including, without limitation, cell viability and / or protein translation activity.

As they are referred to herein, ribosomal proteins are referred to as "LX" or "SX", where L means "large subunit"; S means "small subunit"; and X in any case is an integer.

As used herein, the term "MIR" refers to multiple isomorphic replacement, a technique used to derive information from phases of crystals treated with heavy atom compounds.

As used herein, the term "molecular graphics" refers to three-dimensional representations of atoms, preferably on a computer screen.

As used herein, the terms "molecular model" or "molecular structure" refer to the three-dimensional arrangement of atoms in a particular object (for example, the three-dimensional structure of atoms comprising a ribosome or a ribosomal subunit, and atoms that comprise a ligand that interacts with a ribosome or a ribosomal subunit, particularly with a large ribosomal subunit, more particularly with a 50S ribosomal subunit).

As used herein, the term "molecular modeling" refers to a method or procedure that can be performed with or without a computer to make one or more models, and, optionally, to make predictions about structure activity relationships. of ligands. The methods used in molecular modeling range from molecular graphs to computational chemistry.

As used herein, the term "molecular replacement" refers to a method that involves the generation of a model of a ribosome or a ribosomal subunit whose atomic coordinates are unknown, orienting and positioning the atomic coordinates described in the present invention in the cell unit of the unknown ribosome crystals so as to better explain the observed diffraction model of the unknown crystal. The phases can be calculated after this model and combined with the amplitudes observed to give the atomic coordinates of the ribosome or the unknown ribosomal subunit. This type of method is described, for example, in The Molecular Replacement Method, (Rossmann, M.G., ed.), Gordon & Breach, New York, (1972).

As used herein, the term "non-covalent bond" refers to an interaction between atoms and / or molecules that does not involve the formation of a covalent bond between them.

As used herein, the term "peptidyltransferase site" refers to the locus in the large ribosomal subunit where the peptide bonds are synthesized.

As used herein, the term "polypeptide exit tunnel" refers to the channel that passes through the large ribosomal subunit from the peptidyltransferase site outside the ribosome through which the newly synthesized polypeptides pass.

As used herein, the term "protein synthesis inhibitor" refers to any molecule that can reduce, inhibit or otherwise interrupt the synthesis of proteins or polypeptides in a ribosome.

As used herein, the term "P site" refers to the locus occupied by a peptidyl-rRNA at the time it participates in the peptide bond forming reaction.

As used herein, the term "ribofunctional locus" refers to a region of the ribosome or ribosomal subunit that participates, either actively or passively, in the synthesis of proteins or polypeptides in the ribosome or ribosome subunit and / or exports or translocates a protein or polypeptide out of a ribosome. The ribofunctional locus may include, for example, a portion of a peptidyltransferase site, a site A, a site P, a site E, a binding domain of the elongation factor, a polypeptide exit tunnel, and a domain of union of a signal recognition particle (SRP). It is understood that the ribofunctional locus will not only have a certain topology but also have a particular surface chemistry defined by atoms that, for example, participate in the hydrogen bond (for example, proton donors and / or acceptors), have properties specific electrostatics and / or hydrophilic or hydrophobic character.

As used herein, the term "ribosomal subunit" refers to one of the two subunits of the ribosome that can function independently during the initiation phase of protein synthesis, but which together constitute a ribosome. For example, a prokaryotic ribosome comprises a 50S subunit (large subunit) and a 30S subunit (small subunit).

As used herein, the term "ribosome" refers to a complex comprising a large ribosomal subunit and a small ribosome subunit.

As used herein, the expression "binding domain of signal recognition particles" refers to the portion of the ribosome that interacts directly with the signal recognition particle.

As used herein, the term "space group" refers to the arrangement of symmetry elements of a crystal.

As used herein, the expression "symmetry operation" refers to an operation in the space group since it places the atoms in an asymmetric unit in the corresponding atoms in another asymmetric unit.

As used herein, the term "maclado" refers to a single macroscopic crystal containing microscopic domains of the same symmetry that differ significantly in orientation such that the diffraction models of all are superimposed. In a glass, the mosaic blocks, or domains, are oriented in such a way that some point in one direction and others point in a second direction, clearly different, and the directions are such that the diffraction model generated by a group of blocks it falls exactly on the diffraction model of the other group.

As used herein, the term "not maclado" refers to a crystal cell whose domains are aligned. Domains are also known as "mosaic blocks." Most crystals differ as if they were sets of mosaic blocks. You can think of these as small regions, perfectly ordered, within the larger crystal, which in general is not so well ordered. Each block has the same symmetry and packing of the cell unit as all the others.

As used herein, the expression "cell unit" refers to a block of basic parallelepiped shape. The entire volume of the glass can be constructed by regular assembly of such blocks. Each cell unit comprises a complete representation of the model unit, whose repetition forms the crystal.

II.

 Structure and use of the large ribosomal subunit

TO.

 Atomic structure of the large ribosomal subunit at a resolution of 2.4 A, initial refinement

The present invention is based, in part, on the development of a new method for preparing ribosome crystals. The new method provides 50S ribosomal subunit crystals that are much thicker than those available before and that can diffract X-rays at a resolution of approximately 2.4 A. The method eliminates the crushing of crystals that obstruct the progress in the Determination of the crystalline structure of the 50A ribosomal subunit of H. marismortui for many years. The method of preparation of the 50S ribosomal subunit crystals is discussed below.

The present invention is also based, in part, on the atomic structure of the 50S ribosomal subunit crystal of H. marismortui that has been derived from an electron density map of 2.4 A resolution that was experimentally phased using atom derivatives heavy The atomic coordinates that define the large ribosomal unit were deposited on July 10, 2000, in the Protein Data Bank (PDB) at the Research Collaboratory for Structural Bioinformatics (RCSB) (Berman et al. (2000) Nucleic Acid Research 28 , 235242; http: // www.rcsb.org/pdb/) under the PDB access number ID: 1FFK.

In addition, the present invention is based, in part, on the derivation of the atomic coordinates of the following model that is briefly summarized here and discussed in detail in the following sections of the report. This model includes 2,811 of the 2,923 nucleotides of 23S rRNA, the 122 nucleotides of its 5S rRNA, and the structures for the 27 proteins that are well ordered in the subunit.

The secondary structures of both 5S and 23S rRNAs are remarkably close to those deduced for them by phylogenetic comparison. The secondary structure of the 23S rRNA divides it into 6 large domains, each of which has a highly asymmetric tertiary structure. Despite the irregularities of their forms, the domains fit together in an interlaced manner to give a compact mass of RNA that is almost isometric. Proteins are dispersed throughout the structure, concentrated largely on their surface, but they are much less abundant in the subunit regions that are of primary functional significance for protein syntheses - the 30S subunit interface, binding regions for tRNA and the active site of peptidyltransferase. The most striking feature of many of these proteins are the irregular irregular structures of their loops and terms, which penetrate between the RNA helices. The primary role of most proteins in the subunit seems to be the stabilization of the three-dimensional structure of its rRNA.

1. Preparation of the crystal for the 50S ribosomal subunit and structure determination

Several experimental proposals were used to extend the resolution of the electronic density maps of the 50S ribosomal subunit of H. marismortui from 5 A to 2.4 A including improvements in the crystals. A back extraction procedure was developed to reproduce the crystals in a reproducible manner that are much thicker than those available before and can diffract at a resolution of 2.2 A (see, Example 1). Briefly, the crystals were grown at room temperature in hanging droplets by vapor diffusion of seeded solutions retro-extracted from precipitated subunits. The resulting crystals had maximum dimensions of 0.5 x 0.5 x 0.2 mm and were harvested after three weeks. The crystallization of crystals that obstructed progress for many years (Ban et al. (1999) supra) was eliminated by adjusting the crystal stabilization conditions (see, Example 1). The crystals were stabilized by gradual transfer in a solution containing 12% PEG 6000, 22% ethylene glycol, 1.7 M NaCl, 0.5 M NH4Cl, 100 mM potassium acetate, 30 mM MgCl2 and 1 mM of CdCl2, pH 6.2, and were instantly frozen (flash) in liquid propane. Reducing the salt concentration below 1.7 M NaCl (KCl) increases the tendency of the crystals to become mixed. At salt concentrations as low as 1.2 M almost all the crystals were maclados.

All X-rays used for high resolution phaseing were collected in the Brookhaven National Synchrotron Light Source except for two sets of native data used, which were collected in the Advanced Foton Source in Argonne (see Example 2) (Table 1). Derivatives of osmium pentamine (132 sites) and iridium hexamine (84 sites) proved to be the most effective in producing isomorphic replacement and anomalous dispersion phase information at a resolution of 3.2 A (see Example 2 ). Averaging the inter-crystal density, which had contributed significantly to a lower resolution, did not help a resolution beyond approximately 5 A. The electron density maps improved markedly and their resolutions extended, eventually to 2.4 A, using the solvent spinning procedure in CNS (Abrahams et al. (1996) Acta Crystallogr. D 52: 30; Brunger et al. (1998) Acta Crystallogr. D Biol. Crystallogr. 54: 905-921).

Table 1 Statistics for data collection, phase determination and model construction

Data statistics

M1RAS1 LOOK2 Native1 Os (NH3) 52B

3-3B 3B2B

UO2F5 Native2 Ir (NH3) 6 Os (NH3) 6 Ta6Br12 Conc. Heavy atoms -30.0 0.5 -20.0 4.5 3.0 (mM) Embedded time (hs) -1.5 4 -24 hs 24 hours 24 hours Sites no. -132 20 -84 38 9 Resolution (A) (�) 90-2.4 (25-2.4) 40-3.5 (3.6-3.5) 40-3.8 (3.9- 3.8) 30-2.9 (3.0-2.9) 30-3.2 30-3.5 (3.6-3.5) 30-3.8 (3.32-3.22 ) (3.27-3.20) (3.97-3.80)

� (A) 1.00 1.14 1.30 1.00 1,075 1.14 1,255 Observations 6,089,802 1,308,703 596,166 2,832,360 1,823,861 1,646,468 1,288,524

Unique 665.928 429.761 313.863 390.770 541.488 488.275 346.745Redun. (�) 9.1 (6.5) 3.0 (2.5) 1.9 (1.6) 7.2 3.4 4.3 (4.2) 3.7 Complete (�) 95, 6 (71.0) 99.4 (96.8) 92.0 (54.1) 97.1 93.8 98.1 (99.0) 99.5 I / �I (last bin) 25, 5 (1.9) 13.5 (3.3) 8.9 (1.6) 18.0 (6.4) 12.0 (2.6) 10.6 (2.7) 10.8 ( 3.2) Runion (�) 8.6 (69.1) 7.2 (32.0) 9.1 (37.9) 11.2 (36.9) 8.5 (29.5) 12, 1 (46.0) 12.1 (40.5) 02 (anus) (�) - 2.8 (1.0) 1.5 (1.0) - 2.63 (1.48) 1.8 (1.0) 2.42 (1.18) Runion (anus) -6.2 8.0 - 6.7 6.9 Riso - 14.1 (22.7) 26.4 (47.0) - 12.9 (28.1) 19.5 (39.4)

Phasing Statistics

Resolution layers (A): - 73,200 reflections per bin 30.0 5.1 4.0 3.5 3.2 Total MIRAS1 (FOM) 0.52 0.31 0.14 -0.32 Os (NH3) 52B Power Phasing 0.87 0.72 0.66 -0.75 Phasing Power (SAD) 1.40 0.58 0.26 -0.75 Rcullis (Central) 0.62 0.65 0.67 -0, 65 UO2F5 Phasing power 0.47 0.33 0.28 -0.36 Phasing power (SAD) 0.46 0.25 --0.36

(continuation)

Phasing Statistics

Resolution layers (A): -73,200 reflections per bin

Rcullis (centric) 0.72 0.77 0.75 -0.75 MIRAS2 (FOM) 0.48 0.40 0.28 0.12 0.33 Ir (NH3) 63B Phasing power 1.02 0.92 0 , 78 0.66 0.89 Phasing power (SAD) 2.02 1.60 1.22 0.83 1.47 Rcullis (central) 0.58 0.63 0.70 0.74 0.63 Os ( NH3) 63B Phasing power 0.62 0.57 0.58 0.58 0.59 Phasing power (SAD) 0.47 0.39 - 0.42 Rcuillis (central) 0.78 0.78 0.78 0.76 0.78 Ta6Br122B (Used for SAD phase only) Phase power (SAD) 2.77 0.35 0.13 -1.19 FOM 0.76 0.51 0.31 0.14 0.37 ( MIRAS1BMIRAS2BSAD)

Model Statistics

Resolution range (A) 90.0-2.4 rms deviations: Average B factors (A2) Reflections 577.304 Unions (A) 0.0064 All atoms 37.4 Rsrysa (%) 25.2 Angles (0) 1 , 19 23S rRNA 32.3 Rlibre (%) 26.1 Dihedrics (0) 28.8 5S rRNA 43.2 Impropios (0) 1.68 B factors minimum / max. (A2) 70 / 107.9

, wavelength; Redun., Redundancy; (�) last resolution layer. Riso: FPH-FP / FPH, where FPH and FP are the derivative and amplitudes of the native structure factor, respectively. Rsim: i I (h) -I (h) i /: I (h) i, where I (h) is the intermediate intensity after reflections. Phasing power: quadratic mean of the isomorphic difference divided by the residual quadratic mean lack of closure. Rcullis: (FPH-FP FH (calc)) / FPH-FP), where FPH is the structure factor of the derivative and FP is that of the native data. The sum is valid only for central reflection. FOM (demerit figure): mean cosine value of the error in the phase angles. Abbreviations: MIRAS: multiple isomorphic replacement, anomalous dispersion; SAD: anomalous diffraction of simple round length; FOM: figure of merit. Except for the regions obscured by disorder, the experimentally phased, electron density map of 2.4 A resolution was of sufficient quality so that protein sequencing errors and nucleic acid could be identified and corrected. Each nucleotide could be adjusted individually and the difference between A and G was usually clear without reference to the chemical sequence, as was the distinction between purines and pyrimidines (Figure 1).

The subtraction of the atomic model of the experimental electronic density map does not leave a significant density except for water and ions, showing that the model explains the entire macromolecular density. A preliminary refinement of the model was achieved using a mixed objective in the CNS program (Brunger et al. (1998) above). The model was further refined in real space against the 2.4 A electronic density map using the TNT program (Tronrud (1997), Macromolecular Crystallography, Part B, Methods In Enzymology), which gave a model with a free R factor of 0.33. An additional round of mixed objective refinement of the atomic positions and the B factors using CNS led to the structure described below. Its free R factor is 0.27 (Table 1).

2. Sequence adjustment and protein identification.

The 23S rRNA sequence was adjusted on the nucleotide nucleotide electron density map starting from its sarcin / ricin loop sequence (A2691-A2702) (E. coli numbers A2654 to A2665), whose position had been determined at a resolution of 5 A (Ban et al. (1999) supra). Guided by the available information about the secondary structures of 23S RNAs (Gutell, RR (1996), "Comparative Sequence Analysis and the Structure of 16S and 23S rRNA," Ribosomal RNA. Structure, Evolution, Processing, and Function in Protein Biosynthesis, (Dahlberg A. and Zimmerman B., eds.), CRC Press, Boca Raton, FL pp. 111-128), the electron density of remaining RNA clearly accommodated the 5S RNAR sequence. The interpretation of the protein electron density corresponding to the protein was more complicated because each protein region had to be chemically identified before the appropriate sequence could be adjusted in it, but about 4,000 amino acid residues were adjusted in density electronics.

The 50S subunit of H. marismortui appears to contain thirty-one proteins, and there are sequences in the Swiss-Prot database for twenty-eight of the thirty-one proteins, including one, HMS6 or L7ae, which was originally assigned to the small ribosome subunit (Whittmann-Liebold et al. (1990) supra). The three remaining proteins were identified using the sequences of the ribosomal proteins of eukaryotes and other species of archeas as guides. No electron density was found for one of the large ribosomal subunit proteins of H. marismortui in the sequence database, LX. Either the assignment of LX to the large subunit is an error, or LX is associated with a messy region of the subunit, or LX is absent from the subunits examined in general.

The 2.4 A resolution electronic density map does not have a clear electronic density for the L1, L10, L11, and L12 proteins, whose positions are known from previous studies of X-rays and / or low resolution electron microscopy. These proteins are components of the two lateral protrusions of the subunit, which are both poorly arranged in these crystals. L1 is the only protein component of one of them (Oakes, M. et al. (1986)), Structure, Function and Genetics of Ribosomes, (Hardesty, B. and Kramer, G., eds.) Springer-Verlag, New York, NY, 47-67) and is evident in the 9A resolution density maps of the subunit (Ban et al. (1998) supra), but not at higher resolutions. L10, L11 and L12 are components of the other protuberance, which is often referred to as the "stem" of L7 / L12 (Oakes et al. (1986) supra). L11 and the RNA to which it binds were located on the 5 A electron density map of the large subunit resolution of H. marismortui (Ban et al. (1999) supra) using the crystalline structures determined independently of that complex (Conn GL et al. (1999) Science 284: 1171-1174; Wimberly et al. (1999) Cell 97: 491-502). A protein fragment (approximately 100 residues) that is associated with the RNA stem that supports the L11 complex can be seen on the map with a resolution of 2.4 A. Based on the location, it has to be part of L10 . There is no electronic density corresponding to L12 in view of any resolution, but it is known that the L12 tetramer is attached to the ribosome through L10, and it is known that the L10 / L12 assembly is flexible under certain circumstances (Moller et al. (1986) Structure, Function, and Genetics of Ribosomes, supra, pp. 309-325), which may explain its invisibility here.

The structures of the eubacterial homologs of the L2, L4, L6, L14, and L22 proteins have been previously determined in whole or in part (see Table 2). L2, L6 and L14 were initially located on the 5 A resolution map (Ban et al. (1999) supra). L4 and L22 have now been identified and positioned in the same way. The electronic density corresponding to most of the remaining proteins was assigned by comparing the chain lengths and the motifs of sequences deduced from the electronic density map with known sequence lengths, guided by the available information about the relative positions of the proteins (Walleczek et al. (1988) EMBO J. 7: 3571-3576) and protein interactions with 23S rRNA and 5S rRNA (Ostergaard et al. (1998) J. Mol. Biol. 284: 227-240). Each of the electron density regions of the proteins so identified is well represented by its amino acid sequence.

The most important protein identified by sequence similarity was L7ae, which first appeared to be L30e. The identification of the L30e seemed plausible because the yeast structure of the L30 clearly overlaps with the electronic density of the L7ae, and the structure of the RNA to which the L7ae binds closely resembles that of the RNA to which it attaches. unites the yeast of L30 (Mao, H. et al. (1999) Nat. Struct. Biol. 6: 1139-1147). However, the HMS6 sequence, which by similarity of sequences is a member of the L7ae protein family, is better suited to electron density. Four of the other proteins identified by sequence similarity, L24e, L37e, L37ae and L44e, contain zinc finger motifs. The rat homologs of L37e and L37ae were predicted as being zinc finger proteins based on their sequences (Wool et al. (1995) supra), and this prediction helped identify their counterparts in H. marismortui.

Table 2

Large subunit proteins from Haloarcula marismortui

Interactions5

Name1

Hmlg2 Lng3 Conf4 one 2 3 4 5 6 5S Protein

L1�

211 glb B any

L2t

RL8 239 glbBext B B B B (L37ae)

L3

RL3 337 (L37ae) B B B B L14, L24e, (L13)

L4t

RL4 246 glbBext B B TO (L18e), (L24), (L37e)

L5

RL11 176 glb B B L18

L6

RL9 177 glb TO TO B (L13)

L10�

RP0 348 glb? B L12

L11�

RL12 161 glb any

L12�

RL1 / 2 115 glb L10

L13

RL13a 145 glb B TO TO (L3), (L6)

L14

RL23 132 glb B B B L3, L24e

L15

RL27a 164 glbBext B B B (L18e), (L32e)

L18

RL5 186 glbBext TO B B L5, L21e

L19

RL19 148 glbBext B B B TO any

L22

RL17 154 glbBext B TO B B B B any

L23

RL23a 84 glb TO B L29, (L39e)

L24

RL26 119 glbBext B (L4)

L29

RL35 70 glb B L23

L30

RL7 154 glb B B any

L18e

RL18 115 glb B (L4), (L15)

L21e

RL21 95 glb B B TO L18

L24e

RL24 66 glb TO B L3, L14

L31e

RL31 91 glb B B B any

L32e

RL32 240 glb TO B (L15)

L37e

RL37 56 glbBext B B B TO (L4)

L39e

RL39 49 its T B B (L23)

L44e

RL36a 92 glbBext B TO B (L15e)

L7ae

R17a 110 glb TO L15e

L10e

RL10 163 glb B B TO any

L15e

RL15 184 glbBext B TO TO TO B (L44e), L7ae

L37ae

RL37a 72 glbBext B B B L2

The upper protein block includes all those known to have eubacterial homologs of the same name. The second block indicates the proteins found in the large ribosomal subunit of H. marismortui that only have eukaryotic counterparts (Wittmann-Liebold et al. (1990) supra). Their names are followed by the letter "e" to distinguish them from eubacterial proteins that would otherwise have the same name. The third block are large subunit proteins for which there is not yet a sequence of H. marismortui. They are identified by sequence homology using standard L names. 1 The structures of all or part of the homologs of the following proteins were previously determined: LI (Nevskaya et al. (2000) Struct. Fold. Des. 8: 363), L2 (Nakagawa, A. et al. (1999) EMBO J. 18: 1459-1467), L4 (Wahl et al. (2000) EMBOJ. 19: 807-818), L6 (Golden et al. (1993) EMBOJ. 12: 4901-4908), L11 (Conn et al. (1999) supra; Wimberly et al. (1999) supra; Markus et al. (1997) Nature Struct. Biol. 4: 70-77), L12 (Leijonmarck, M. et al. (1980) Nature 286: 824-827), L14 (Davies et al. (1996) Structure 4: 55-66), L22 (Unge et al. (1998) Structure 6: 1577-1586), L30 (Wilson et al. (1986) Proc. Nat. Acad. Sci. USA 83: 7251-7255). All other structures, except 10, have been determined just in this study. 2 Rat homologue. Rat equivalents in relation to H. marismortui proteins are from (Mao et al. (1999) supra). 3 Sequence chain length. 4 Conformation: glb = globular; ext = extension 5 Protein interactions with the 6 domains of 23S rRNA, 5S rRNA and other proteins are specified. (B) implies that the interaction is substantial. (A) implies a weak, tangential interaction. The names of the proteins shown in parentheses imply that the interactions are weak; otherwise, the interaction is substantial. All entries thus designated describe proteins that are not fully represented in the electronic density maps described herein. The summary information is derived from sources in the literature and is included here to complete. t The structure available for this protein in isolation does not include the extension (s) reported here.

3. General appearance of the subunit.

In its crown view (see Figure 2), the large ribosomal subunit, which is approximately 250 A across, presents its surface that interacts with the small subunit to the observer with the three projections that radiate 5 of that pointed surface. Although the protuberance that includes L1 is not visible on the electron density map of 2.4 A resolution, the structure of L1, which has been determined independently (Nikonov et al. (1996) EMBO J. 15: 1350-1359) , it has been positioned approximately on lower resolution maps (Ban et al. (1998) above) and is included here to guide the reader. It is evident that, except for its two lateral protuberances, the large ribosomal subunit is monolithic. There is no indication of a division of its structure in

10 topologically separated domains. In addition, partly because it does not have an obvious domain substructure, but also because it is so large, it is impossible to encompass it by looking at it as a whole. To give the reader a sense of how this is assembled, the subunit has to be divided into its chemical components.

4. Secondary structure of RNA.

All base pairs in the 23S rRNA of H. marismortui stabilized by at least two hydrogen bonds

were identified using a computer program that investigated the structure for the donors and acceptors of hydrogen bonds separated by less than 3.2 A. The bases joined by at least two of such bonds were considered paired if the angle between their perpendicular was less than 450, and the angle between links and perpendicular of the bases was also less than 450. Based on the results of this analysis, a secondary structure diagram was prepared in the standard format for 23S / 28S RNAs (see Figure 3). The secondary structure predicted for this molecule by phylogenetic comparison was remarkably correct, but I do not find all the tertiary matings and failed to identify the interactions involving conserved bases. In addition to the base pairs of almost each type, the RNA contains numerous examples of well-known secondary structure motifs, such as base triplets, tetrabucles, and cross-strand purine stems, but no secondary structure motifs have been identified so far. surprisingly new.

The secondary structure of this 23S rRNA consists of a central loop that is closed by a terminal stem, from which 11 more or less complicated stems / loops radiate. It is customary to describe the molecule as consisting of 6 domains, and number their helical stems starting sequentially from the 5 'end (see, Figure 4) (Leffers et al. (1987) supra). The division of the molecule into domains shown in Figure 4 deviates from standard practice with respect to helix 25, which is usually considered part of domain I. It is placed in domain II because it interacts more strongly with domain II than with the other elements of domain I.

There are five sequences longer than the 10 nucleotides in 23S rRNA whose structures cannot be determined from the 2.4A resolution map due to the disorder. Together they represent 207 of the 232 nucleotides that are missing in the final model. The disordered regions are: (1) all of helix 1, (2) the distal end of helix 38, (3) helix 43/44 to which ribosomal protein L11 binds, (4) the end of stem loop / loop 69, and (5) helix 76/77/78, which is the RNA structure to which L1 binds. To complete, these regions are included in Figure 3 (in gray) with their secondary structures determined phylogenetically.

5. General architecture of ARNr.

The six domains of 23S rRNA and 5S rRNA all have complicated convolute forms, which nonetheless agree with each other to produce a monolithic, compact RNA mass (see Figures 4 (A) and 4 (B)). Thus, despite the organization of its RNAs at the level of the secondary structure, in three dimensions, the large subunit is a giant, individual domain. In this respect, it is quite different from the small subunit, which is a more flattened object, which is not monolithic. Even in low resolution electron micrographs, the small subunit consists of three structural domains, each of which, in turn, contains one of the three domains of secondary structure of its RNA (Noller et al. (1990) The Ribosome: Structure, Function, and Evolution, supra, pp. 73-92). This qualitative difference between the two subunits may reflect a requirement for conformational flexibility that is greater for the small subunit.

Domain I, which resembles a fungus (see, Figure 4 (E)), is located at the back of the particle, behind and below the L1 region. The thin part of the domain begins in the vicinity of domain VI, which is where its first and last residues are located. Helices 1 and 25 encompass the particle at the back and then the domain expands into a more globular structure, larger, below and behind the L1 region.

The largest of the six domains of 23S rRNA, domain II, which represents most of the back of the particle, has three protrusions that extend to the side of the interface of the subunit of the particle (see, Figure 4 (F)). One of them (helix 42-44) is the RNA portion of the L7 / L12 stem, which is known to interact with elongation factors, is not well ordered in these crystals. The protrusion of the second domain II is helix 38, which is the longest, unbranched stem in the particle. It begins at the back of the particle, bends approximately 90 degrees and protrudes into the small subunit between domains V and 5S rRNA. The third region, helix 32-35.1, points directly towards the small subunit and its end, the stem loop / loop 34, interacts directly with the small ribosomal subunit (Culver et al. (1999) Science 285: 2133- 2135). This loop emerges at the interface of the subunit between domains III and IV.

Domain III is a compact globular domain that occupies the lower left region of the subunit in the crown view (see, Figure 4 (G)). It resembles a four-pointed star with the origin of the domain (stem / loop 48) and stem / loops 52, 57, and 58 forming the tips. The most extensive contacts of domain III are with domain II, but also interact with domains I, IV and VI. Unlike the other domains, domain III hardly interacts with domain V; The only contact is a unique van der Waals contact that involves an individual basis for each domain.

Domain IV represents most of the surface of the 50S subunit interface that contacts the 30S subunit (see, Figure 4 (H)). It forms a large diagonal flat surface patch on that side of the subunit, and connects domains III and V on the back of the particle. Helices 67-71 are the most prominent feature of domain IV, and they form the front edge of the cleft of the active site, which is clearly visible at low resolution (see, Figure 2). This is one of the few regions of the 23S rRNA that is not extensively stabilized by ribosomal proteins. The helix 69 in the middle of this crest interacts with the long penultimate stem of the 16S rRNA in the small ribosomal subunit and can be seen as a divisor separating the tRNAs attached to site A of the tRNA attached to site P.

Domain V, which is found between domains IV and II in the middle of the subunit, is known to be intimately involved in the activity of ribosome peptidyltransferase. Structurally the domain can be divided into three regions (see, Figures 4 (I) and 4 (J)). The first begins with helix 75 and finally forms the binding site for the L1 protein. The second, consisting of helices 80-88, forms the bulk of the region of the central protuberance, and is supported on the back by the 5S rRNA and domain II. The third region, which includes helices 89-93, extends into domain VI and helps stabilize the junction region of the ribosome elongation factor.

The smallest domain in the 23S rRNA, domain VI, which forms a large part of the subunit surface immediately below the L7 / L12 stem, resembles a letter X with a horizontal bar at the bottom (see, Figure 4 (K)). An interesting region of this domain is the sarcina-ricin loop (SRL) (stem / loop 95), whose structure has been extensively studied in isolation (Szewczak et al. (1995) J. Mol. Biol. 247: 81-98 ). SRL is essential for factor binding, and ribosomes can be inactivated by cleavage of simple covalent bonds in this loop (Wool et al. (1992) TIBS 17: 266-269). As the nucleotide protection data suggest, the main groove of this loop is exposed to the solvent (Moazed et al. (1988) Nature 334: 362-364), and its conformation is stabilized by proteins and through interaction with the domain V that involves two bases on the minor side of the slot. The nucleotides involved are A 2699 and G 2700 in domain VI, and A 2566 and G 2567 in domain V.

The 5S ribosomal RNA, which is effectively the seventh RNA domain in the subunit, consists of three stems that radiate outward from a common junction called the A loop (see, Figure 4 (D)). In contrast to what is seen in the crystal structure of fragment 1 of E. coli 5S rRNA (Correll et al. (1997) Cell 91: 705-712), arm 2/3 of the molecule's helix is stacked on your 4/5 arm of the propeller, do not propel it 1 (see Figure 4 (L)). This arrangement results from a twisted conformation of loop A residues that involves two triple stacked bases. Indeed, from the point of view of the secondary structure, the arm 2.3 of the helix of loop A or rRNA 5S is quite remarkable. Nowhere else in the subunit is there a greater concentration of unusual mating and secondary structure of rolled RNA.

6. Conservation of sequences and interactions in 23S rRNA.

While 23S / 28S rRNAs contain many conserved sequences, they also vary substantially in chain length. The shorter 23S / 28S RNAs are distinguished from their longer counterparts by truncating,

or even the elimination of whole stems / loops, and by comparison of the sequences, a minimal structure can be identified that is shared by all (Gerbi (1995) Ribosomal RNA: Structure, Evolution, Processing and Function in Protein Biosynthesis, supra, pp. 77-88). The expansion sequences in the 23S rRNA of H. marismortui, that is, the sequences it contains that are larger than the minimum, are shown in Figure 5 in green. They are largely absent from the surface of the interface of the subunit of the particle, but they are abundant on the posterior surface, far from their active sites. This is consistent with the low resolution electron microscopy observations that suggest that the region of the large subunit whose structure is most conserved is the surface that interacts with the small subunit (Dube et al. (1998) Structure 6: 389-399) .

There are two kinds of sequences conserved in rRNA 23S. One contains residues concentrated in the regions of active sites of the large subunit. The second class consists of much shorter sequences dispersed throughout the particle (Figure 5: red sequences). The SRL sequence in domain VI and the cluster of conserved residues belonging to domain V, which are located in the lower part of the peptidyltransferase slit are members of the first class. They are preserved because they are essential for substrate binding, factor binding and catalytic activity. Most of the residues in the second class of conserved residues are involved in the inter-and intra-domain interactions that stabilize the tertiary structure of 23S rRNA. Adenosines are disproportionately represented in this class. The predominance of A among the residues conserved in the ARNrs has been noted in the past (Ware et al. (1983) Nucl. Acids. Res.

22: 7795-7817).

In addition to their dependence on A-dependent motifs, the tertiary structure of the 23S rRNA domains and their relative positions are stabilized by familiar elements of the tertiary structure, such as RNA zippers and tetrabucle / tetrabucle receptor motifs (Moore, PB (1999) Annu. Rev. Biochem. 68: 287300), and in many places, base pairs and triples stabilize the interactions of the sequences that belong to the different components of the secondary structure of the 23S rRNA.

It is interesting that the 5S rRNA and the 23S rRNA do not interact extensively with each other. The few RNA / RNA interactions involved involve the skeletons of the 4/5 arm of the helix or the 5S rRNA and the skeleton of the 38S helix of the 23S rRNA. Most of the free energy and all the specificity or binding of 5S rRNA to the large ribosomal subunit seems to depend on its extensive interactions with proteins that act as a modeling clay by attaching it to the rest of the ribosome.

7. Proteins in the 50S ribosomal subunit.

The structures of twenty-seven proteins found in the large ribosomal subunit of H. marismortui (Table 2) have been determined. Twenty-one of these protein structures have not been previously established by any homologue, and the structures of the six that have homologs of known structure have been reconstructed on the electronic density map using their H. marismortui sequences. In addition, there are structures available for homologs of L1, L11 and L12 of H. marismortu, which cannot be visualized on the electron density map of 2.4 A resolution. Only the structure of L10 is unknown even among the thirty-one proteins known to date.

Not all of these structures are complete. For example, an entire domain of L5 is missing in electronic density, probably due to disorder. L32e is also remarkable in this regard. About twenty residues of its N-end are not seen on the electronic density map, and the electronic density map suggests that its C-terminal residue is covalently linked with most N-ends of its visible residues.

Of the thirty large subunit ribosomal proteins whose structures are known, 17 are globular proteins, similar in nature to the thousands whose structures are found in the Protein Data Bank (Table 2). The thirteen remaining proteins have either globular bodies with extensions protruding from them ("glbBext") or are fully extended ("ext"). Their extensions frequently do not have an obvious tertiary and in many regions they also do not have a secondary structure (see Figure 6). These extensions may explain why many ribosomal proteins have resisted crystallization in isolation. The exceptions that prove the rule are L2 and L4, which are proteins that belong to the "glbBext" class. The L2 protein was crystallized and its structure was resolved only after its extensions were removed (Nakagawa et al. (1999) supra), and one of the two L4 regions that are extended in the ribosome is disordered in the crystalline structure of the L4 intact (Wahl et al. (2000) supra).

Except for the L1, L7, L10 and L11 proteins, which form the tips of the two lateral protrusions, the 50S subunit proteins do not extend significantly beyond the sheath defined by the RNA (see, Figure 7). Its globular domains are found largely outside the particle, often housed in spaces and crevices formed by the folding of RNA. Thus, unlike the proteins in the spherical viruses, the proteins of the large ribosomal subunit do not form a shell around the nucleic acid with which they are associated, and unlike the proteins in the nucleosomes, they are not surrounded by nucleic acid either. Instead, the proteins act like a mortar that fills the spaces and cracks between "RNA bricks."

The distribution of proteins on the surface of the subunit is almost uniform, except for the cleavage of the active site and the flat surface that interacts with the 30S subunit. In the crown view the proteins are found around, on the periphery of the subunit (see, Figure 7 (A)), but when viewed from the opposite side to the junction site of the 30S subunit (the "back side") , seem to form an almost uniform lattice over its entire surface (see, Figure 7 (B)). Similarly, the lower surface of the subunit, which includes the exit of the polypeptide tunnel, is full of proteins (see, Figure 7 (C)). In fact, the 5 proteins that surround the exit of the tunnel can have a role in the secretion of proteins since they are part of the surface oriented towards the membrane and the translocon when the membrane and the secreted proteins are synthesized.

Although Figure 7 shows chains of proteins that disappear in the inner ribosome, the degree to which proteins penetrate the body of the particle can only be fully appreciated when RNA is separated. The interior of the particle is not protein-free, but it is poor in protein compared to the surface of the particle. The extended tentacles of the polypeptide, many of which emanate from globular domains on the surface, penetrate the interior, filling the spaces between the neighboring elements of the secondary structure of the RNA (see, Figure 8 (E)). The extraneous structures of these extensions are explained by their interactions with the RNA.

Although extended non-globular structures are rare in the protein database, they are not unknown. The terms of extended proteins frequently form inter-protein contacts, for example, in viral capsids, presumably adopting fixed structures only after capsid formation. The basic "tails" of histones can behave in the same way when nucleosomes are formed. The Final sequences of capsid proteins are often positively charged, and in the crystalline structures of viruses, the electron density for these sequences frequently disappears within the virus where these sequences presumably interact with asymmetrically ordered nucleic acid. The interactions observed in the ribosome could be useful models for these viral interactions.

The interactions of extended polypeptides and RNA in the large subunit, which stabilizes its massive nucleic acid structure, results in a marafa of RNA and protein in the center of the subunit (see, Figures 8 (A) and 8 (B)) . It is difficult to imagine an object like this forming from its components efficiently in a way that is not a highly orderly way. Chaperones may be required to avoid agglomeration of the extended regions of these proteins, which are likely to be disordered outside the context provided by the rRNA, and to handle folding of the rRNA.

8. Proteins and RNA interactions.

As the protein penetrates the large subunit to a surprising degree, there are only a few segments of the 23S rRNA that do not interact with the protein. Of the 2923 nucleotides in the 23S rRNA, 1157 nucleotides make at least one van der Waals contact with the protein (see, Figure 8 (D)), and there are only ten sequences longer than twenty nucleotides into which no nucleotide enters in contact with the protein. The longest of such sequences contains forty-seven nucleotides, and is the part of domain IV that forms the crest of the cleft of the active site.

The extent of interactions between RNA and protein that occurs when the large subunit is formed can be estimated quantitatively. Using Richards' algorithm (Lee, B. et al. (1971) J. Mol. Biol. 55: 379-400) and a 1.7 A radio probe to compute accessible surface areas, it can be shown that 180,000 A2 of surface is covered when the subunit is formed from its isolated components, but completely structured. This is approximately half of its total surface area. The average is approximately 6,000 A2 per protein. While this is a huge amount compared to the covered surface when the majority of protein oligomers are formed, it should be recognized that the ribosome assembly has to be accompanied by a large loss in conformational entropy that does not occur when most The oligomerized proteins. The ends and loops of the extended protein of ribosomal proteins are almost certainly flexible in isolation, and in the absence of the protein, RNA is probably also quite flexible. Thus, concealment of a large amount of surface area may be required to provide the energy necessary to fix the structures of these molecules.

All proteins in the particle except L12 interact directly with the RNA and all but seven of the thirty remaining proteins interact with two domains of rRNA or more (Table 2). The "champion" in this regard is L22, which is the only protein that interacts with the RNA sequences that belong to the 6 domains of the 23S rRNA (see, Figure 8 (C)). Protein-mediated interactions between the 5S rRNA and the 23S rRNA are particularly extensive. The L18 protein binds helix 1 and helix 2/3 or rRNA 5S to helix 87 of rRNA 23S. The L31e protein mediates an interaction between the same part or the 5S rRNA and domains II and V. The C loop is linked to the V domain by the L5 protein and the D loop is linked to domains II and V by the L10e protein. Whatever other function they play, it is clear that an important function of these proteins is the stabilization of the relative orientations of adjacent RNA domains. Several also help to secure the tertiary structures of the domains with which they interact.

Since most ribosomal proteins interact with many RNA sequences and the number of proteins largely exceeds the number of RNA domains, it is hardly a surprise that any rRNA domain interacts with multiple proteins (Table 2). The V domain, for example, interacts with thirteen proteins, some intimately and a few in passing.

It is clear that the oligonucleotide binding experiments on which information about the RNA binding properties of ribosomal proteins have been underestimated have underestimated their potential to interact with the RNA. The high affinity RNA binding site identified in a protein by such an experiment may indeed be important for ribosome assembly, but its many weaker interactions with other sequences are likely to be lost, and these may also be vital for Ribosome structure Most ribosomal proteins bind cross-linked with RNA and cross-linking is impossible without multiple interactions. Similar considerations may apply to proteins that are components of other ribonucleoproteins such as splenicosome.

Of the seven proteins that interact with only one domain, three (L1, L10, L11) participate directly in the protein synthesis process. Instead of being included in the ribosome to ensure that the RNA adopts the proper conformation, it seems more appropriate to see the RNA as being structured to ensure its correct placement. Three others (L24, L29, L18e) interact with various elements of the secondary structure within the domains to which they join and presumably function to stabilize the tertiary structures of their domains. The last of the simple RNA domain proteins, L7ae, is a puzzle. On the one hand, it cannot function as an RNA stabilizing protein because it interacts only with a short, individual sequence, in domain I, and on the other hand, it is far from important functional sites in the subunit, the peptidyltransferase site and the binding site of the factor. It is quite close to L1, however, which seems to be important for the function of site E (Agrawal et al. (1999) J. Biol. Chem. 274: 8723-8729), and is probably involved in that activity.

While many ribosomal proteins interact primarily with RNA, a few interact significantly with other proteins. The most important structure generated by protein-protein interactions is the protein cluster composed of L3, L6, L 14, L 19 and L24e that is close to the factor binding site. The surface of the protein they provide may be important for factor interactions.

The structure presented above illuminates the strengths and weaknesses of the proposals for complex assemblies that depend on the determination of the structures of the components in isolation. The structures of the globular domains of the protein homologs in the large ribosomal subunit of H. marismortui are largely the same as those of the corresponding domains in the intact subunit, although adjustments in domain positions are sometimes required . Consequently, these structures were very useful to locate the proteins and interpret the electron density maps of lower resolution. However, for obvious reasons, the structures of the extended tails and loops of ribosomal proteins cannot be determined in the absence of the RNAs that give them structure, and the feasibility of strategies that depend on producing low-protein RNA complexes. The molecular weight of all the RNA contacts required to fix the structures of such proteins seems remote. RNA is also a problem. While the sarcin / ricin loop has an almost equal structure in isolation than it does in the ribosome, the structure or 5S rRNA in isolation differs in some aspects from what is observed in the ribosome, and the structure of the P loop isolated (Puglisi et al. (1997) Nat. Struct. Biol. 4: 775-778) does not resemble the structure of the P loop in the ribosome. Clearly a "structural genomics" proposal to the ribosome, which could have involved determining the structures of all its proteins and all possible rRNA fragments, would not have worked. Nor could it be successful for other macromolecular sets.

B. The structural basis of ribosome activity in the synthesis of peptide binding

Analysis of the atomic coordinates discussed in section IIA above along with the additional atomic coordinates of a ribosomal subunit complexed with several analogs, similarly refined, allow an analysis of the ribosomal function. Accordingly, the present invention is also based on complexed crystals of the 50S ribosomal subunit of complex Haloarcula marismortui or with the analogue of the transition state of Yarus, the CCdA-p-pure, or with a mini-helix analog of an aminoacyl -ARNr. The present invention provides the structures of both complexes. The atomic coordinates of the structure of both complexes were deposited on July 26, 2000, in the Protein Data Bank (PDB) of the Research Collaborative for Structural Bioinformatics (RCSB) (Berman et al. (2000) Nucleic Acid Research 28: 235-242; htttp: // www.rcsb.org/pdb/) with access numbers PDB ID: 1FFZ (ribosome complex / CCdA-p-pure 50S) and PDB ID: 1FG0 (ribosome / analog aa-tRNA 50S ).

As discussed below, the complete atomic structures of the large ribosomal subunit and its complexes with two substrate analogs show that the ribosome is a ribozyme. The complete atomic structures also provide information about the catalytic properties of your active site of all RNA. Both analogs of the substrate are contacted exclusively by conserved rRNA residues of domain V of 23S rRNA; there are no protein side chains closer than about 18 A with respect to the peptide bond that is synthesized. The mechanism of the peptide bond synthesis seems to resemble the reverse of the deacination step in serine proteases, where the base of A2486 (A2451) in E. coli has the same general base role as His57 in chymotrypsin. The unusual pKa that A2486 has to possess to perform this function is probably derived from its hydrogen bonding to G2482 (G2447) that interacts with a built-in phosphate that could stabilize unusual two-base tautomers, which is required for catalysis. The polypeptide exit tunnel is largely formed by RNA but has significant contributions from the L22, L39 and L4 proteins and its output is surrounded by the L19, L22, L23a, L24 and L29 proteins.

The CCdA of the Yarus analogue joins the so-called P loop and therefore has to be at the P site. Only the terminal CCA of the aa-tRNA analogue is visible, but as it interacts appropriately with the A loop (Kim et al. (1999) Molec. Cell 4: 859-864), has to be at site A. The puromycin group occupies the same place in both structures, and there are no proteins near that site. Therefore, the catalytic activity of the active site has to completely depend on the RNA. The N3 of A2486 (E. coli A2451) is the titillable group closest to the peptide bond that is being synthesized and probably functions as a general basis for facilitating nucleophilic attack by the amino-group of the site A substrate. this capacity, the pKa of this base has to be around 5 units greater than normal.

1. Structures of complex analogues of the substrate.

To establish how the substrates interact at site A and site P of the large subunit, two analogs of the substrate were used. One of the analogs, which was designed to simulate the acceptor stem of an aa-tRNA and bind to site A, was a twelve-base pair RNA hairpin with an extension of four amino acylated nucleotides at its 3 'end (see, Figure 9). The sequence used was that of the tyr acceptor stem of rRNA and is terminated with puromycin, since it is itself a tyrosyl-A76 analog. The second analog used was the analogue of the transition state of Yarus, CCdA-p-puromycin. As in the case of the analogue of the site A substrate, it is expected that the puromycin of the Yarus inhibitor will bind at site A, while its CCdA part should bind at site P.

The positions of the Yarus inhibitor and the rRNA acceptor stem analog were determined by embedding these molecules in 50S ribosomal subunit crystals of H. marismortui, measuring the diffraction data at a resolution of 3.2 A and calculating the difference in electron density maps (Welch et al. (1997) Biochemistry 36: 6614-6623). The complex maps were also calculated using 2Fo (complexized) -Fo (not complexed) as coefficients, to examine the deviations in the positions of the ribosome residues that occur when these analogs are joined (see, Figure 10 (B) and Table 3).

Table 3

Statistics for data collection and scale

Crystal

Native A Native B CcdAp-Pure Minihelice

Embedded Time (hours)

- - 24 24

Embedding Concentration (! M)

- - 100 100

Wavelength (A)

1.0 1.0 1.0 1.00

Observations

1,571,171 1,344,877 2,590,726 2,712,813

nico

284,533 369,167 367,284 447,121

Redundancy

5.5 3.6 7.0 6.0

Resolution limits (A)

70.0-3.2 70.0-3.0 70.0-3.0 70.0-2.8

(High resolution bin) �

(3.26-3.20) (3.05-3.00) (3.23-3.17) (3.08-3.02)

Full

94.1 (96.0) 98.9 (99.3) 98.6 (99.9) 99.6 (100)

1 / he

14.6 (4.0) 10.8 (3.1) 11.0 (2.8) 10.7 (2.9)

Runiont

10.2 (40) 11.5 (38) 18.8 (84) 14.3 (72)

Native Riso A:

- - 6.8 (20.8) 14.4 (25.2)

Native Riso B:

- - 12.6 (27.4) 17.5 (31.0)

�The statistics between parentheses are calculated for the high resolution bin used in the map calculations, which, as indicated, was sometimes lower resolution than the high resolution bin used in data reduction. t Runion: i I (h) -I (h) i I I (h) i 'where I (h) is the average intensity after reflection. : Riso: FPH -FP FpH where FPH and FP are the amplitudes of the native and embedded crystal structure factor, respectively.

A model for the entire Yarus inhibitor could be adjusted on the difference density map (see Figure 10 (A)), and the complex electron density map shows the N3 of A2486 (2451) within a distance 5 of the hydrogen bond of an oxygen that does not form a phosphoramide bridge (see Figure 10 (B)). The two Cs of the inhibitor, which correspond to C74 and C75 of the peptidyl-tRNA, have the Watson-Crick bases paired with G2285 (2252) and G2284 (2251) in the P loop, respectively (see Figure 11 (A)). The interaction of C74-G2285 (2252) was predicted by the results of Noller et al. (Noller et al. (1992) Science 256: 1416-1419). The dA, which corresponds to A76 of an tRNA at the P site, does not have the paired bases, but rather is stacked in the ribose

10 of A2486 and is joined by a hydrogen bond to the 2'OH of nucleotide A2485 (see Figure 12 (B)).

Only the CC-puromycin part of the acceptor stem analog of the mini-helix showed an ordered electron density on its difference electron density map (see Figure 10 (C)). The C75 of the CCA acceptor stem has the Watson-Crick bases paired with G2588 (2553) of the A loop, while the C74 is more messy and does not have the paired bases, but appears to be stacked on a ribosome base. Dimethyl A inhibitor

The puromycin A site is positioned identically to the dimethyl A of the Yarus inhibitor. In addition, the dimethyl A of puromycin, which is the A76 equivalent of a tRNA of site A, interacts with the A loop in a similar way to that the A76 of the CCA of the P site interacts with the P loop (see Figure 11 (B )).

The most notable of the various conformational changes in the ribosome induced by the union of the transition state analog is the arrangement of the A2637 base (2602), which is disordered in the unbound enzyme

20 (see Figure 11 (B)). It is positioned between the CCA attached at site A and the CCA attached at site P. The base of U2620 (2585) also moves so that it can make a hydrogen bond with the 2 'hydroxyl of the A76 ribose in site A, and U2619 and G2618 deviate to allow the A76 to be positioned. Smaller deviations of the positions in the positions of A2486 are observed, whose N3 is close to the oxygen that does not form a phosphate bridge, and one of the residues G with which it interacts, G2102 (2482).

2. Location and chemical composition of the peptidyltransferase site.

The inhibitors are linked to a site made entirely of 23S rRNA without proteins nearby, proving that the ribosome is a ribozyme. Both the Yarus inhibitor and the aa-tRNA analogue of site A bind to the large subunit at the bottom of a large, deep groove at the entrance to the 100A long polypeptide exit tunnel that runs through the back of the subunit (see Figure 12). This site is surrounded by nucleotides that belong to the central loop of domain V of 23S RNA, the "peptidyltransferase loop." The nucleotides of the single strand portions of this loop make the closest approach to phosphate that simulates the tetrahedric carbon intermediate. In general, helices that extend from the peptidyltransferase loop in 2 secondary structure diagrams of 23S rRNA also extend away from the active site in the tertiary structure (see Figure 13). Although there are 13 proteins that interact with the V domain (see Figure 14 (A)), there are no globular proteins in the vicinity of the inhibitor. The closest polypeptides are the non-globular extensions of several proteins (L2, L3, L4, L10e) that penetrate deeply into the V domain and approach the active site (see Figure 14 (B)). These extensions fill many of the gaps between the RNA helices of the V domain, neutralize phosphate skeleton loading, and presumably stabilize the domain structure and its association with other regions of the RNA. However, none of its side chain atoms are closer than approximately 18 A to the phosphorus of the phosphate group of the inhibitor, marking the site where the peptide bonds are formed. In addition, the substrate analogue is completely enclosed in a rRNA cavity that is so tightly packed that there is no possibility that an unidentified peptide could pass nearby (see Figure 15). Thus, the catalytic entity in the ribosome has to be RNA.

Two of the proteins with long ends or loops that penetrate the scaffold of the RNA domain of domain V are proteins that could not be previously excluded from their participation in the reaction of peptidyltransferase L2 and L3 (Noller (1991) Ann. Rev. Biochem. 60: 191-227). Noller and colleagues (Noller et al. (1992) supra) found that under conditions that prevent denaturation of RNA, the extensive digestion of 50S subunits of Thermus thermophilus with proteases followed by phenol extraction and other agents that disrupt protein interactions -RNA did not remove several peptides from the subunit that were less than 10,000 by molecular weight. The structure becomes evident because these protein fragments were particularly resistant to protease treatments. Although the protease treatment could digest the domains of the globular protein on the surface of the large subunit, it could not remove the long ends or loops that penetrate deep into the 23S rRNA because they are sequestered within the rRNA and thus protected from excision, regardless of globular domains.

3. Active site of peptidyltransferase.

The RNA surrounding the substrate analogs is closely packed, very similar to the region of the active site of a protein enzyme and the nucleotides in contact with the inhibitor are larger than 95% conserved in the three realms of life (see Figure 15 ). So, it is clear that the ribosome is a ribozyme, but what gives RNA its catalytic power?

Without pretending to be limited by theory, the residue most likely to be involved in the catalysis, presumably as a general basis, is A2486, whose N3 is approximately 3 A of the phosphoramide oxygen from the Yarus inhibitor which is the analog of carbonyl oxygen of a nascent peptide bond and about 4 A of the amide corresponding to the nitrogen of the peptide bond amide that is being synthesized. Commonly, the pKa of N1 of adenosine monophosphate is approximately 3.5 and that of its N3 is perhaps 2 pH units lower (Saenger (1984) Principles of Nucleic Acid Structure, (CR Cantor, eds.), Springer Advanced Texts in Chemistry, Springer-Verlag, New York, NY), and for the A2486 to function as a general base, its pKa would have to be raised to 7 or more. The structure of the crystal itself suggests that its pKa is indeed quite unusual. The N3 of A2486 can only be linked by a hydrogen bond to the phosphate oxygen, as noted, if this (or phosphate oxygen) is protonated. The distance between these two atoms is approximately 3 A indicating that a hydrogen bond actually exists between them. Since the crystal is at pH 5.8, this implies that the pKa of N3 is greater than 6. Muth and Strobel have measured the corresponding pKa of A in 23S RNA of E. coli by examining its dimethyl sulfate reactivity as a function of pH and have concluded that it is 7.6, although they cannot be sure from their experiments if the pKa they measured is that of N3 or that of N1 (Muth et al. (2000) Science 289: 947-950). As there is no other functional group of titratable RNA available, closer than approximately 7 A to the nascent peptide bond, there is no other group available that functions as a general basis.

There are several environmental characteristics of A2486 (2451) that could affect its pKa. The pKa of the N3 of A2486 (2451) can be significantly increased in part by a mechanism of charge forwarding, analogous to that occurring at the active site of the serine proteases, with the phosphate of A2485 (2450) incorporated performing a function similar to of the incorporated chymotrypsin Asp102 carboxylate. The experimental 2.4A electronic density map establishes in an unambiguous way the hydrogen bond interactions in this most critical region of the active site (see Figure 16 (A)). The N6 of A2486 interacts with the 06 atoms of G2482 (2447) and G2102 (2061) (see Figure 16 (B)). The N2 of G2482 is also interacting with an oxygen that does not bridge the phosphate group of A2485 (2450) that is among the total of the 3 phosphate groups most inaccessible to the solvent (826, 1497 and 2485) in the large ribosomal subunit for which we don't see any neutralizing counterion on the 2.4 A map of

resolution. The weak density that may correspond to a molecule of water is linked by hydrogen bonding to the other oxygen that does not form a bridge. A neutralizing counterion is not observed in this structure. The incorporated phosphate of A2485 could subtract the proton from the exocyclic N2 of G2482 to neutralize its negatively deposited negative energy charge. This in turn would stabilize the otherwise tautomer imino of that base. The interaction of a G2482 imino with A2486 can also stabilize the A2486 imino tautomer that would result in a negative charge on its N3 where it was deprotonated (see Figure 16 (C)). In this way a part of the negative electrostatic charge that originated in the phosphate deposited from A2485 could be forwarded to N3 of A2486, thus increasing its pKa.

A second characteristic of the environment of the catalytic site that can affect its stability, tautomeric state and distribution of the electrostatic charge is a monovalent cation attached. A potassium ion or a sodium ion interacts with 06 of G2482 and G2102 as well as with three other bases. Its identity as a potassium ion is established by its continued continuation and by an independent experiment that shows that a rubidium ion can bind to this site. The monovalent ion could also stabilize non-standard tautomers, but its expected influence on the pKa of A2486 is less obvious. Previous biochemical experiments have shown the importance of potassium for the activity of peptidyltransferase (Monro (1967) supra; Maden et al. (1968) supra) and this binding site could be responsible for this effect.

It could also be the case that the stabilization of an imino tautomer by an incorporated phosphate explains the expected greater pKa of a catalytic cytosine at the active site of the hepatitis delta virus ribozyme (FerreD'Amare et al. (1998) Nature 395: 567-574; Naharo et al. (2000) Science 287: 1493-1497). In this case, it is observed that a skeleton phosphate, whose accessibility to the solvent is similar to that of the A2485 in the ribosome, is joined by a hydrogen bond to the N4 of C, and the protonated form of the imine tautomer imine of that neutralizing C phosphate, promoting the function of its N3 as a general acid (Naharo et al. (2000) supra).

4. Catalytic mechanism of peptide bond formation.

The proximity of the N3 of A2486 (2451) to the peptide bond that is being synthesized and the nature of the catalyzed reaction suggests a chemical mechanism of the peptide synthesis that is analogous to the reverse of the deactivation step seen in the serine proteases during hydrolysis of peptides (Blow et al. (1969) Nature 221: 337-340; Steitz et al. (1982) Ann. Rev. Biophys. Bioeng. 11: 419-444). In this mechanism, the basic form of His57 subtracts a proton from the amino group of the peptide product since it attacks the acyl-Ser195. The formation of the carbon intermediate of the tetrahedric carbonyl is stabilized by interaction of the oxyanion formed with amides of the skeleton (the "oxyanion hole"); His57 transports the proton acquired from -NH2 to Ser195 when the tetrahedric intermediate decomposes.

The residue A2486 (2451) appears to be analogous to His57 in chymotrypsin and that peptidyl-rRNA is analogous to acyl-Ser195. Thus, the N3 of A2486, with its very high pKa, subtracts a proton from the aminoacyl-tRNA amino group linked to site A that facilitates the nucleophilic attack of that amino group on the carbonyl carbon that acylates the 3'OH of the tRNA at site P (see Figure 17 (A)). In contrast to serine proteases, however, the tetrahedric intermediate oxyanion is close to the protonated N3 of A2486 (A2451) rather than being near a separate oxyanion binding site. Thus, it could be that the protonated N3 of A2486 stabilizes the formation of oxygen by hydrogen bonding with it, as seen in the Yarus inhibitor complex (see Figure 17 (B)). The N3 of A2486 could then subsequently transfer its proton to the 3 'hydroxyl of the tRNA bound to the P site, which is released as the peptide is diverted to the tRNA bound to the A site (see Figure 17 (C)).

An additional question is how is catalyzed hydrolysis of peptidyl tRNA at the P site avoided before the release of the next aa-tRNA appropriate to site A? From this complex it seems that water would not be excluded from access to the peptidyl link to the tRNA of site P, if site A were vacant. An analogous problem was commented by Koshland in the 1960s (Koshland, Jr. (1963) Cold Spring Harbor Symp. Quant. Biol. 28: 473-489), who theorized that hexokinase does not hydrolyse ATP in the absence of glucose, since that water should bind perfectly well to the binding site used by the 6-hydroxyl glucose. The proposed response was induced adjustment, that is, the hexokinase is not catalytically competent until glucose binds and produces a conformational change that orientates the substrates and catalytic groups optimally. This really seems to be the case (Bennett, Jr. et al. (1978) Proc. Natl. Acad. Sci. USA 75: 4848-4852). Similarly, it could be that the catalytic A2486 and / or the peptidyl substrate are not properly oriented or that the binding site for the -NH2 group is blocked by a ribosome base redirected in the absence of aa-tRNA at site A. The base of U2620 appears to be near A2486 in the ligand-free structure, and may serve as the appropriate plug that prevents spontaneous hydrolysis of peptidyl-tRNA.

Thus, it seems that this RNA enzyme uses the same catalytic principles as a protein enzyme. First, a great catalytic increase is achieved by precisely targeting the two reagents, the aminoacyl-tRNA of the A site and the carbonyl carbon of the peptidyl-tRNA of the P site. This can be achieved, in part, by the interactions of the CCA ends of the tRNAs of site A and site P with loop A and loop P, respectively. Secondly, acid-base catalysis and stabilization of the transition state are achieved by an enzymatic functional group (A2486 (2451) in this case) whose chemical properties are appropriately altered by the environment of the active site. Third, similar chemical principles can be used by RNA and protein enzymes to alter the pKa of functional groups. An incorporated Asp 102 carboxylate acting through His57 alters the nucleophilicity of Ser195 in chymotrypsin (Blow et al. (1969) supra). In the ribosome a phosphate inaccessible to the solvent that can act through G2482 (2447) alters the nucleophilicity of the N3 of A2486 (2451). It could be that the RNA molecules "learned" how to use the chemical principles of catalysis significantly before the protein molecules did.

5. Union to tRNA.

While it is not possible to bind the tRNA molecules either to the A site or to the P site in these crystals for steric reasons, it is possible to place the tRNA molecules of the A, P and E sites in the large ribosomal subunit in the same orientation relative observed by Cate et al. in his crystallographic study of the 70S ribosome of Thermus aquaticus. The coordinates of the three tRNA molecules at the relative positions seen on ribosome 70S can be coupled to the large ribosomal subunit of Haloarcula marismortu in a way that avoids steric shocks and places the acceptor stems of tRNAs at sites A and P near the positions of the CCAs we have observed attached to the A loop and the P loop (see Figure 18). Although nucleotides C74 and C75 were modeled in a different conformation on the 7.8 A ribosome map, C74 residues of CCAs at sites A and P may be connected to residue 72 of tRNAs at sites A and P coupled through a modeled residue 73, and it seems that tRNA molecules fit well on the surface of the subunit. Unexpectedly, this modeling places the tRNA molecules bound to the E, P and A sites in close proximity to the three ribosomal proteins. Proteins L5 and L10e are near the tRNAs at site P and site

A. As these two proteins also interact with the 5S rRNA, this observation raises the possibility that the 5S rRNA and some of its associated proteins could help stabilize the positioning of the tRNAs attached to the ribosome and is consistent with the fact that the 5S rRNA increases ribosomal activity, but it is not absolutely essential for this (Moore, Ribosomal RNA & Structure, Evolution, Processing and Function in Protein Biosynthesis (1996), supra, pp. 199-236). The L44e protein appears to interact with the tRNA of site E and may contribute to the activity of site E. According to this coupling experiment, the tRNA of site A interacts with the highly conserved stem loop 2502-2518 (2467-2483) which together with the L10e forms a large concave surface that brings the tRNA into contact on the T-stem, using exactly the same binding site exploited by EF-Tu (Gutell et al. (2000) supra).

The examination of the relationships between the CCAs bound at sites A and P and the rRNAs to which they are connected as well as their interactions with the ribosome also lead to some understanding of the translocation. Immediately after the formation of the new peptide bond and the deactivation of the tRNA of the P site, it is known that the acceptor end of the tRNA of the P site moves to the E side and that that of the rRNA of the A site moves to the P site (Blobel et al. (1970) J. Cell. Biol. 45: 130-145). The approximate modeling of the 3 tRNA molecules in the large subunit suggests some possible contributions to this process. First, there are two base pairs between the P site RNA and the P loop and only one between the A site and the A loop. Moving from site A to site P increases base pairing, although there must be a concomitant attraction of the tRNA from the P site deacrylated to a site E. In addition, the CCAs attached to the A and P loops are related by a rotation of 1800, while the tRNAs to which they are attached are not. Thus, the relationships of the CCAs with the stems of the acceptor cannot be the same at both sites and may not be equally stable. If the conformation of the tRNA from site A is less stable, then moving an tRNA from site A to site P could be energetically favored.

6. Polypeptide exit tunnel.

It seems very likely from the structure that all nascent polypeptides pass through the polypeptide exit tunnel before emerging from the ribosome, since there seems to be no other exit. We can now answer two important questions about the operation of the polypeptide exit tunnel: (1) Why are nascent proteins not sticking to their walls? Does Teflon have the wonderful property of not sticking to the denatured egg proteins, so how has the ribosome achieved a similar surface that does not stick to the denatured proteins that have to pass through the tunnel? (2) Are proteins folded to some degree in the tunnel giving the ribosome a chaperon-like function?

The length of the tunnel on the side of the peptide synthesis at its exit is approximately 100 A, quite consistent with the length of the nascent polypeptide that is protected from proteolytic cleavage by the ribosome (Moazed et al. (1989) Nature 342: 142 ) and the minimum length required for the recognition of antibodies at the exit (Picking et al. (1992) Biochemistry 31: 2368-2375). The tunnel is largely straight, except for a curve 20 to 35 A from the center of the peptidyltransferase (see Figure 19). Its diameter varies from approximately 20 A at its widest point to a narrow point of approximately 10 A at the beginning and at a position 28 A from the tunnel exit with an average diameter of approximately 15 A. As the smallest hole through from which the product of the polypeptide has to pass, only barely houses the diameter of a diameter of a helix �, it seems unlikely that significant folding of the protein beyond the formation of the helix could occur within the ribosome.

Most of the tunnel surface is formed by the I-V domains of the 23S rRNA, but the non-globular regions of the L22, L4 and L39 proteins that not only fill some of the gaps in the RNA scaffolding also make significant contributions. , but also form significant portions of the tunnel wall (see Figure 19). The largest protein contributor to the surface of the tunnel is L22, whose long hairpin loop is between the RNA segments of domains I to IV and is approximately parallel to the axis of the tunnel. Unlike other tunnel proteins, the L39 protein does not have a globular domain on the surface of the particle and is almost completely deposited in domains I and III below the L23 protein. It is interesting that the nucleotides of the 23S rRNA that form the tunnel wall are predominantly of loops in the secondary structure of the 23S rRNA (see Figure 19). As it progresses through the tunnel from the active site, a nascent polypeptide first finds domain V followed 20 A further by domains II and IV and proteins L4 and L22. The last half of the tunnel is formed by domains I and III and the L39e protein.

The narrowest part of the tunnel is formed by the L22 and L4 proteins that approach the tunnel from opposite sides that form what appears to be a gated opening (see Figure 19C). The function of this constriction, if it has one, is not evident. It may be the place where the nature of the nascent chain is felt and the information transmitted to the surface of the particle, perhaps through L22 or L4. The L22 fork at the site of this hole and the 23S rRNA that interacts with it are highly conserved; Its globular portion is located adjacent to the tunnel exit on the surface that the translocon has to face during protein secretion (see Figure 19).

The "non-sticky" character of the tunnel wall must reflect a lack of structural complementarity and polarity with respect to any protein sequence or conformation it encounters. The surface of the tunnel is largely hydrophilic and includes groups of hydrogen bonds of bases, phosphates of the skeleton and side chains of polar proteins (see Figure 19). While there are many hydrophobic groups (sugars, bases, protein side chains) that also face the tunnel, there are no hydrophobic surface patches large enough to form a significant binding site for hydrophobic sequences in the nascent polypeptide. Since the tunnel is approximately 20 A in diameter and filled with water and the newly synthesized polypeptide can presumably move freely, the union of a peptide to the tunnel wall would result in a large loss of entropy that would have to be compensated by a large area of complementary interaction that is larger than 700 A (Chothia et al. (1975) Nature 256: 705-708). Similarly, although the Arg and Lys side chains of the nascent peptide can actually interact with the phosphates exposed in the tunnel, the degree of structural complementarity and the net bond energy obtained after displacing joined counterions must be too small to overcome great unfavorable entropy of immobilization that would result from the union of the peptide. Thus, although the ribosome tunnel is made primarily of RNA, the nature of its surface is reminiscent of the inner surface of the chaperonin, GroEL (Xu et al. (1998) J. Struct. Biol. 124: 129141) in its conformation not binding. Only in the conformation that exposes a large hydrophobic surface does GroEL bind a denatured protein.

There are six proteins (L19, L22, L23, L24, L29 and L31e) located at the exit of the tunnel, which face the translocon on which the ribosome engages during protein secretion. There is evidence that the ribosome binds the translocon even after extensive digestion of its protein by the protease, implying that the interaction between the translocon and the ribosome is mediated by RNA. The proximity of these proteins to the translocon, however, leads us to think what role, if it has, could have in the process of protein secretion. Recent data from the Dobberstein laboratory show that the N-terminal domain of SRP54, the G protein of the signal recognition particle involved in sephalic peptide binding, can cross-link with ribosomal proteins L23 and L29. These two proteins are adjacent to each other and at the tunnel exit (see Figure 19).

7. Evolution.

The in vitro evolution of RNA oligonucleotides has produced small RNA molecules that can bind molecules such as the Yarus inhibitor effectively or catalyze the reaction of peptidyltransferase (Zhaug et al. (1998) Chem. Biol. 5: 539-553; Welch et al. (1997) supra). The sequence and secondary structure of one of these selected RNAs is reminiscent of the peptidyltransferase loop in domain V of 23S rRNA (Zhaug et al. (1998) supra). The most important similarity is a five nucleotide sequence that is identical to a sequence in the V domain that includes the A2486, catalytic G2482 and the deposited phosphate of A2485. Notably, all the groups involved in the proposed load forwarding system to activate the A2486 in the ribosome, are present in the selected ribozyme in vitro. Thus, although the surrounding structural context is probably different, it seems plausible that this artificially developed ribozyme uses the same mechanisms as the ribosome to divert the pKa from an adenine and also uses them as a basis for peptide synthesis. A second RNA (Welch et al. (1997) supra) contains a 12 nucleotide loop that includes a 9 base sequence identical to that found in the same region of the peptidyltransferase loop.

The remarkable similarities between the sequences that contain the key catalytic elements found in the active site of the ribosome peptidyltransferase and the sequences of the selected in vitro RNAs that have related activities make it evident that the appearance of a small RNA domain capable of catalyzing the Peptidyltransferase was a plausible first step in the evolution of protein synthesis in the ribosome. The first peptides synthesized by this enzyme that synthesizes the primordial peptide could have been random polymers or copolymers, and it may have worked with substrates as simple as an amino acylated CCA. The basic peptides of the types observed to form non-globular extensions that co-fold with the 23S rRNA could have been among the first synthesized peptides that were functionally useful. Such peptides may have improved the stability of the protoribosome and other early ribozymes such as the more sophisticated peptides of the current ribosome seem to do.

C. Atomic structure of the large ribosomal subunit at 2.4 A resolution, complete refinement

The three-dimensional structure of the large ribosomal subunit of Haloarcula marismortui has now been completely refined to 2.4 A resolution. The model includes 2876 RNA nucleotides, 3701 amino acids from 28 ribosomal proteins, 117 magnesium ions, 88 monovalent cations, and 7898 water molecules. Many of its proteins consist of a globular domain exposed on the surface and one or more non-globular, basic extensions that are deposited in the rRNA. Half of these include common motifs in non-ribosomal proteins including, for example: RRM domains, SH3 type cylinders and zinc fingers. Proteins that have significant sequence and structural similarity, such as L15 and L18e, perform interactions with essentially identical rRNAs.

More particularly, the 50S subunit of H. marismortui has been completely rebuilt and refined by successive turns of energy gradient minimization and refinement of factor B using CNS (Brunger et al. (1998) supra). Ribosomal proteins and rRNA were completely reconstructed using the "O" software program (Jones, TA et al. (1991) Acta Crystallogr. A46: 110-119) with 2Fo-Fc electronic density maps before solvent modeling and metal ions Modeling errors in proteins were identified using PROCHECK (Laskowski et al. (1993) J. App /. Cryst. 26: 283-291) and by Fo-Fc map inspection. Difference maps also helped in the identification of errors in the rRNA, most often associated with sugar folds. In the process, some adjustments were made in the amino acid conformations, the sequence recording and in the sequences themselves. The sequence changes made were largely limited to L10e, L15e, and L37Ae, the only three proteins of H. marismortui 50S that had not been sequenced directly. In addition, fifty-one amino acids were added to the model described in section IIA with forty-four of these from L10 at the base of the stem of L7 / L12 and L39e that cover a portion of the wall of the polypeptide exit tunnel. Less adjustments were made to the structure of the rRNA. Forty-nine new nucleotides were modeled and refined, mainly in helices 43 and 44 in domain II of rRNA 23S. In addition, the sugar fold or conformation around the glycosidic bond was adjusted for some nucleotides. The refinement process was monitored for the quality of the electronic density maps calculated using phases derived from the model as well as R / Rlibre values. The fully refined model now includes 2876 RNA nucleotides, 3701 amino acids, 210 metal ions and 7898 water molecules. The model is refined to an R / Rlibre of 18.9% / 22.3% and has an excellent geometry (Table 4).

Solvent modeling began with the generation of a list of possible magnesium ions obtained by an automatic peak selection using CNS. Peaks greater than 3.5 were selected on the Fo-Fc maps positioned within the coordination distance of the inner sphere of the magnesium of 1.9 - 2.1 A of the N or O atoms. The resulting list was examined manually. and only the peaks with a clear octahedron coordination geometry such as magnesium were selected.

Monovalent cations were identified based on isomorphic differences between crystals embedded in rubidium and native crystals of the 50S subunit of H. marismartui. As the native crystals were stabilized in the presence of 1.6M NaCl, these sites were initially modeled and refined as NaB1. The refinement of these sites as KB1 almost always resulted in unusually high temperature factors, with two exceptions where we have modeled KB1. Most monovalent sites in the 50S subunit appear to be occupied by NaB1 in our crystals, however, these sites are probably occupied by KB1 in vivo.

Waters were selected as peaks greater than 3.5 A of the O or N atoms. Individual factor B values were used to assess the allocation of water molecules. A number of refined waters to factors B decrease more significantly than the surrounding RNA and protein atoms. In many cases it was found that these peaks were metal ions or chloride ions. A small number of low factor B water molecules was retained in the final model because they could not be assigned unambiguously like other species. As a result of the addition of metal ions and water molecules, the final model now contains 98, 547 non-hydrogen atoms. The refinement and model statistics for the large ribosomal subunit are summarized in Table 4.

Table 4

Refinement statistics and model for the 50S subunit of H. marismortui

Space group

C2221

a = 211.66A, b = 299.67A, c = 573.77A

Total non-hydrogen atoms

98,542

RNA atoms

61,617

Protein atoms

28,800

Water molecules

7,893

Magnesium ions

117

Potassium ions

 2

Sodium ions

      86

Chloride ions

22

Cadmium ions

5

Refinement Statistics:

Resolution Range

15.0 - 2.4 A

Number of reflections used in refinement

623,525

Number of reflections for cross validation

6,187

Work

       18.9%

Free

       22.3%

to a coordinate error (cross validation)

0.35A (0.43A)

Luzzati coordinate error (cross validation)

0.29A (0.35A)

Deviations from the ideal:

r.m.s.d. of union lengths

0.0052A

r.m.s.d. of the angles of union

1,130

r.m.s.d. of dihedrics

15.70

r.m.s.d. improper

2,120

Protein statistics of the Ramachandran chart:

Waste in most favored regions

2704 (86.6%)

Waste in additional allowed regions

379 (12.1%)

Waste in generously permitted regions

27 (0.9%)

Waste in rejected regions

13 (0.4%)

Statistics of the average factor B (Å2):

All atoms (high / low)

44.3 (10.1 / 133.7)

RRNA

41.2 (11.78 / 125.0)

Protein

49.7 (13.9 / 92.5)

Waters

41.89 (9.58 / 115.4)

The refinement has also allowed additional modeling of L10, L39e, and the binding site of L11 in the 23S rRNA. In addition, it has been found that some reasons, for example, RRM topologies, SH3 type cylinders and zinc fingers are common in 50S proteins and each recognizes rRNA in many different ways. Proteins that have a significant three-dimensional homology, however, such as L15 and L18e as well as L18 and S11, perform essentially identical interactions in rRNA. Additional structural homologies are also evident between 50S proteins and non-ribosomal proteins. The solvent-exposed surfaces of these globular protein domains are rich in aspartate and glutamate residues, while irregular extensions of the proteins penetrate the ribosome RNA nucleus. These extensions are often highly conserved, and their abundance of arginine, lysine and glycine residues is important for their function. Collectively, the results show evolutionary connections between many ribosomal proteins and illustrates that the interactions between protein-RNA in the ribosome, although largely idiosyncratic, share some common principles.

D. Antibiotic binding sites

In addition to the preceding structural studies, the structure of the large ribosomal subunit of H. marismortui has been determined complexed with each of nine different antibiotics. More specifically, crystals of the large ribosomal subunit of H. marismortui have been embedded with one of the following antibiotics: anisomycin, blasticidine, carbomycin A, tylosin, sparsomycin, virginiamycin M, spiramycin, azithromycin, linezolide or erythromycin. The structure of the large ribosomal subunit complexed with each antibiotic was then resolved based on the X-ray diffraction data generated for each crystal.

Briefly, a small amount of a concentrated antibiotic solution was added to a large subunit crystal suspended in stabilization solution and incubated for several hours. After freezing and other procedures normally used to prepare such crystals for experimental use, the X-ray diffraction data of the antibiotic containing crystals was collected. Since the crystals were isomorphic with respect to those from which the structure described above is derived, the phases obtained for the native crystal were combined with the diffraction intensities obtained from the crystal embedded in antibiotic to obtain a structure for the latter. The position of the antibiotic in the crystal to which it was attached is most clearly revealed in the electronic density maps of difference, which are electronically density maps computed using the phases just referenced and the amplitudes obtained by subtracting the amplitudes of crystals that do not contain antibiotics of the amplitudes (in adequate scale) of those that contain antibiotics. Using the preceding methods, it was possible to determine the atomic coordinates that show the spatial relationship between particular antibiotics and their binding sites within the large ribosomal subunit. It is considered that similar methods can be used to solve the structure of other antibiotics complexed with the large ribosomal subunit.

The atomic coordinates of the large ribosomal subunit complexed with anisomycin are indicated in a table on a compact disk, Disk ND 1 under the file name anisomycin.pdb, a more refined set of these is represented in a table on a compact disk, Disk ND 1 under the file name ANISOMYC.PDB, and deposited in the Protein Data Bank of the RCSB with the PDB access number ID: 1K73. In addition, Figure 20 shows the spatial relationship between the antibiotic anisomycin and the large ribosomal subunit.

The atomic coordinates of the large ribosomal subunit complexed with blasticidine are indicated in a table on the compact disc Disk ND 1 under the file name blasticidin.pdb, a more refined set of these is represented in a table on a compact disk, Disk ND 1, under the file name BLASTICI.PDB, and deposited in the Protein Data Side of the RCSB with the PDB access number ID: 1KC8. Figure 21 shows the spatial relationship between the antibiotic blasticidine and the large ribosomal subunit. For guidance, Figure 21 also includes a substrate for the site.

The atomic coordinates of the large ribosomal subunit complexed with carbomycin A are indicated in a table on the compact disk, Disk ND 1 under the filename carbomycin.pdb, a more refined set of these is represented in a table on a compact disk, Disk ND 1 under the file name CARBOMYC.PDB, and deposited in the Protein Data Bank of the RCSB with the PDB access number ID: 1K8A. Figure 22 shows the spatial relationship between antibiotic carbomycin A and the large ribosomal subunit. Figure 22 also shows a portion of the polypeptide exit tunnel.

The atomic coordinates of the large ribosomal subunit complexed with tylosin are indicated in a table on a compact disk, Disk ND 1 under the file name tylosin.pdb, a more refined set of these is represented in a table on a compact disk, Disk ND 1, under the file name TYLOSIN.PDB, and deposited in the Protein Data Bank of the RCSB with the PDB access number ID: 1K9M. Figure 22 shows the spatial relationship between the antibiotic tylosin and the large ribosomal subunit. Figure 22 also shows a portion of the polypeptide exit tunnel.

The atomic coordinates of the large ribosomal subunit complexed with sparsomycin are indicated in a table on a compact disk, Disk ND 1, under the file name sparsomycin.pdb, a more refined set of these is represented in a table on a compact disk, Disk ND 1, under the filename SPARSOMY.PDB. Figure 23 shows the spatial relationship between the antibiotic sparsomycin and the large ribosomal subunit. For guidance, Figure 23 also shows a substrate for the P site.

The atomic coordinates of the large ribosomal subunit complexed with virginiamycin M are indicated in a table on a compact disk, Disk ND 1, under the file name virginiamycin.pdb, a more refined set of these is represented in a table on a compact disk , Disk ND 1, under the file name VIRGINIA.PDB. Figure 24 shows the spatial relationship between the antibiotics virginiamycin M as well as carbomycin A, and the large ribosomal subunit.

The atomic coordinates of the large ribosomal subunit complexed with spiramycin are indicated in a table on a compact disk, Disk ND 1, under the file name spiramycin.pdb, a more refined set of these is represented in a table on a compact disk, Disk ND 1, under the file name SPIRAMYC.PDB, and deposited in the Protein Data Bank of the RCSB with the access number PDB ID: 1KD1. Figure 25 shows the spatial relationship between the antibiotic spiramycin and the large ribosomal subunit.

The atomic coordinates of the large ribosomal subunit complexed with azithromycin are indicated in a table on a compact disk, Disk ND 1, under the file name AZITHROM.PDB, a more refined set of these is represented in a table on a compact disk, Disk 1, under the filename azithromycin.pdb. Figure 26 shows the spatial relationship between the antibiotic azithromycin and the large ribosomal subunit.

The atomic coordinates of the large ribosomal subunit complexed with linezolide are indicated in a table on a compact disk, Disk ND 1, under the file name LINEZOLI.PDB, a more refined set of these is represented in a table on a compact disk, Disk ND 1, under the file name linezolid.pdb. Figure 27 shows the spatial relationship between the antibiotic linezolide and the large ribosomal subunit.

The atomic coordinates of the large ribosomal subunit complexed with erythromycin are indicated in a table on a compact disk, Disk N.D 1, under the file name erythromycin.pdb. Figure 28 shows the spatial relationship between the erythromycin antibiotic and the large ribosomal subunit.

Figure 29 shows the spatial orientations of several antibiotics, namely blasticidine, anisomycin, virginiamycin M and carbomycin A, when they bind to their respective antibiotic binding sites within the large ribosomal subunit. For the purpose of guiding the reader, the positions of the P site, the A site and the polypeptide exit tunnel are shown in Figure 29. As is evident, these antibiotics bind to, or contact, specific locations. within the large ribosomal subunit to interrupt protein biosynthesis. For example, it seems that blasticidine binds to the large ribosomal subunit in the vicinity of the P site; anisomycin and virginiamycin bind to the large ribosomal subunit in the vicinity of site A; and carbomycin, spiramycin, tylosin, azithromycin and erythromycin (macrolide 5), all bind to the large ribosomal subunit in the vicinity of the polypeptide exit tunnel adjacent to the peptidyltransferase site.

From Figure 29, it is clear that one skilled in the art can identify some portions of each antibiotic that contact regions in the large ribosomal subunit. Knowing their spatial relationship with respect to each other, the skilled artisan can generate a hybrid antibiotic molecule comprising a portion of a first mold antibiotic and a portion of a second antibiotic of a different mold. The two portions may be joined by a chemical linker so as to maintain the spatial orientation of one portion with respect to the other portion. As a result, the hybrid antibiotic can simultaneously bind each of the regions of the ribosomal subunit typically linked by each mold antibiotic. The design and testing of such molecules is discussed in more detail below.

Figure 30 (A) shows the tylosin molecule bound within the polypeptide exit tunnel. Figure 30 (A) shows an enlarged portion of the large ribosomal subunit with the tylosin antibiotic attached at the top of the polypeptide exit tunnel adjacent to the peptidyltransferase site. Figures 30 (B) and 30 (C) are views showing each half of a large ribosomal subunit cut along the polypeptide exit tunnel and are provided to guide the reader to show the tylosin binding site in relation to the large ribosomal unit as a whole. Figure 30 (A) also shows two cavities defined by the wall of the exit tunnel of the polypeptide and are indicated as "cavity 1" and "cavity 2". In addition, Figure 30 (A) also shows a disacarid binding pocket. The direction in which the chains of the newly synthesized polypeptide exits the ribosome through the polypeptide exit tunnel is indicated by an arrow.

E. Experimental techniques that exploit X-ray diffraction data

Based on the Xn-ray diffraction model obtained from the assembly of the molecules or atoms in a crystalline solid, the electronic density of that solid can be reconstructed using tools well known to those skilled in the art of crystallography and ray diffraction techniques X. Additional phase information extracted either from diffraction data and available in published literature, and / or from supplementary experiments, can then be used to complete the reconstruction.

For basic concepts and procedures for collecting, analyzing and using X-ray diffraction data for the construction of electronic densities see, for example, Campbell et al. (1984) Biological Spectroscopy, The Benjamin / Cummings Publishing Co., Inc., (Menlo Park, CA); Cantor et al. (1980) Biophysical Chemistry, Part II: Techniques for the study of biological structure and function, W.H. Freeman and Co., San Francisco, CA; A.T. Brunger (1993) X-PLOR Version 3.1: A system for X-ray crystallography and NMR, Yale Univ. Pr., (New Haven, CT); M.M. Woolfson (1997) An Introduction to X-ray Crystallography, Cambridge Univ. Pr., (Cambridge, UK); J. Drenth (1999) Principles of Protein X-ray Crystallography (Springer Advanced Texts in Chemistry), Springer Verlag; Berlin; Tsirelson et al. (1996) Electronic Density and Bonding in Crystals: Principles, Theory and X-ray Difraction Experiments in Solid State Physics and Chemistry, Inst. Of Physics Pub .; U.S. Patent No. 5,942,428; U.S. Patent No. 6,037,117; U.S. Patent No. 5,200,910 and U.S. Patent No. 5,365,456 ("Method for Modeling the Electronic Density of a Crystal").

A molecular model can then be progressively constructed using experimental electronic density information and further refined against X-ray diffraction data resulting in a correct solid molecular structure.

F. Structural determination of other large ribosomal subunits

It is understood that the person skilled in the art, when provided with the atomic coordinates of a first macromolecule, can use this information to quickly and easily determine the three-dimensional structure of a different but structurally related macromolecule. For example, the atomic coordinates that define the large ribosomal subunit of H. marismortui can be used to determine the structure of the large ribosomal subunit of other species either as an isolated subunit, complexed with the small subunit, or complexed with functionally important ligands, for example: aminoacyl tRNA; several protein synthesis factors, such as elongation factor G, elongation factor Tu, termination factor or recycling factor, in their conformational states GTP and GDP; and protein synthesis inhibitors, for example, antibiotics. In addition, the coordinates of the H. marismortui subunit can also be used to solve the structures of the ribosomal complexes with components of the protein secretion machinery, for example, the signal recognition particle, and the translocon.

If the crystal being examined contains a macromolecule of unknown structure and no additional information is available, additional experiments may sometimes be required to determine the relevant phases of the macromolecule. These studies may frequently take a long time and have an uncertain success (Blundell et al. (1976) supra). However, when additional information, for example, structural and / or crystallographic information, is available for molecules related in some way to the macromolecule of interest, then the process of solving the structure of the molecule of interest is a not so difficult task. And it doesn't consume so much time.

Therefore, the person skilled in the art can use information collected from the structure previously resolved to develop a three-dimensional model of a new molecule of interest. In addition, the art expert can use a variety of proposals to explain the three-dimensional structure of the new molecule. The proposals may depend on the crystals of the molecule of interest being available and / or that the molecule of interest has a counterpart whose structure has already been determined.

In a proposal, if the molecule of interest forms crystals that are isomorphic, that is, they have the same dimensions of cell unit and spatial group as the related molecule whose structure has been determined, then the phases and / or coordinates for the Related molecule can be directly combined with newly observed amplitudes to obtain electron density maps and, consequently, atomic coordinates of the molecule of interest. The resulting atomic maps and / or coordinates can then be refined using standard refinement techniques known in the art. In another proposal, if the molecule of interest is related to another molecule of known three-dimensional structure, but crystallizes in a different cell unit with different symmetry, the person skilled in the art can use a technique known as molecular replacement to obtain useful phases of the coordinates of the molecule whose structure is known (Blundell et al. (1976) supra). This proposal was reportedly used in the determination of the 50S subunit structure of Deinococcus radiodurans (Harms J. et al., Cell 107 (5): 679-88; Schlunzen F. et al., (2001) Nature 413 (6858): 814-21). These phases can then be used to generate an electronic density map and / or atomic coordinates for the molecule of interest. In another proposal, if there are no crystals available for the molecule of interest but it is homologous to another molecule whose three-dimensional structure is known, the person skilled in the art can use a process known as homology modeling to produce a three-dimensional model of the molecule of interest. . It is considered that other proposals may be useful to derive a three-dimensional model of a molecule of interest. Accordingly, the information concerning the crystals and / or atomic coordinates of a molecule can greatly facilitate the determination of the structures of the related molecules.

The molecular replacement method, initially developed by Rossmann and Blow in the 1960s, is now routinely used to establish the crystalline structures of macromolecules of unknown structures using the structure of a homologous molecule, or one in a different ligation state (MG Rossmann, ed. "The Molecular Replacement Methods", Int. Sci. Rev. J. No. 13, Gordon & Breach, New York, NY (1972); Eaton Lattman, "Use of Rotation and Translation Functions," HW Wyckdef, CHW Hist. (SN Timasheff, ed.) Methods in Enzymology, 115: 55-77 (1985)). For an example of the application of molecular replacement, see, for example, Rice,

P.A. & Steitz, T.A. (1994) EMBO J. 13: 1514-24.

In molecular replacement, the three-dimensional structure of the known molecule is positioned within the cell unit of the new crystal, finding the orientation and position that provides the best agreement between the observed diffraction amplitudes and those calculated from the coordinates of the positioned subunit. . From this modeling, approximate phases can be derived for the unknown crystal. To position a known structure in the unit cells of an unknown but related structure, three rotation angles and three translations related to the origin of the cell unit have to be determined. The investigation of the rotation is carried out looking for a concordance between Patterson's function of the search and the objective structures based on their relative orientation (the function of rotation). X-PLOR (Brunger et al. (1987) Science 235: 458-460; CNS (Crystallography & NMR System, Brunger et al., (1998) Acta Cryst. Sect. D 54: 905921), and AMORE: an Automatic Package for Molecular Replacement (Navaza, J. (1994) Acta Cryst. Sect. A, 50: 157163) are computer programs that can execute the searches of the rotation and translation function. Once the orientation of a molecule of test, the position of the molecule must be found using a translational search Once the known structure has been positioned in the cell unit of the unknown molecules, the phases for the diffraction data observed from the coordinates can be calculated Atoms of structurally related atoms of known molecules Using the calculated phases and X-ray diffraction data for the unknown molecule, the person skilled in the art can generate a map of electronic density and / or atomic coordinates of the molecule of interest .

As an example, it is considered that a three-dimensional model of a ribosomal subunit different from that derived from H. marismortui can be generated through molecular replacement. In this method, the H. marismortui subunit structures are positioned within the cell unit of the new crystal, finding the orientation and position that provides the best concordance between the observed diffraction amplitudes and those calculated from the coordinates of the positioned subunit. A starting electron density map calculated using 2Fhkl (observed) - Fhkl (calculated), where F (observed) are the amplitudes of diffraction that have been measured by crystals of the unknown structure, and F (calculated) are the amplitudes of Calculated diffraction of the subunit structure of H. marismortui positioned. The refinement of the initial model can be performed since it is standard in the field of macromolecular crystallography.

The 50S structure of H. marismortui can also be used to establish the structure of a 70S ribosome or a 50S ribosome for which an electron density map has been calculated, at a resolution that would otherwise be too low to be interpreted, while that a resolution map 5 could not be interpreted in de novo atomic terms, a plausible atomic model can be constructed by readjusting the 50S structure of H. marismortui to a lower resolution map (for example, 4.5 h at 8 h). This readjustment can be combined with homology modeling to obtain a three-dimensional model of a ribosome or a ribosomal subunit of a different species. It is considered that similar procedures can be used to determine the structure of the eukaryotic 60S subunit and / or a eukaryotic ribosome.

In general, the success of a molecular replacement to solve structures depends on the fraction of the structures that are related and their degree of identity. For example, if approximately 50% or more of the structure shows a r.m. difference. Among the corresponding atoms in the range of approximately 2 hours or less, the known structure can be used successfully to solve the unknown structure.

Homology modeling, also known as comparative modeling or knowledge-based modeling, can be used to generate a three-dimensional model for a molecule based on the known structure of homologs. In general, the method may comprise one or more of the following steps: alignment of the sequence of an amino acid or nucleic acid of an unknown molecule against the sequence of an amino acid or nucleic acid of a molecule whose structure has been previously determined; identify structurally conserved and structurally variable regions; generate the atomic coordinates for the nucleus residues (structurally conserved) of the unknown structure of those of the unknown structures; generate conformations for the other residues (structurally variable) in the unknown structure; build conformations of side chains; and refine and / or evaluate the unknown structure.

As an example, since the nucleotide sequences of all known 50S subunit RNAs can be aligned with respect to each other and to the 23S and 5S rRNAs of H. marismortui, it is possible to construct models of the structures of other 50S ribosomal rRNAs. , particularly in the tunnel regions and active sites, using the structure of H. marismortui. Similarly, homologous proteins can also be modeled using similar methodologies. Useful methods for the analysis of comparative RNA sequences are known in the art and include visual methods and number model methods, as well as methods that employ chi-square statistics, phylogenetic algorithms or empirical algorithms. Descriptions of some of the methods mentioned above are available, for example, at http://www.ARN.icmb.utexas.edu/; Gutell (1996), "Comparative Sequence Analysis and the Structure of 16S and 23S rRNA", Ribosomal RNA. Structure, Evolution, Processing, and Function in Protein Biosynthesis, (Dahlberg A. and Zimmerman B., eds.) CRC Press. Boca Raton, pp. 111-128; Guttell et al. (1993) Nucl. Acid Res. 21: 3055-3074; Schnare et al. (1996) J. Mol. Biol. 256: 701719. Particularly useful visual inspection methods include comparing a particular position in a secondary structure diagram of H. marismortui with the residues located in the analogous position in a secondary structure diagram of E. coli. A software program that is particularly useful in homology modeling includes XALIGN (Wishart, D. et al., (1994) Cabios 10: 687-88). See also, U.S. Patent Do not.

5,884,230.

To model the rRNA of a new species, bases of the hRNA of H. marismortui can be replaced, using a computer graphics program such as "O" (Jones et al., (1991) Acta Cryst. Sect. A, 47: 110-119), for those of the homologous rRNA, where they differ. In many, if not most, the same orientation of the base will be appropriate. The insertions and deletions may be more difficult and speculative, but the rRNA that forms the peptidyltransferase site and the portion of the tunnel closest to it is highly conserved essentially without insertions and deletions. Automated web-based homology modeling can be performed using, for example, the SWISS-MODEL computer programs available through Glaxo Welcome Experimental Research in Geneva, Switzerland, and WHATIF available on EMBL servers.

For other descriptions of homology modeling, see, for example, Gutell R.R. (1996), supra; Gutell R.R., et al. (1993) Nucleic Acids Res. 21: 3055-3074; Schnare et al. (1996) J. Mol. Biol., 256: 701-719; Blundell et al. (1987) Nature 326: 347-352; Fetrow and Bryant (1993) Bio / Technology 11: 479-484; Greer (1991) Methods in Enzymology 202: 239-252; and Johnson et al. (1994) Crit. Rev. Biochem. Mol. Biol. 29: 1-68. An example of homology modeling can be found, for example, in Szklarz G.D. (1997) Life Sci. 61: 2507-2520.

As mentioned above, the large ribosomal subunit of prokaryotes and eukaryotes, and eukaryotic mitrochondria are structurally conserved. The amino acid sequences of the large ribosomal subunit of prokaryotes and eukaryotes may be aligned due to the evolutionary conservation of amino acid residues that are important for the three-dimensional structure, nature and shape of the binding sites for the substrates and the site catalytic. This similarity in the amino acid sequence of the large homologous ribosomal subunit allows the construction of models, through homology modeling, for molecules whose crystalline structures have not been resolved.

The new ribosome or large ribosomal subunit structures determined using the crystals and / or atomic coordinates of H. marismortui can then be used for the design of structure-based drugs using one or more of the proposals described below. This information can then be used to design molecules that selectively bind and interrupt the synthesis of proteins in the pathogens of the pathogens while leaving the ribosomes of a relatively unaffected host.

G. Rational drug design

1. Introduction

It is considered that the atomic coordinates that define a large ribosomal subunit of interest, whether derived from one or more X-ray crystallography, molecular modeling, homology modeling or molecular replacement, can be used in the rational drug design (RDD) to dissect a new molecule of interest, for example, new modulators (for example, inducers, mimetics or inhibitors) of ribosomal function. Furthermore, it is considered that using the principles disclosed herein, the person skilled in the art can design, perform, test, refine and use new protein synthesis inhibitors specifically created by genetic engineering to reduce, interrupt, or otherwise, or inhibit ribosomal function in an organism or species of interest. For example, using the principles discussed herein, the person skilled in the art can create, by genetic engineering, new molecules that specifically target and inhibit ribosomal function in a pathogen, for example, a particular prokaryotic organism, while preserving ribosomal function. in a host, for example, a eukaryotic organism, specifically a mammal, and more specifically, a human being. As a result, the atomic coordinates provided and discussed herein allow the person skilled in the art to design new antibiotics that can kill some pathogenic organisms, while having little or no toxicity in the intended receptor, for example, a human being.

It is considered that the RDD using atomic coordinates of the large ribosomal subunit can be facilitated more quickly through a computer-assisted drug design (CADD) using conventional computer hardware and software known and used in the art. The candidate molecules can be de novo designed or can be designed as a modified version of an existing molecule, for example, a pre-existing antibiotic, using conventional methodologies. Once designed, the candidate molecules can be synthesized using standard methodologies known and used in the art. After the synthesis, the candidate molecules can be traced with respect to bioactivity, for example, for their ability to reduce or inhibit ribosomal function, their ability to interact with, or bind to a ribosome, or a ribosomal subunit. Based in part on these results, the candidate molecules can be iteratively refined using one or more of the preceding steps to produce a more convenient molecule with a desired biological activity. The resulting molecules can be useful for the treatment, inhibition or prevention of the biological activities of the target organisms, thus killing the organism or preventing its growth. Alternatively, the resulting molecules can be useful for the treatment, inhibition or prevention of microbial infections in any organism, particularly in animals, more particularly in humans.

In summary, the tools and methodologies provided in the present invention can be used to identify and / or dissect molecules that bind and / or interact conveniently with ribosomes and ribosomal subunits. Basically, the procedures use an iterative process by which the molecules are synthesized, tested and characterized. New molecules can be disseminated based on the information gained in the testing and the characterization of the initial molecules and then such newly identified molecules can be tested and characterized. This series of processes can be repeated as many times as necessary to obtain molecules with binding properties and / or desirable biological activities. The methods for identifying candidate molecules are discussed in more detail below.

2. Identification of candidate molecules

It is considered that the design of the candidate molecules of interest can be facilitated by conventional ball and bar modeling procedures. However, in view of the size and complexity of the large ribosomal subunit, it is considered that the ability to dissect candidate molecules can be significantly improved using computational design and modeling protocols.

to. Molecular Modeling

It is considered that the design of the candidate molecules, as discussed in detail below, can be facilitated using conventional computers or work terminals, commercially available, for example, from Silicon Graphics Inc. and Sun Microsystems, which operate, for example, operating systems based on UNIX, Windows NT or IBM OS / 2, and capable of operating conventional computer programs for molecular modeling and rational drug design.

It is understood that any computer system that has the general characteristics presented in Figure 31 may be useful in the practice of the invention. More specifically, Figure 31 is a schematic representation of a typical computer work terminal that has in electrical communication (100) with another route, for example, an internal bus or an external network, a central processing unit (101), a random access memory (RAM) (102), a read-only memory (ROM) (103), a monitor or terminal (104), and optimally an external storage device, for example, a floppy disk, a CD ROM, or a magnetic tape (105).

The computer-based systems of the invention preferably comprise a data storage medium having a sequence of a ribosome or a ribosomal subunit or a fragment stored therein, and / or atomic coordinate data / ray diffraction data X of the present invention and the hardware and software means necessary to support and implement a means of analysis. As used herein, "a computer system" or "a computer-based system" refers to the hardware media, the software media, and the data storage media used to analyze the sequence, data from X-ray diffraction, and / or the atomic coordinates of the invention. As used herein, the term "data storage media" is understood to refer to any memory that can store sequence data, atomic coordinates, and / or X-ray diffraction data, or an access means of memory that can access elements that have the atomic coordinates of the present invention engraved therein.

In one embodiment, a ribosome or a ribosomal subunit, or at least a subdomain thereof, a sequence of amino acids and nucleic acids, X-ray diffraction data and / or atomic coordinates of the present invention are recorded on a computer-readable medium. . As used herein, it is understood that the expression "computer readable medium" means any media that can be read and that can be accessed directly by a computer. Such media include, but are not limited to: magnetic storage media, such as floppy discs, hard disk storage media and magnetic tape; optical storage media such as optical discs or CD-ROM; electric storage media such as RAM and ROM; and hybrids of these categories such as magnetic / optical storage media. One skilled in the art can easily appreciate how any computer-readable medium currently known can be used to create an element comprising a computer-readable medium that has a sequence of amino acids and / or nucleotides, ray diffraction data recorded therein. X, and / or atomic coordinates of the present invention.

As used herein, it is understood that the term "recorded" means any process for storing information in a computer-readable medium. One skilled in the art can easily adopt any of the methods currently known for recording information in a computer-readable medium to generate elements comprising a sequence of amino acids or nucleotides, atomic coordinates and / or X-ray diffraction data of the present invention. .

A variety of data storage structures are available to a person skilled in the art to create a computer-readable medium that has amino acid and / or nucleotide sequences, atomic coordinates and / or X-ray diffraction data recorded on it. The present invention. The choice of the data storage structure will generally be based on the means chosen to access the stored information. In addition, a variety of programs and data processing formats can be used to store sequence information, X-ray data and / or the atomic coordinates of the present invention in the computer-readable medium. The preceding information, data and coordinates may be represented in a word processing text file, formatted in commercially available software, such as WordPerfect and MICROSOFT Word, or represented in the form of an ASCII file, stored in an application. of databases, such as DB2, Sybase, Oracle, or similar. An expert in the art can easily adapt any number of structuring processor formats (for example, text files or databases) to obtain a computer-readable medium that has the information of the present invention engraved therein.

By providing a computer-readable medium that has a sequence of a ribosome or ribosomal subunit, and / or atomic coordinates stored in it, a person skilled in the art can normally access the sequence, and / or atomic coordinates to model a ribosome or a ribosomal subunit, a subdomain of these, a mimetic, or a ligand of these. Computing algorithms are publicly and commercially available, allowing the skilled person to access these data provided in a computer-readable medium and analyze them for molecular modeling and / or RDD. See, for example, Biotechnology Software Directory, MaryAnn Liebert Publ., New York, NY (1995).

Although computers are not required, molecular modeling can be facilitated primarily by using computers to construct realistic models of a ribosome, a ribosomal subunit, or a portion thereof. Molecular modeling also allows the modeling of new smaller molecules, for example ligands, agents and other molecules, which bind to a ribosome, a ribosomal subunit, or a portion thereof. The methods used in molecular modeling range from molecular graphs (i.e., three-dimensional representations) to computational chemistry (i.e. calculations of physical and chemical properties) to make predictions about the union of the smallest molecules or their activities; to dissect new molecules; and to predict new molecules, including ligands such as drugs, for chemical synthesis.

For basic information on molecular modeling, see, for example, M. Schlecht, Molecular Modeling on the PC (1998) John Wiley &Sons; Gans et al., Fundamental Principals of Molecular Modeling (1996) Plenum Pub. Corp .;

N.C. Cohen, ed., Guidebook on Molecular Modeling in Drug Design (1996) Academic Press; and W.B. Smith, Introduction to Theoretical organic Chemistry and Molecular Modeling (1996). US patents that provide detailed information on molecular modeling include, for example: U.S. patents. Nos. 6,093,573; 6,080,576; 6,075,014; 6,075,123; 6,071,700; 5,994,503; 5,884,230; 5,612,894; 5,583,973; 5,030,103; 4,906,122; Y

4,812.12.

Three-dimensional modeling may include, but is not limited to, making three-dimensional representations of structures, drawing illustrations of structures, constructing physical models of structures, and determining related ribosome structures, ribosome subunits and ribosome / ligand complexes and ribosomal subunit / ligand, using known coordinates. Appropriate coordinates are entered into one or more computer programs for molecular modeling, as is known in the art. By way of illustration, a list of useful computer programs to view or manipulate three-dimensional structures include: Midas (University of California, San Francisco); MidasPlus (University of California, San Francisco); MOIL (University of Illinois); Yummie (Yale University); Sybil (Tripos, Inc.); Insight / Discover (Biosym Technologies); MacroModel (Columbia University); Quanta (Molecular Simulations, Inc.); Cerius (Molecular Simulations, Inc.); Alchemi (Tripos, Inc.); LabVision (Tripos, Inc.); Rasmol (Glaxo Research and Development); Ribbon (University of Alabama); NAOMI (Oxford University); Explorer Eyechem (Silicon Graphics, Inc.); Univision (Cray Research); Molscript (Uppsala University); Chem-3D (Cambridge Scientific); Chain (Bailor College of Medicine); O (Uppsala University); GRASP (Columbia University); X-Plor (Molecular Simulations, Inc .; Yale University); Spartan (Wavefunction, Inc.); Catalyst (Molecular Simulations, Inc.); Molcadd (Tripos, Inc.); VMD (University of Illinois / Beckman Institute); Sculpt (Interactive Simulations, Inc.); Procheck (Brookhaven National Library); DGEOM (QCPE); RE VIEW (Brunell University); Modeller (Birbeck College, University of London); Xmol (Minnesota Supercomputing Center); Expert Protein (Cambridge Scientific); HyperChem (Hypercube); MD Display (University of Washington); PKB (National Center for Biotechnology Information, NIH); ChemX (Chemical Design, Ltd.); Cameleon (Oxford Molecular, Inc.); and Iditis (Oxford Molecular, Inc.).

A proposal for RDD is to look for known molecular structures that can join a site of interest. Using molecular modeling, RDD programs can see a range of different molecular molecular structures that can be adjusted to a site of interest, and by moving them on the computer screen or through computing, you can decide which structures fit really well with the site (William Bains (1998) Biotechnology from A to Z, second edition, Oxford University Press, p. 259).

An alternative but related proposal begins with the known structure of a complex with a small molecule ligand and models modifications of that small molecule in an effort to make additional favorable interactions with a ribosome or a ribosomal subunit.

The present invention allows the use of molecular and computer modeling techniques to design and select new molecules, such as antibiotics or other therapeutic agents, that interact with ribosomes and ribosomal subunits. Such antibiotics and other types of therapeutic agents include, but are not limited to, antifungals, antivirals, antibacterials, insecticides, herbicides, miticides, rodenticides, etc.

To facilitate molecular modeling and / or RDD, the person skilled in the art can use some or all of the atomic coordinates deposited in the Protein Data Bank of the RCSB with the access number PDB ID: 1FFK, 1JJ2, 1FFZ, IFG0, 1K73, 1KC8, 1K8A, 1KD1 or 1K9M, and / or the atomic coordinates contained in Disk N.D1. In addition, the person skilled in the art, using the preceding atomic coordinates, can generate additional atomic coordinates through, for example, molecular modeling using, for example, homology modeling techniques and / or molecular replacement, which together define at least a portion of a ribosome model of another species of interest. Using the aforementioned atomic coordinates, the person skilled in the art can design protein synthesis inhibitors that can be adapted to be effective against the ribosomes of one or more species, but have little or no effect on the ribosomes of other species. Such inhibitors can be competitive inhibitors. As used herein, the term "competitive inhibitor" refers to an inhibitor that binds to the active form of a ribosome or ribosomal subunit at the same sites as its substrate (s) or tRNA (s). , thus competing directly with them. The term "active form" of a ribosome or a ribosomal subunit refers to a ribosome or a ribosomal subunit in a state that makes it capable of a protein synthesis. Competitive inhibition can be completely reversed by increasing the substrate or tRNA concentration.

This invention also allows the design of molecules that act as non-competitive inhibitors of protein synthesis. As used herein, the term "non-competitive inhibitor" refers to a molecule that inhibits the functional activity of a ribosome or ribosomal subunit by binding to a different site in the ribosome or ribosome subunit than what its substrates do, or tRNA Such inhibitors can often bind to the ribosome or ribosomal subunit with the substrate or tRNA and not to the ribosome or ribosome subunit itself. Non-competitive inhibition cannot be completely reversed by increasing the concentration of the substrate. These inhibitors may bind to, all or a portion of, the active sites or other regions of the large ribosomal subunit already bound to their substrate and may be potent and less specific than known competitive inhibitors competing for the active sites of the subunit. Large ribosomal or by union to the large ribosomal subunit.

Similarly, non-competitive inhibitors that bind to and inhibit protein synthesis, whether or not they are bound to another chemical entity, can be designed using the atomic coordinates of large or complex ribosomal subunits that comprise the large ribosomal subunit. of this invention. As used herein, the term "non-competitive inhibitor" refers to an inhibitor that can bind either to the free form.

or to the substrate or to the bound tRNA of the ribosome or ribosomal subunit.

Those skilled in the art can identify inhibitors as competitive or non-competitive by computer-adjusted enzymatic kinetic data using the standard equation according to Segel, IH, (1975) Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and Steady-State Enzyme Systems, (Wiley Classics Library). It should be understood that non-competitive inhibitors according to the present invention can bind the same different binding sites or sites.

Alternatively, the atomic coordinates provided by the present invention are useful for designing improved analogs of known inhibitors of protein synthesis or for designing new classes of inhibitors based on the atomic structures and crystal coordinates of the ribosomal subunit complex / CCdA-p -pure 50S and the ribosomal subunit complex / aa-tRNA analog 50S. This provides a new way to design protein synthesis inhibitors with high specificity, stability and other drug-like qualities (Lipinski et al. (1997) Adv. Drug Deliv. Rev. 23: 3).

The atomic coordinates of the present invention also allow probing the three-dimensional structure of a ribosome or ribosome subunit or a portion thereof with molecules composed of a variety of different chemical characteristics to determine optimal sites for interaction between candidate inhibitors and / or activators and the ribosome or ribosomal subunit. For example, high resolution atomic coordinates based on X-ray diffraction data collected from solvent-saturated crystals allow the determination of where each type of solvent molecule sticks. The small molecules that bind to these sites can then be designed and synthesized and then tested for their inhibitory activity (Travis, J. (1993) Science 262: 1374). In addition, any known antibiotic, inhibitor or other small molecule that binds to the large subunit of H. marismortui can be embedded in large subunit crystals of H. marismortui and its exact mode of binding can be determined from the maps Electronic density difference. These molecules may represent major compounds from which better pharmaceutical-type compounds can be synthesized.

b. Identification of the target sites.

The atomic coordinates of the invention allow the person skilled in the art to identify the target locations in a ribosome or a large ribosomal subunit that can serve as a starting point in a rational drug design. As a threshold, the atomic coordinates of the invention allow the person skilled in the art to identify specific regions within a ribosome or a ribosomal subunit that are involved in protein synthesis and / or protein secretion outside the ribosome. In addition, the atomic coordinates of the invention allow an expert in the art to also identify portions of these regions that are conserved or that are not conserved between different organisms. For example, identifying portions of these regions that are conserved among some pathogens, for example, some prokaryotes, but are not conserved in a

For example, a host organism, for example, a eukaryotic, more preferably a mammal, the person skilled in the art can design molecules that selectively inhibit or interrupt the synthesis activity of pathogen proteins but not the host's ribosomes. Furthermore, by analyzing the regions that are either conserved or not conserved among some pathogens, it may be possible to disseminate broad-spectrum or narrow-spectrum inhibitors of protein synthesis, for example, antibiotics, when a particular need arises.

10 Figure 32 is a schematic representation of a large ribosomal subunit that identifies a variety of illustrative target sites that appear to participate in protein synthesis within the ribosome and / or export

or translocation of the newly synthesized protein outside the ribosome. Target sites include, for example, the P site (200), the A site (201), the peptidyltransferase center (202), the peptidyltransferase site (203) which includes at least a portion of the P site and the site A, a domain that links a factor (204) including, for example, the domain

15 that binds EF-Tu and the domain that binds EF-G, the polypeptide exit tunnel (205) including cavities defined by the wall of the exit tunnel, and the binding domain of the signal recognition particle ( 206).

As an example, inspection of the atomic coordinates of the 50S ribosomal subunit of H. marismortui has identified a variety of target regions that can serve as a basis for the rational drug design of inhibitors of new or modified protein synthesis. Target regions include the peptidyltransferase site, the A site, the P site, the polypeptide exit tunnel, some cavities arranged in the wall of the polypeptide exit tunnel (eg, cavity 1 and cavity 2), and some antibiotic union pockets. The residues that together define at least one portion of each of the preceding regions are identified in the following tables. However, it is considered that the same sites or similar target sites can be identified in a ribosome or ribosomal unit of interest using the principles described in the

25 present. In addition, these principles can be employed using any of the primary sets of atomic coordinates provided herein or any additional sets of atomic coordinates, for example, sets of secondary atomic coordinates, which can be generated by molecular modeling of any ribosome or ribosomal subunit of interest.

Table 5A identifies residues in the 50S ribosomal subunit of H. marismortui that together define at least

30 a portion of the peptidyltransferase ribosomal site (layer 5.8 A). In addition, Table 5A identifies the corresponding residues that define at least a portion of the ribosomal site of peptidyltransferase in E. coli, Rattus, humans, and the large subunit in the human mitochondria. Table 5B identifies residues in the 50S ribosomal subunit of H. marismortui that together define a wider portion of the peptidyltransferase ribosomal site (layer 5.8 A-12.6 A). Unconserved residues were identified by comparison of

35 sequences of the 23S rRNA structure of H. marismortui or a ribosomal protein that forms the aforementioned sites with the corresponding aligned genomic DNA sequences encoding either the homologous 23S RNA or the ribosomal protein of the other organisms.

Table 5A Residues that define the ribosomal peptidyltransferase site (5.8 A envelope) It was determined that the residues that jointly define the peptidyltransferase site (5.8 A envelope) were those residues in the 50S ribosomal subunit that are within 5 , 8 angstroms of the CC-puromycin-A site ligand atoms and the CCdA-PO2 part of the CCdA-p transition state inhibitor

H. Marismortui residue

Residue corresponding to E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

G2102

G2061 G3632 G3876 G1054

A2103

A2062 A3633 A3877 A1055

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

C2106

C2065 C3636 C3880 C1058

G2284 G2252

G2251 G2252 G3971 G3918 G4156 G4157 G1145 G1146

G2286

G2253 G3919 G4158 G1147

A2474

A2439 A4106 A4345 A1256

H. Marismortui residue

Residue corresponding to E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

G2482

G2447 G4114 G4353 G1264

A2485

A2450 A4117 A4356 A1267

A2486

A2451 A4118 A4357 A1268

C2487

C2452 C4119 4358 C1269

A2488

A2453 U4120 U4359 A1270

U2528

U2493 U1460 U4399 U1310

G2529

G2494 G4161 G4400 A1311

C2536

C2501 C4168 C4407 C1318

A2538

A2503 A4170 A4409 A1320

G2540

G2505 G4172 G4411 G1322

U2541 G2542

U2506 C2507 U4173 C4174 U4412 C4413 U1323 U1324

G2543

G2508 G4175 G4414 G1325

G2588

G2553 G4220 G4459 G1370

U2589

U2554 U4221 U4460 U1371

U2590

U2555 U4222 U4461 U1372

C2608

C2573 C4240 C4479 C1390

G2617

G2582 G4249 G4488 A1399

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2620

U2585 U4252 U4491 U1402

C2636

C2601 C4268 C4507 C1418

A2637

A2602 A4269 A4508 A1419

G2638

G2603 G4270 G4509 G1420

5 puromycin (Access codes: 1fg0 and 1 ffz, respectively) using the MIDASa program. Conserved residues in 23S rRNA were determined by comparing sequences of the 23S rRNA of H. marismortui with genomic DNA sequences encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). Sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA), for which the default parameters were used.

10 a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619 c) Genbank access number J01695 d) Genbank access number 2624399 e) Genbank access number M11167 f) Genbank access number 13683 Table 5B. Waste defining the ribosomal peptidyltransferase site (envelope 5.8 A-12.6)

H. Marismortui residue

Waste corr. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

C1750

A1672 C2692 C2830 A732

C1982

C1941 C3515 C3757 A934

C1983

C1942 C3514 C3758 C935

U1984

U1943 U3515 U3759 U936

U1985

U1944 C3516 C3760 U937

U1996

U1955 U3527 U3771 U948

C2006 A2007

C1965 A1966 C3537 A3538 C3781 A3782 U958 A959

G2073

G2032 C3603 C3847 G1025

A2100

A2059 A3630 A2874 A1052

A2103

A2062 A3633 A3877 A1055

U2107

C2066 U3637 U3881 U1059

U2282

U2249 U3915 U4154 U1143

G2283

G2250 G3916 G4155 G1144

C2287

C2254 C3920 C4159 C1148

G2288

G2255 G3921 G4160 G1149

C2309

G2275 C3942 C4181 C1154

G2472

G2437 U4104 U4343 U1254

U2473

U2438 U4105 U4344 U1255

C2476

U2441 C4108 C4347 C1258

A2479

G2444 A4111 A4350 A1261

G2480

G2445 G4112 G4351 G1262

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

U2484

U2449 U4116 U4355 U1266

G2489

G2454 G4121 G4360 G1271

H. Marismortui residue

Waste corr. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

A2490

G2455 G4122 G4361 C1272

C2526

U2491 U4158 U4397 U1308

U2527

U2492 U4159 U4398 U1309

G2529

G2494 G4161 G4400 A1311

C2530

G2495 A4162 A4401 C1312

U2531

C2496 U4163 U4402 G1313

A2532

A2497 C4164 C4403 A1314

C2534

C2499 U4166 U4405 C1316

U2535

U2500 U4167 U4406 G1319

G2537

G2502 G4169 G4408 U1317

A2538

A2503 A4170 A4409 A1320

U2539

U2504 U4171 U4410 U1321

G2540

G2505 G4172 G4411 G1322

G2544

G2509 G4176 G4415 G1326

U2587

U2552 U4219 U4458 U1369

C2591

C2556 C4223 C4462 C1373

C2592

G2257 A4224 A4463 A1374

A2604

G2569 G4236 G4475 C1386

G2605

G2570 G4237 G4476 C1387

U2607

A2572 A4239 A4478 A1389

G2609

G2574 G4241 G4480 G1391

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

G2616

G2581 G4248 G4487 G1398

G2617

G2582 G4249 G4488 A1399

U2621

U2586 U4253 U4492 A1404

A2622

A2587 A4254 A4493 C1403

G2627

G2592 G4259 G4498 G1409

U2628

U2593 U4260 U4499 G1410

G2634

G2599 G4266 G4505 U1416

A2635

A2600 A4267 A4506 C1417

H. Marismortui residue

Waste corr. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

G2638

G2603 G4270 G4509 G1420

G2639

U2604 G4271 G4510 G1421

U2640

U2605 U4272 U4511 U1422

G2643

G2608 G4275 G4514 G1425

U2645

C2610 U4277 U4516 U1427

L3 protein

P1

NP� S2 S2 P2

L10E protein

D109 G110

XX XX T113 G114 T113 G114 XX XX

L10E protein

R112

XX R115 R116 XX

It was determined that the residues that jointly define the peptidyltransferase site (envelope 5.8 A-12.6 A) were those residues in the 50S ribosomal subunit that are within 5.8-12.6 angstroms of the atoms of the ligand CC-puromycin-A site and the atoms of the CCdA-PO2 part of the transition state inhibitor of 5 CCdA-p-puromycin (Access codes: 1fg0 and 1 ffz, respectively) using the MIDASa program. Conserved residues in 23S rRNA were determined by comparing sequences of the 23S rRNA of H. marismortui with the corresponding aligned genomic DNA sequences encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the case of ribosomal proteins, comparisons were also made between the structural sequences of H. marismortui proteinasg and the corresponding aligned protein sequences encoding homologous proteins in E.colii, rattus norvegicusj, humanok or human mitochondria l. Sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) by using default parameters.

�NP means that no homologous residue has been identified. XX means that no homologous protein has been identified in these species.

15 a) a. T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27,36-37.

b) Genbank access number AF034619 c) Genbank access number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

20 f) Genbank accession number 13683 g) Genbank accession number AAA86859 for the L3 protein; Genbank accession number 15825950 for the L10E protein.

i) Genbank accession number CAA26460 for the L3 protein.

j) Genbank accession number P21531 for the L3 protein; Genbank accession number NP 112362 for the L10E protein.

25 k) Genbank accession number NP 000958 for the L3 protein; Genbank accession number NP 006004 for the L10E protein.

l) Genbank accession number P09001 for the L3 protein.

Table 6A identifies residues in the ribosomal subunit of H. marismortui 50S that together define at least a portion of ribosomal site A (envelope 5.8 A). In addition, Table 6A identifies the corresponding residues that define at least a portion of the ribosomal site A in E. coli, Rattus, human, and large subunit of human mitochondria. Table 6B identifies the residues in the H. marismortui 50S ribosomal subunit that together define a broader portion of ribosomal site A (envelope 5.8 A-12.6 A). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 6A Wastes that define ribosomal site A (5.8 A envelope)

Marismortui residue

H. Corresponding residue in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

G2102

G2061 G3632 G3876 G1054

C2104

C2063 C3634 C3878 C1056

A2485

A2450 A4117 A4356 A1267

A2486

A2451 A4118 A4357 A1268

C2487

C2452 C4119 C4358 C1269

A2488

U2453 U4120 U4359 A1270

U2528

U2493 U4160 U4399 U1310

G2529

G2494 G4161 U44009 A1311

C2536

C2501 C4168 C4407 C1318

U2541

U2506 U4173 U4412 U1323

C2542

C2507 C4174 C4413 U1324

G2543

G2508 G4175 G4414 G1325

G2588

G2553 G4220 G4459 G1370

U2589

U2554 U4421 U4460 U1371

U2590

U2555 U4222 U4461 U1372

C2608

C2573 C4240 C4479 C1390

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 G4490 U1401

U2620

U2585 U4252 U4491 U1402

It was determined that the residues that jointly define site A (5.8 A envelope) were those residues in the

50S ribosomal subunit that are within 5.8 angstroms of the ligands of the CC-puromycin 10 A site (PDB access code 1fg0) through the MIDASa program. Conserved residues in 23S rRNA were determined

by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding

Genomic aligned DNA sequences encoding the 23S homologous rRNA (E. colic, Rattus norvegicusd, humanoe

or human mitochondria f). Sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) by using default parameters.

15 a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399 e) Genbank access number M11167 f) Genbank access number 13683 Table 6B. Waste defining ribosomal site A (enveloped 5.8 A-12.6 A)

Marismortui residue

H. Corresponding residue in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C1750

A1672 C2692 C2830 A732

C1982

C1941 C3513 C3757 A934

C1983

C1942 C3514 C3758 C935

U1984

U1943 U3515 U3759 U936

U1985

U1944 C3516 C3760 U937

U1996

U1955 U3527 U3771 U948

G2073

G2032 C3603 C3847 G1025

A2100

A2059 A3630 A3874 A1052

A2103

A2062 C3633 A3877 A1055

C2105

C2064 C3635 C3879 C1057

C2106

C2065 C3636 C3880 C1058

U2107

C2066 U3637 U3881 U1059

G2284

G2251 G3917 G4156 G1145

G2285

G2252 G3918 G4157 G1146

A2474

A2439 A4106 A4345 A1256

A2479

G2444 A4111 A4350 A1261

G2480

G2445 G4112 G4351 G1262

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

U2484

U2449 U4116 U4355 U1266

G2489

G2454 G4121 G4360 G1271

A2490

G2455 G4122 G4361 C1272

C2526

U2491 U4158 U4397 U1308

U2527

U2492 U4159 U4398 U1309

C2530

G2495 A4162 A4401 C1312

U2531

C2496 U4163 U4402 G1313

A2532

A2497 C4164 C4403 A1314

C2534

C2499 U4166 U4405 C1316

U2535

U2500 U4167 U4406 U1317

G2537

G2502 G4169 G4408 G1319

A2538

A2503 A4170 A4409 A1320

U2539

U2504 U4171 U4410 U1321

G2540

G2505 G4172 G4411 G1322

G2544

G2509 G4176 G4415 G1326

U2587

U2552 U4219 U4458 U1369

C2591

C2556 C4223 C4462 C1373

G2592

G2257 A4224 A4463 A1374

A2604

G2569 G4236 G4475 C1386

G2605

G2570 G4237 G4476 C1387

U2607

A2572 A4239 A4478 A1389

G2609

G2574 G4241 G4480 G1391

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

G2616

G2581 G4248 G4487 G1398

G2617

G2582 G4249 G4488 A1399

U2621

U2586 U4253 U4492 C1403

C2636

C2601 C4268 C4507 C1418

A2637

A2602 A4269 A4508 A1419

A2637

A2602 A4269 A4508 A1419

G2638

G2603 G4270 G4509 G1420

G2639

U2604 G4271 G4510 G1421

U2640

U2605 U4272 U4511 U1422

G2643 U2645

G2608 C2610 G4275 U4277 G4514 U4516 G1425 U1427

L3 protein

P1

NP� S2 S2 P2

It was determined that the residues defining site A (enveloped 5.8 A-12.6 A) were those residues in the 50S ribosomal subunit that are within 5.8-12.6 angstroms of the ligand atoms of the CCpuromycin A site (PDB access code 1fg0) using the MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui 5 with the corresponding sequences of the aligned genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or human mitochondria f ). In the cases of ribosomal proteins, comparisons were also made between the sequence g protein structure of H. marismortui and the corresponding sequences of aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicusj, human or human mitochondria. The sequence alignments were determined by the program

10 MegAlign (DNASTAR, Madison, Wisconsin, USA) with default parameters. �NP means that no homologous residue was identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695 d) Genbank access number 2624399 e) Genbank access number M11167 f) Genbank access number 13683

5 g) Genbank accession number AAA86859 for the L3 protein. i) Genbank accession number CAA26460 for the L3 protein. j) Genbank accession number P21531 for the L3 protein. k) Genbank NP 000958 access number for the L3 protein. l) Genbank accession number P09001 for the L3 protein.

10 Table 7A identifies residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of ribosomal site A (envelope 5.8 A). In addition, Table 7A identifies the corresponding residues that define at least a portion of the ribosomal P site in E. coli, Rattus, human, and large subunit of human mitochondria. Table 7B identifies the residues in the H. marismortui 50S ribosomal subunit that together define a broader portion of the P · ribosomal site (envelope 5.8 A-12.6 A). Waste not

15 conserved have been identified as previously described with respect to Tables 5A and 5B. Table 7A Waste defining the ribosomal site P (enveloped 5.8 A)

Marismortui residue

H. Corresponding residue in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

C2106

C2065 C3636 C3880 C1058

G2284

G2251 G3917 G4156 G1145

G2285

G2252 G3918 G4157 G1146

G2286

G2253 G3919 G4158 G1147

A2474

A2439 A4106 A4345 A1256

A2485

A2450 A4117 A4356 A1267

A2486

A2451 A4118 A4357 A1268

U2619

U2584 U4251 U4490 U1401

U2620

U2585 U4252 U4491 U1402

C2636

C2601 C4268 C4507 C1418

A2637

A2602 A4269 A4508 A1419

G2638

G2603 G4270 G4509 G1420

It was determined that the residues that jointly define the P site (5.8 A envelope) were those residues in the ribosomal 50S ribosomal subunit that are within 5.8 angstroms of the atoms of the CCdA-PO2 part of the state inhibitor of CCdA-p-puromycin transition (Figure 10a) (PDB access code 1 ffz) using the

20 MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding sequences of the aligned genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or human mitochondria f) . The sequence alignments were determined using the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters.

 a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37. b) Genbank access number AF034619 c) Genbank access number J01695 d) Genbank access number 2624399 e) Genbank access number M11167 f) Genbank access number 13683 Table 7B. Waste defining the ribosomal site P (wrapped 5.8 A-12.6 A)

Marismortui residue

H. Corresponding residue in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C1982

C1941 C3513 C3757 A934

C2006

C1965 C3537 C3781 U958

A2007

A1966 A3538 A3782 A959

G2102

G2061 G3632 G3876 G1054

A2103

A2062 A3633 A3877 A1055

U2107

C2066 U3637 U3881 U1059

U2282

U2249 U3915 U4154 U1143

G2283

G2250 G3916 G4155 G1144

C2287

C2254 C3920 C4159 C1148

G2288

G2255 G3921 G4160 G1149

C2309

C2275 C3942 C4181 C1154

C2472

G2437 U4104 U4343 U1254

U2473

U2438 U4105 U4344 U1255

C2476

U2441 C4108 C4347 C1258

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

U2484

U2449 U4116 U4355 U1266

C2487

C2452 C4119 C4358 C1269

A2488

A2453 U4120 U4359 A1270

U2528

U2493 U4160 U4399 U1310

G2529

G2494 G4161 G4400 A1311

C2530

G2495 A4162 A4401 C1312

U2531

C2496 U4163 U4402 G1313

A2532

A2497 C4164 C4403 C1314

C2534

C2499 U4166 U4405 C1316

U2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

A2538

A2503 A4170 A4409 A1320

G2540

G2505 G4172 G4411 G1322

U2541

U2506 U4173 U4412 U1323

C2542

C2507 C4174 C4413 U1324

C2608

C2573 C4240 C4479 C1390

G2611

G2576 G4243 G4482 G1393

G2618

G2583 G4250 G4489 G1400

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

G2627

G2592 G4259 G4498 G1409

U2628

U2593 U4260 U4499 G1410

G2634

G2599 G4266 G4505 U1416

A2635

A2600 A4267 A4506 C1417

G2639

U2604 G4271 G4510 G1421

U2640

U2605 U4272 U4511 U1422

G2643

G2608 G4275 G4514 G1425

L10E protein

D109

XX T113 T113 XX

G110

XX G114 G114 XX

R112

XX R115 R116 XX

It was determined that the residues that jointly define the P site (envelope 5.8 A-12.6 A) were those residues in the 50S ribosomal subunit that are within the atoms of the CCdA-PO2 part of the transition state inhibitor CCdA-p-puromycin (Figure 10a) (PDB 1fffz access codes) using the MIDASa 5 program. The residues conserved in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding aligned genomic DNA sequences encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the case of ribosomal proteins, comparisons were also made between the structural sequences of H. marismortui proteinasg and the corresponding aligned protein sequences encoding homologous proteins in E. colii

10 rattus norvegicusj, humanok or human mitochondria. Sequence alignments were determined using the MegAlign program (DNASTAR, Madison, Wisconsin, USA) under the use of default parameters. �� XX means that no homologous protein has been identified in these species.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37. b) Genbank access number AF034619

15 c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683 g) Genbank access number 15825950 for the L10E protein.

i) There is no Genbank Access Number for the L10E protein.

j) Genbank NP 112362 access number for the L10E protein.

k) Genbank NP 006004 access number for the L10E protein.

5 l) There is no Genbank Access Number that compares with the L10E protein

Table 8A identifies the residues in the H. marismortui 50S ribosomal subunit that are within 10 angstroms of a hypothetical nascent polypeptide within the ribosomal exit tunnel (envelope 10 A). In addition, Table 8A identifies the corresponding residues that define at least a portion of the ribosomal polypeptide exit tunnel in E. coli, Rattus, human, and large subunit of human mitochondria. On the other hand, in

10 Table 8B identifies residues in the H. marismortui 50S ribosomal subunit that are found within 10-15 angstroms of a nascent polypeptide within the ribosomal exit tunnel (10-15 angstroms envelope). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 8A Waste defining the ribosomal exit tunnel (wrapped 10 A)

Marismortui residue

H. Corresponding residue in E-coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A60

A64 C729 � C682 PN�

G88

G93 G764 G713 NP

G89

A94 G765 G714 NP

A90

A95 C766 C715 NP

A462

C456 C1112 C1195 NP

U465

U459 G1116 G1199 NP

A466

A460 G1117 G1200 NP

G467

C461 G1118 G1201 NP

U468

C462 G1119 A1202 NP

G75

G460 U1127 U1208 NP

A476

A470 C1128 C1209 NP

A477

A471 A1129 U1210 NP

C478

A472 C1130 C1211 NP

U488

A482 C1140 C1221 NP

A497

G491 G1151 G1233 NP

A498

A492 G1152 A1234 NP

A513

A507 A1167 G1249 NP

G514

A508 A1168 A1250 NP

A767

A676 A1425 A1503 NP

U835

U744 U1491 U1570 NP

C839

U746 G1495 G1574 NP

U840

U747 U1496 U1575 NP

A841

G748 G1497 G1576 NP

A844

A751 A1500 A1579 NP

U845

A752 A1501 A1580 NP

U1359

U1255 U2191 U2327 U351

C1360

G1256 C2192 C2328 G352

C1361

C1257 U2193 U2329 U353

U1362

U1258 U2194 U2330 C354

G1363

G1259 G2195 G2331 C355

G1364

A1260 G2196 G2332 A356

A1424

U1318 G2255 G2391 G412

G1425

C1319 U2256 U2392 U413

C1426

C1320 G2257 G2393 A414

A1427

A1321 A2258 A2394 A415

C1428

A1322 A2259 A2395 A416

U1429

C1323 C2260 C2396 U417

G1430

G1324 A2261 A2397 NP

C1439

G1333 A2270 A2406 U426

U1440

G1334 C2271 C2407 A427

G1441

C1225 A2272 A2408 G428

A1442

A1336 U2273 U2409 U429

A1689

A1814 C2635 C2771 A680

C1690

C1615 A2636 A2772 U681

G1837

U1781 A3368 A3612 C831

U1838

U1782 U3369 U3613 C832

A2054

A2013 A3585 A3829 A1006

A2055

A2014 A3586 A3830 A1007

A2055

A2014 A3586 A3830 A1007

C2056

A2015 A3587 A3831 A1008

U2057

U2016 C3588 C3832 U1009

C2098

G2057 A3628 A3872 A1050

G2099

A2058 G3629 G3873 G1051

A2100

A2059 A3630 A3874 A1052

A2101

A2060 A3631 A3875 A1053

G2102

G2061 G3632 G3876 G1054

A2103

A2062 A3633 A3877 A1055

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

A2485

A2450 A4117 A4356 A1267

A2486

A2451 A4118 A4357 A1268

C2487

C2452 C4119 C4358 C1269

A2488

A2453 U4120 U4359 A1270

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

A2538

A2503 A4170 A4409 A1320

U2539

U2504 U4171 U4410 U1321

G2540

G2505 G4172 G4411 G1322

U2541

U2506 U4173 U4412 U1323

C2542

C2507 C4174 C4413 U1324

C2608

C2573 C4240 C4479 C1390

G2611

G2576 G4243 C4482 G1393

U2620

U2585 U4252 U4491 U1402

G2616

G2581 G4248 G4487 G1398

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2621

U2586 U4253 U4492 C1403

A2637

A2602 A4269 A4508 A1419

C2644

U2609 U4276 U4515 U1426

U2645

C2610 U4277 U4516 U1427

G2646

C2611 U4278 U4517 U1428

C2647

C2612 U4279 U4518 C1429

L4 protein

E59

E51 E65 E65 R70

S60

V52 S66 S66 G71

F61

T53 W67 W67 F72

G62

G54 G68 G68 E73

S63

S55 T69 T69 Q74

G64

G56 G70 G70 E75

R65

K57 R71 R71 R76

G56

NP� A72 A72 G78

Q67

NP V73 V73 L79

A68

NP A74 A74 A80

H69

K58 R75 R75 D81

V70

P59 I76 I76 L82

P71

W60 P77 P77 H83

K72

R61 R78 R78 P84

L73

Q62 V79 V79 D85

D74

G64 G81 G81 F87

G75

T65 G82 G82 A88

R76

G66 G83 G83 T89

A77

R67 T84 T84 A90

L22 protein

E20

H9 N21 N21 Q73

E121

S81 K124 K124 P177

Q122

M82 M125 M125 P178

Q123

K83 R126 R126 P179

G124

R84 R127 R127 P180

R125

185 R128 R128 E181

K126

M86 T129 T129 P182

L22 protein

P127

P87 Y130 Y130 P183

R128

R88 R131 R131 K184

A129

A89 A132 A132 A186

M130

K90 H133 H133 V187

G131

G91 G134 G134 A188

R132

R92 R135 R135 H189

A133

A93 I136 I136 A190

S134

D94 N137 N137 K191

A135

R95 P138 P138 E192

W136

196 Y139 Y139 Y193

N137

L97 M140 M140 1194

Q140

T100 P143 P143 F197

L39E protein

N18

XX N20 N20 N98

S19

XX R21 R21 H106

R20

XX P22 P22 R107

V21

XX I23 123 I108

P22

XX P24 P24 G109

A23

XX Q25 Q25 D110

Y24

XX W26 W26 F111

V25

XX I27 I27 I112

M26

XX R28 R28 D113

L27

XX M29 M29 V114

K28

XX K30 K30 S115

T29

XX T31 T31 E116

D30

XX G32 G32 G117

E31

XX N33 N33 P118

R35

XX Y37 Y37 H135

N36

XX N38 N38 N136

H37

XX S39 S39 L137

K38

XX K40 K40 Q138

R39

XX R41 R41 R146

R40

XX R42 R42 R147

H41

XX H43 H43 H156

R44

XX R46 R46 R170

L29E protein

N45

XX T47 T47 S171

It was determined that the residues in the envelope of 10 angstroms were those residues in the 50S ribosomal subunit that are within 10 angstroms of the atoms of a recently synthesized peptide model positioned in the center of the exit tunnel, for which it was used the MIDASa program. Conserved residues in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding aligned genomic DNA sequences encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondrial humanof). In the case of ribosomal proteins, comparisons were also made between the structural sequences of proteinasg of H. marismortui and the corresponding aligned protein sequences encoding homologous proteins in E.colii, rattus norvegicus j, humanok or human mitochondria. Sequence alignments were determined with the MegAlign program (DNASTAR,

10 Madison, Wisconsin, USA UU.) By using default parameters. �NP means that no homologous residue has been identified. XX means that no homologous protein has been identified in these species.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619 c) Genbank accession number J01695 d) Genbank access number 2624399 e) Genbank access number M11167

5 f) Genbank access number 13683

g) Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS22 for the L22 protein; Genbank accession number P22452 for the L39E protein. i) Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the

L22 protein; there is no Genbank accession number for the L39E protein

10 j) Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein; Genbank accession number CAA57900 for the L39E protein. k) Genbank accession numbers P36578, NP 00959, S39803 and T09551 for the L4 protein; access number to

Genbank XP 057521 for the L22 protein; Genbank accession number NP 000991 for the L39E protein. l) Genbank XP access number 049502 for the L4 protein; Genbank XP accession number 051279 for the L22 protein; Genbank accession number NP 059142 and NP 542984 for the L39E protein. Table 8B Waste defining the ribosomal peptide exit tunnel (wrapped 10 A-15 A)

Marismortui residue

H. Corresponding residue in E-coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

U22

U25 G686 G637 NP�

G23

G26 G687 G638 NP

G24

G27 G688 G639 NP

C57

C61 C726 U679 NP

C58

U62 U727 U680 NP

A59

A63 U728 U681 NP

G61

U65 C730 C683 NP

A86

A91 A762 A711 NP

C87

U92 C763 C712 NP

G91

C96 U767 U716 NP

U454

U448 C1104 C1187 NP

C461

C455 C1111 C1194 NP

A463

A457 G1114 G1197 NP

G469

G463 G1120 G1203 NP

U470

U464 C1122 NP NP

A473

G467 C1125 U120 NP

C474

G468 G1126 C1207 NP

G479

G473 G1131 G1212 NP

C480

G474 C1132 G1213 NP

A485

A479 U1137 A1218 NP

A486

A480 C1138 C1219 NP

G487

G481 C1139 G1220 NP

A489

A483 U1141 G1222 NP

C490

C484 C1142 C1223 NP

C491

C485 C1143 G1224 NP

A495

G489 U1147 C1228 NP

G496

C490 C1148 C1229 NP

G499

G493 G1153 A1235 NP

G512

G506 G1166 G1248 NP

C515

C509 G1169 G1251 NP

C633

U576 G1284 G1365 NP

G634

G577 C1285 C1366 NP

A635

G578 A1286 A1367 NP

G636

G579 A1287 A1368 NP

C637

U580 U1288 U1369 NP

C638

C581 G1289 G1370 NP

C764

C673 C1422 C1500 NP

G765

G674 G1423 G1501 NP

A766

A675 A1424 A1502 NP

G836

G745 G1492 G1571 NP

U837

NP� A1493 A1572 NP

C838

NP C1494 C1573 NP

C842

A749 C1498 C1577 NP

A843

A750 A1499 A1578 NP

A846

A753 U1502 U1581 NP

A882

A789 A1538 A1617 NP

U883

U790 U1539 U1618 NP

C884

C791 C1540 C1619 NP

G885

A792 G1541 G1620 NP

C889

C796 C1545 C1624 NP

C890

G797 A1546 A1625 NP

A1358

A1254 A2190 A2326 U350

C1365

C1261 U2197 U2333 A357

C1366

A1262 G2198 G2334 G358

A1367

U1263 G2199 G2335 A359

U1368

A1264 U2200 U2336 U360

A1369

A1265 A2201 A2337 A361

U1419

U1313 U2250 U2386 C407

C1423

G1317 U2254 Y2390 U411

U1432

U1326 C2263 C2399 U419

C1436

C1330 U2267 U2403 U423

A1437

G1331 G2268 G2404 G424

G1438

G1332 A2269 A2405 U425

G1443

G1337 G2274 G2410 C430

G1688

G1613 G2634 G2270 G679

A1691

A1616 G2637 G2773 U682

C1692

C1617 C2638 C2774 A683

A1836

A1780 A3367 A3611 A830

A1839

A1783 U3370 U3614 A833

U2052

U2011 C3583 C3827 U1004

G2053

G2012 G3584 G3828 G1005

G2058

U2017 C3589 C3833 U1010

U2059

G2018 NP NP G1011

G2073

G2032 C3603 C3847 G1025

G2097

G2056 A3627 A3871 G1049

C2106

C2065 C3636 C3880 C1058

U2107

C2066 U3637 U3881 U1059

G2284

G2251 G3917 G4156 G1145

G2285

G2252 G3918 G4157 G1146

U2473

U2438 U4105 U4344 U1255

C2475

C2440 C4107 C4346 C1257

C2476

U2441 C4108 C4347 C1258

A2479

G2444 A4111 A4350 A1261

G2480

G2445 G4112 G4351 G1262

U2484

U2449 U4116 U4355 U1266

G2489

G2454 G4121 G4360 G1271

U2528

U2493 U4160 U4399 U1310

G2529

G2494 G4161 G4400 A1311

U2531

C2496 U4163 U4402 G1313

A2532

A2497 C4164 C4403 A1314

C2534

C2499 U4166 U4405 C1316

U2535

U2500 U4167 U4406 U1317

G2543

G2508 G4175 G4414 G1325

G2588

G2553 G4220 G4459 G1370

U2607

A2572 A4239 A4478 A1389

U2610

C2575 U4242 U4481 U1392

A2612

A2577 A4244 A4483 A1394

G2613

G2578 G4245 G4484 U1395

C2614

C2579 C4246 C4485 C1396

U2615

U2580 U4247 U4486 U1397

G2617

G2582 G4249 G4488 A1399

A2622

A2587 A4254 A4493 A1404

C2636

C2601 C4268 C4507 C1418

G2639

U2604 G4271 G4510 G1421

G2642

G2607 A4274 A4513 G1424

G2643

G2608 G4275 G4514 G1425

U2648

U2613 A4280 A4519 U1430

L4 protein

T56

T48 T62 T62 E67

P57

R49 S63 R63 / S63n S68

A58

A50 A64 A64 L69

R78

A68 H85 H85 P91

R79

R69 R86 R86 R92

V80

S70 S87 S87 L96

Q82

NP Q89 Q89 Q98

A83

S72 G90 G90 V99

L4 protein

V84

I73 A91 A91 A100

K85

K74 F92 F92 M101

L22 protein

R19

NP S20 S20 R72

R21

A10 L22 L22 I74

Q22

R11 R23 R23 K75

V119

G79 A122 A122 G175

G120

P80 P123 P123 P176

S138

K98 S141 S141 Q195

P139

R99 S142 S142 Q196

L24 protein

K81

N74 K89 K89 Y95

R83

A76 N91 N91 Y97

G84

T77 G92 G92 I98

E85

G78 T93 T93 G99

L29 protein

G37

G35 �� G38 G38 XX

G38

NP G39 G39 XX

A39

Q36 A40 A40 XX

P40

L37 A41 A41 XX

E41

Q38 S42 S42 XX

P43

S40 L44 L44 XX

L37E protein

G4

XX  G4  G4 XX

T5

XX  T5  T5 XX

L39E protein

L14

XX K16 K16 K94

D15

XX Q17 Q17 A95

N16

XX K18 K18 S96

Q17

XX Q19 Q19 Q97

W24

XX W26 W26 F111

W42

XX W44 W44 L157

R43

XX R45 R45 R158

L39E protein

D46

XX K48 K48 R172

It was determined that the residues in the envelope of 10-15 angstroms were those residues in the 50S ribosomal subunit that are within 10 -15 angstroms of the atoms of a recently synthesized peptide model positioned in the center of the exit tunnel, for which used the MIDASa program. Conserved residues in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. 5 marismortui with the corresponding aligned genomic DNA sequences encoding the homologous 23S rRNA

(E. colic, Rattus norvegicusd, humanoe or mitochondrial humanof). In the case of ribosomal proteins, comparisons were also made between the structural sequence of H. marismortui proteinag and the corresponding aligned protein sequences encoding homologous proteins in E.colii, rattus norvegicusj, humanok or mitochondria human. Sequence alignments were determined with the MegAlign program (DNASTAR,

10 Madison, Wisconsin, USA UU.) By using default parameters. �NP means that no homologous residue has been identified. XX means that no homologous protein has been identified in these species.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619B

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS22 for the L22 protein; Genbank accession number R5HS22 for the L24 protein; Genbank accession number R5HS29 for the L29 protein; Genbank accession number P32410 for the L37E protein; Genbank accession number P22452 for the L39E protein.

i) Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein; Genbank accession number R5EC24 for the L24 protein; Genbank accession number R5EC29 for the L29 protein; there is no Genbank accession number for the L37E protein; There is no Genbank accession number for the L39E protein.

j) Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein; Genbank accession number P12749 for the L24 protein; Genbank accession number R5RT35 for the L29 protein; Genbank accession number CAA47012 for the L37E protein; Genbank accession number CAA57900 for the L39E protein.

k) Genbank accession numbers P36578, NP 00959, S39803, and T09551 for the L4 protein; Genbank accession number XP-057521 for the L22 protein; Genbank accession number AAA60279 and NP 000978 for the L24 protein; Genbank accession number AAA51648 for the L29 protein; Genbank accession number NP 000988 for the L37E protein; Genbank accession number NP 000991 for the L39E protein.

l) Genbank XP access number 049502 for the L4 protein; Genbank XP accession number 051279 for the L22 protein; Genbank XP accession number 056380 for the L24 protein; there is no Genbank accession number for the L29 protein; there is no Genbank accession number for the L37E protein; Genbanks accession number NP 059142 and NP 542984 for the L39E protein.

n) R63 is present in the GenBank sequences T09551 and S39803, while S63 is present in the GenBank sequences P36578 and NP 000959

Figure 30 shows a region of the large ribosomal subunit in which an antibiotic is ligated. Figure 30 (A) shows an enlarged portion of the large ribosomal subunit with the tylosin antibiotic attached at the top of the polypeptide exit tunnel adjacent to the peptidyltransferase site. Figures 30 (B) and 30 (C) are views showing each half of a large ribosomal subunit cut along the polypeptide exit tunnel and are intended to guide the reader to show the tylosin binding site with respect to to the great ribosomal subunit as a whole. Figure 30 (A) also shows two cavities defined by the wall of the polypeptide exit tunnel bearing the designation "cavity 1" and "cavity 2". In addition, Figure 30 (A) also shows a disacarido joint pocket. The direction in which the newly synthesized polypeptide chains exit the ribosome through the polypeptide exit tunnel is indicated by an arrow.

Table 9 identifies the residues in the H. marismortui 50S ribosomal subunit that together define a first cavity within the wall of the polypeptide exit tunnel (cavity 1). In addition, Table 9 identifies which of the residues define corresponding residues of cavity 1 in E. coli, Rattus, human and human mitochondria. Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 9. Waste defining cavity 1 in the ribosomal peptide exit tunnel

H. Marismortui residue

Residue corresponding to E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

C474

G468 G1126 C1207 NP�

A766

A675 A1424 A1502 NP

H. Marismortui residue

Residue corresponding to E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

A767

A676 A1425 A1503 NP

U768

A677 A1426 A1504 NP

U883

U790 U1539 U1618 NP

C884

C791 C1540 C1619 NP

G885

A792 G1541 G1620 NP

A886

A793 A1542 A1621 NP

U888

C795 C1544 C1623 NP

C889

C796 C1545 C1624 NP

C890

G797 A1546 A1625 NP

U1359

U1255 U2191 U2327 U351

G1837

U1781 A3368 A3612 C831

A2100

A2059 A3630 A3874 A1052

A2101

A2060 A3631 A3875 A1053

G2102

G2061 G3632 G3876 G1054

A2103

A2062 A3633 A3877 A1055

C2475

C2440 C4107 C4346 C1257

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

A2479

G2444 A4111 A4350 A1261

A2538

A2503 A4170 A4409 A1320

L4 protein

P57

R49 S63 R63 / S63n S68

A58

A50 A64 A64 L69

E59

E51 E65 E65 R70

S60

V52 S66 S66 G71

F61

T53 W67 W67 F72

G62

G54 G68 G68 E73

S63

S55 T69 T69 Q74

G64

G56 G70 G70 E75

R65

K57 R71 R71 R76

Q67

NP� V73 V73 L79

V70

P59 176 176 L82

H. Marismortui residue

Residue corresponding to E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

P71

W60 P77 P77 H83

L4 protein

K72

R61 R78 R78 P84

L73

Q62 V79 V79 D85

D74

G64 G81 G81 F87

G75

T65 G82 G82 A88

R76

G66 G83 G83 T89

The cavities residues were identified, for which the MIDASa program was used. Conserved residues in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding aligned genomic DNA sequences encoding the 23S rRNA homologob (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, they were also performed

5 comparisons between the structural sequence of H. marismortui proteinag and the corresponding aligned protein sequences encoding homologous proteins in E. colii rattus norvegicusj, humanok or human mitochondria. Sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA). �NP means that no homologous residue has been identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis and R. Langridge. (1988). "The MIDAS Display System," J. Mol. Graphics, 6 10 (1): 13-27, 36-37. b) Genbank access number AF034619 c) Genbank access number J01695 d) Genbank access number 2624399 e) Genbank access number M11167 15 f) Genbank access number 13683 g) Genbank access number P12735 for L4 protein. i) Genbank accession number CAA26461 for the L4 protein. j) Genbank accession number JC4277 for the L4 protein. k) Access numbers to Genbank P36578, NP 00959, S39803 and T09551 for the L4 protein.

20 l) Genbank XP access number 049502 for the L4 protein. n) R63 is present in the GenBank sequences T09551 and S39803 while S63 is present in the GenBank sequences P36578 and NP 000959.

Table 10 identifies residues in the H. marismortui 50S ribosomal subunit that together define a second cavity in the wall of the polypeptide exit tunnel (cavity 2). In addition, Table 10 identifies

25 of which residues define corresponding residues of cavity 2 in E. coli, Rattus, human, and human mitochondria. Unconserved residues were identified as previously described with respect to Tables 5A and 5B.

Table 10. Waste defining cavity 2 in the ribosomal peptide exit tunnel

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

U831

C740 U1488 U1567 � NP

U832

U741 C1489 C1568 NP

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

G833

A742 C1490 C1569 NP

G834

A743 NP NP NP

U835

U744 U1491 U1570 NP

G836

G745 G1492 G1571 NP

U837

NP A1493 A1572 NP

C838

NP C1494 C1573 NP

C839

U746 G1495 G1574 NP

U840

U747 U1496 U1575 NP

A841

G748 G1497 G1576 NP

A843

A750 A1499 A1578 NP

A844

A751 A1500 A1579 NP

U845

A752 A1501 A1580 NP

A846

A753 U1502 U1581 NP

C847

U754 C1503 C1582 NP

C848

U755 G1504 G1583 NP

C849

A756 G1505 G1584 NP

C1753

C1675 C2695 C2833 NP

A1754

A1676 A2696 A2834 U734

G1837

U1781 A3368 A3612 C 831

U1838

U1782 U3369 U3613 C832

A1839

A1783 U3370 U3614 A833

G2099

A2058 G3629 G3873 G1051

A2100

A2059 A3630 A3874 A1052

A2103

A2062 A3633 A3877 A1055

U2615

U2580 U4247 U4486 U1397

G2616

G2581 G4248 G4487 G1398

U2621

U2586 U4253 U4492 C 1403

A2622

A2587 A4254 A4493 A1404

G2643

G2608 G4275 G4514 G1425

C2644

U2609 U4276 U4515 U1426

U2645

C2610 U4277 U4516 U1427

G2646

C2611 U4278 U4517 U1428

C2647

C2612 U4279 U4518 C1429

Cavity residues were identified by the MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding aligned genomic DNA sequences encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria humanof). Sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA). �NP means that no homologous residue has been identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

However, Tables 9 and 10 define only two of many cavities arranged within the wall of the polypeptide exit tunnel. However, by using atomic coordinates and the nuclear modeling methodologies described herein, the person skilled in the art can identify the residues (contributed by amino acids, nucleotides or a combination of both) that together define other cavities within the wall of the exit tunnel of polypeptide.

Furthermore, thanks to the use of the atomic coordinates described herein, the person skilled in the art can identify the antibiotic binding site of any antibiotic of interest. This information also provides contact sites between an antibiotic and residues in a ribosome or ribosomal subunit, which can be used advantageously in the design of novel or modified inhibitors of proteins. The binding or contact sites for a variety of antibiotic are set forth in greater detail in the following.

Table 11A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least one portion of an anomyomycin binding pocket (wrapped 5.8 angstroms). In addition, Table 11A identifies the corresponding residues that define at least a portion of the anisomycin binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. Table 11B identifies residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of an anisomycin binding pocket (envelop 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 11A Residues that define the anisomycin binding pocket (5.8 A envelope) It was determined that the residues that define the union pocket (5.8 angstroms envelope) are those residues in the 50S ribosomal subunit that are located within 5 , 8 angstroms of the anisomycin atoms, through the MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the 23S rRNA structure of H. marismortui with the corresponding aligned genomic DNA sequences that

H. Marismortui residue

Residue correresp. in E.coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

A2100v

A2059 A3630 A3874 A1052

G2102v

G2061 G3632 G3876 G1054

A2103v

A2062 A3633 A3877 A1055

G2482v

G2447 G4114 G4353 G1264

A2486v

A2451 A4118 A4357 A1268

C2487h, v

C2452 C4119 C4358 C1269

A2488v

A2453 U4120 U4359 U1270

U2535h, v

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

A2538v

A2503 A4170 A4409 A1320

U2539h, v

U2504 U4171 U4410 U1321

H. Marismortui residue

Residue correresp. in E.coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

G2540v

G2505 G4172 G4411 G1322

U2541v

U2506 U4173 U4412 U1323

U2629v

U2585 U4552 U4991 U1402

5 encode the 23S homologous rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria humanof). Sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27,36-37.

10 b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

15 h) Indicates a residue that is within the hydrogen bonding distance (3.5 angstroms) of the ligand; for example see Jorgensen, WL, Nguyen, TB. J. Comput. Chem. (1993) 14: 195-205.

v) residues in contact with van der Waals with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.).

20 Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen, WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 11B Waste defining the anisomycin binding pocket (5.8 A-12.6 A envelope

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

A631

A574 G1282 � G1363 NP

A632

A575 C1283 C1364 NP

C633

U576 G1284 G1365 NP

G2073

G2032 C3603 C3847 G1025

A2095

A2054 G3625 G3869 A1047

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 A1049

C2098

G2057 A3628 A3872 A1050

G2099

A2058 G3629 G3873 G1051

G2102

G2061 G3632 G3876 G1054

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

G2481

G2446 G4113 G1263 G4352

U2484

U2449 U4116 U4355 U1266

A2485

A2450 A4117 A4356 A1267

G2489

G2454 G4121 G4360 G1271

A2490

G2455 G4122 C G4361 1272

A2532

A2497 C4164 C4403 A1314

C2533

C2498 C4165 C4404 C1315

C2534

C2499 U4166 U4405 C1316

G2537

G2502 G4169 G4408 G1319

C2542

C2507 C4174 C4413 U1324

G2543

G2508 G4175 G4414 G1325

G2588

G2553 G4220 G4459 G1370

U2607

A2572 A4239 A4478 C1389

C2608

C2573 C4240 C4479 C1390

G2609

G2574 G4241 G4480 G1391

U2610

C2575 U4242 U4481 U1392

C2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

A2637

A2602 A4269 A4508 A1419

G2646

C2611 U4278 U4517 U1428

L3

W242

Q150 W257 W257 G246

The residues that define the junction pocket (wrapped 5.8 A-12.6 A) were determined as those residues in the 50S ribosomal subunit that are located within 5.8-12.6 angstroms of the anisomisin atoms for which was used the MIDASa program. The conserved residues were determined by comparison between the proposed secondary structures of H. marismortuib, E. colib, and Rattus norvegicusc, humanod, and mitochondria 5 humanae. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with the use of default parameters. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695 d) Genbank access number 2624399 e) Genbank access number M11167 f) Genbank access number 13683

5 g) Genbank accession number AAA86859 for the L3 protein. i) Genbank accession number CAA26460 for the L3 protein. k) Genbank NP 000964 access number for the L3 protein. l) Genbank accession number P09001 for the L3 protein Table 12A identifies the residues in the H. marismortui 50S ribosome subunit that together define by

10 at least one portion of a blasticidine binding pocket (wrapped 5.8 angstroms). In addition, Table 12A identifies the corresponding residues that define at least a portion of the blasticidine binding pocket in

E. coli, Rattus, human, and large subunit of human mitochondria. Table 12B identifies the residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of a blasticidine junction pocket (enveloped 5.8-12.6 angstroms). Unconserved waste has been identified as

15 described above with respect to Tables 5A and 5B.

Table 12A Residues that define the blasticidine binding pocket (5.8 A envelope) The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosomal subunit that are located within 5, 8 angstroms with respect to blasticidine atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S RNAs

C2106v

C2065 C3636 C3880 C1058

G2284h, v

G2251 G3917 G4156 G1145

G2285h, v

G2252 G3918 G4157 G1146

G2286h, v

G2253 G3919 G4158 G1147

C2287v

C2254 C3920 C4159 C1148

A2007v

A1966 A3538 A3782 A959

C2104v

C2063 C3634 C3878 C1056

C2105h, v

C2064 C3635 C3879 C1057

U2473v

U2438 U4105 U4344 U1255

A2474h, v

A2439 A4106 A4345 A1256

A2485v

A2450 A4117 A1267 A4356

A2486v

A2451 A4118 A1268 A4357

U2620v

U2585 U4252 U4491 U1402

G2627v

G2592 G4259 G4498 G1409

U2628v

U2593 U4260 U4499 G1410

C2629v

C2594 C4261 C4500 A1411

G2634v

G4266 G2599 G4505 U1416

A2635h, v

A2600 A4267 A4506 C1417

C2636h, v

C2601 C4268 C4507 C1418

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

A2637v

A2602 A4269 A4508 A1419

5 comparison of the sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of the aligned genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondriasd humanof). The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) by using default parameters.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 10 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

15 f) Genbank access number 13683

h) Indicates the residue that is within the hydrogen bonding distance (3.5 angstroms) of the ligand, for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs

20 (see: Computer Simulation of Liquids; Allen, M. P. and Tildesley, D. J .; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 12B Waste defining the blasticidine binding pocket (wrapped from 5.8 - 12.6 A)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

G885

A792 G1541 G1620 NP�

U1980

U1939 U3511 U3755 U932

A1981

U1940 G3512 G3756 C933

C1982

C1941 C3513 C3757 A934

C2006

C1965 C3537 C3781 U958

U2008

C1967 U3539 U3783 U960

G2102

G2061 G3632 G3876 G1054

A2103

A2062 A3633 A3877 A1055

U2107

C2066 U3637 U3881 U1059

G2110

G2069 U3640 U3884 G1062

G2111

A2070 G3641 G3885 NP

A2112

A2071 A3642 A3886 NP

G2113

C2072 G3643 G3887 NP

C2114

C2073 C3644 C3888 NP

U2282

U2249 U3915 U4154 U1143

G2283

G2250 G3916 G4155 G1144

G2288

G2255 G3921 G4160 G1149

G2289

G2256 G3922 G4161 A1150

G2471

G2436 G4103 G4342 G1253

C2472

G2437 U4104 U4343 U1254

C2475

C2440 C4107 C4346 C1257

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

U2484

U2449 U4116 U4355 U1266

C2487

C2452 C4119 C4358 C1269

G2529

G2494 G4161 G4400 A1311

C2530

G2495 A4162 A4401 C1312

U2531

C2496 U4163 U4402 G1313

A2532

A2497 C4164 C4403 A1314

C2536

C2501 C4168 C4407 C1318

U2619

U2584 U4251 U4490 U1401

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

G2623

G2588 G4255 G4494 G1405

A2624

A2589 A4256 A4495 A1406

C2625

A2590 C4257 C4496 C1407

G2630

G2595 G4262 G4501 G1412

A2633

A2598 A4265 A4504 A1415

G2638

G2603 G4270 G4509 G1420

G2639

U2604 G4271 G4510 G1421

U2640

U2605 U4272 U4511 U1422

G2642

G2607 A4274 A4513 G1424

G2643

G2608 G4514 G4275 G1425

L2 protein

G204

G235 G214 G214 W281

G205

E236 N215 N215 A282

R206

G237 H216 H216 G283

L10E protein

D109

XX T113 T113 XX

G110

XX G114 G114 XX

M111

XX M115 M115 XX

R112

XX R116 R116 XX

The residues that define the junction pocket (wrapped 5.8 A-12.6 A) were determined as those residues in the 50S ribosomal subunit that are located within 5.8-12.6 angstroms of blasticidine atoms, for which used the MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of aligned genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of H. marismortuig protein and the corresponding aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicus J, humanok or human mitochondria.

10 sequences were determined using the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified. XX means that no homologous protein was identified in that species.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

15 b) Genbank accession number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

20 g) Genbank RSHS2L access number for L2 protein; Genbank accession number 15825950 for the L10E protein.

i) Genbank accession number CAA26463 for the L2 protein; There is no Genbank accession number for the L10E protein.

j) Genbank R5RTL8 access number for the L2 protein; Genbank accession number NP 112362 for the L10E protein.

k) Genbank NP 000964 access number for the L2 protein; Genbank accession number NP 006004 for the L10E protein.

l) Genbank XP access number 004140 for the L2 protein; no Genbank accession number for L10E protein

30 Table 13A identifies residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of a carbomycin binding pocket (wrapped 5.8 angstroms). In addition, Table 13A identifies the corresponding residues that define at least a portion of the carbomycin binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. Table 13B identifies residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of a

35 carbomycin binding pocket (wrapped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 13A Wastes that define the carbomycin binding pocket (5.8 A wrapped)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C839v

U746 G1495 G1574 NP�

C2098v

G2057 A3628 A3872 A1050

G2099h, v

A2058 G3629 G3873 G1051

A2100

A2059 A2630 A3874 A1052

G2102h, v

G2061 G3632 G3876 G1054

A2103h, v

A2062 A3633 A3877 A1055

C2104

C2063 C3634 C3878 C1056

A2486v

A2451 A4118 A4357 A1268

C2487v

C2452 C4119 C4358 C1269

A2538h, v

A2503 A4170 A4409 A1320

U2539m, v

U2504 U4171 U4410 U1321

G2540h, v

G2505 G4172 G4411 G1322

U2541v

U2506 U4173 U4412 U1323

U2620v

U2585 U4252 U4491 U1402

C2644v

U2609 U4276 U4515 A1426

G2646v

C2611 U4278 U4517 U1428

L22 protein

M130v

K90 H133 H133 V187

The residues that define the binding pocket (5.8 A envelope) were determined as those residues in the 50S ribosomal ribosomal subunit that are located within 5.8 angstroms with respect to the 5 carbomycin A atoms for which it was used the MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of aligned genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or human mitochondria. In the case of ribosomal proteins, they were also carried out between the structural sequences of H. marismortui proteins and the corresponding sequence sequences of

10 aligned proteins encoding homologous proteins in E. colii, Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined by the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters. �NP means that no homologous residue was identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 15 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

20 f) Genbank access number 13683 g) Genbank access number R5HS22 for the L22 protein.

h) Indicates the residue that is within the hydrogen bonding distance (3.5 A) of the ligand, for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205. i) Genbank accession number CAA26465 for the L22 protein. j) Genbank accession number P24049 for the L22.5 protein k) Genbank accession number XP 057521 for the L22 protein. 1) Genbank Up access number 051279 for the L22 protein.

m) This residue is actually at 5.85 A. v) Van der Waals contact residues with the antibiotic are defined as those residues within the rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in Lennard-Jones peer potential

10 (see: Computer Simulation of Liquids; Allen, M. P. and Tildesley, D. J .; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 13B Waste that also defines the carbomycin A binding pocket

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C633

U576 G1284 G1365 NP

U835

U744 U1491 U1570 NP

G836

G745 G1492 G1571 NP

U837

NP A1493 A1572 NP

C838

NP C1494 C1573 NP

U840

U747 U1496 U1575 NP

A841

G748 G1497 G1576 NP

A843

A750 A1499 A1578 NP

A844

A751 A1500 A1579 NP

U845

A752 A1501 A1580 NP

A846

A753 U1502 U1581 NP

A882

A789 A1538 A1617 NP

U883

U790 U1539 U1618 NP

U1359

U1255 U2192 U2327 U351

G1837

U178 A3368 A3612 C831

U1838

U1782 U3369 U3613 C832

C2056

A2015 A3587 A3831 A1008

U2057

U2016 C3588 C3832 U1009

G2073

G2032 C3603 C3847 G1025

A2095

A2054 G3625 G3869 A1047

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 G1049

A2101

A2060 A3631 A3875 A1053

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

A2485

A2450 A4117 A4356 A1267

A2488

A2453 U4120 U4359 A1270

G2489

G2454 G4121 G4360 G1271

U2534

C2499 U4166 U4405 C1316

U2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

C2542

C2507 C4174 C4413 U1324

G2543

G2508 G4175 G4414 G1325

U2607

A2572 A4239 A4478 A1389

C2608

C2573 C4240 C4479 C1390

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

C2614

C2579 C4246 C4485 C1396

U2615

U2580 U4247 U4486 U1397

G2616

G2581 G4248 G4487 G1398

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

A2637

A2602 A4269 A4508 A1419

G2643

G2608 G4275 G4514 G1425

U2645

C2610 U4277 U4516 U1427

C2647

C2612 U4279 U4518 U1429

L3

W242

Q150 W257 W257 G246

L4

G62

G54 G68 G68 E73

S63

S55 T69 T69 Q74

G64

G56 G70 G70 E75

R65

K57 R71 R71 R76

G66

NP A72 A72 G78

Q67

NP V73 V73 L79

A129

A89 A132 A132 A186

G131

G91 G134 G134 A188

L4

R132

R92 R135 R135 H189

The residues that define the junction pocket (envelope of 5.8 A-12.6 A) were determined as those residues in the 50S ribosome subunit that are located within 5.8-12.6 angstroms of carbomycin A atoms. , for which MTDASa was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding sequences of aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicusi, humanok, or human mitochondria. The sequence alignments were determined by the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein; Genbank accession number P12735 for the L4 protein.

i) Genbank accession number CAA26460 for the L3 protein; Genbank accession number CAA26461 for the L4 protein.

j) Genbank accession number P21531 for the L3 protein; Genbank accession number JC4277 for the L4 protein.

k) Genbank NP 000958 access number for the L3 protein; Genbanks accession number P36578, NP 000959, S39803, and T09551 for the L4 protein.

l) Genbank accession number P09001 for the L3 protein; Genbank XP access number 049502 for L4 protein

Table 14A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of a tylosin binding pocket (wrapped 5.8 angstroms). In addition, Table 13A identifies the corresponding residues that define at least a portion of the tylosin binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. Table 14B identifies the residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of a tylosin binding pocket (enveloped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 14A Waste defining the tylosin binding pocket (wrapped 5.8 A)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C839v

U746 G1495 G1574 NP�

A841h, v

G748 G1497 G1576 NP

A843v

A750 A1499 A1578 NP

A844h, v

A751 A1500 A1579 NP

U845v

A752 A1501 A1580 NP

A846v

A753 U1502 U1581 NP

G1837

U1781 A3368 A3612 C831

C2098v

G2057 A3628 A3872 A1050

G2099h, and

A2058 G3629 G3873 G1051

A2100v

A2059 A3630 A3874 A1052

G2102v

G2061 G3632 G3876 G1054

A2103h, v

A2062 A3633 A3877 A1055

A2486v

A2451 A4118 A4357 A1268

A2538h, v

A2503 A4170 A4409 A132

U2539Y

U2504 U4171 U4410 U1321

G2540v

G2505 G4172 G4411 G1322

U2541

U2506 U4173 U4412 U1323

U2620m, v

U2585 U4252 U4491 U1402

C2644v

U2609 U4276 U4515 U1426

U2645v

C2610 U4277 U4516 U1427

G2646v

C2611 U4278 U4517 U1428

L22 protein

M130v

K90 H133 H133 V187

The residues that define the union pocket (5.8 A wrapped) were determined as those residues in the

5 50S ribosomal subunit that are located within 5.8 angstroms with respect to tylosin atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding genomic DNA sequences encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the sequence

10 of the marismortui g protein and the corresponding sequences of aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 15 (1): 13-27.36-37.

b) Genbank access number AF034619 c) Genbank accession number J01695 d) Genbank access number 2624399 e) Genbank access number M11167

5 f) Genbank access number 13683 g) Genbank access number R5HS22 for the L22 protein. h) Indicates the residue that is within the hydrogen bonding distance (3.5 A) of the ligand, for example, see

Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205. i) Genbank accession number CAA26465 for the L22 protein.

10 j) Genbank accession number P24049 for the L22 protein. k) Genbank XP access number 057521 for the L22 protein. l) Genbank XP accession number 051279 for the L22 protein. m) This residue is actually at 5.87 A. v) The residues in contact of van der Waals with the antibiotic are defined as those residues within the rmin of

15 any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the torque potential of Lennard-Jones (see: Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

20 Table 14B. Waste defining the tylosin binding pocket (wrapped 5.8 A-12.6 A)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C633

U576 G1284 G1365 NP�

U835

U744 U1491 U1570 NP

G836

G745 G1492 G1571 NP

U837

NP A1493 A1572 NP

C838

NP C1494 C1573 NP

U840

U747 U1496 U1575 NP

C842

A749 C1498 C1577 NP

C847

U754 C1503 C1582 NP

A882

A789 A1538 A1617 NP

U883

U790 U1539 U1618 NP

U1359

U1255 U2191 U2327 U351

G1688

G1613 G2634 G2770 G679

A1689

A1614 C2635 C2771 A680

C1690

C1615 A2636 A2772 U681

C1692

C1617 C2638 C2774 A683

G1694

G1619 G2640 G2776 A685

A1836

A1780 A3367 A3611 A830

C2056

A2015 A3587 A3831 A1008

U2057

U2016 C3588 C3832 U1009

G2073

G2032 C3603 C3847 G1025

A2095

A2054 G3625 G3869 A1047

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 G1049

A2101

A2060 A3631 A3875 A1053

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

C2476

U2441 C4108 C4347 C1258

U2478

C2443 C4110 C4349 U1260

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

A485

A2450 A4117 A4356 A1267

C2487

C2452 C4119 C4358 C1269

A2488

A2453 U4120 U4359 A1270

U2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

C2542

C2507 C4174 C4413 U1324

U2607

A2572 A4239 A4478 A1389

C2608

C2573 C4240 C4479 C1390

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

C2614

C2579 C4246 C4485 C1396

U2615

U2580 U4247 U4486 U1397

G2615

G2581 G4248 G4487 G1398

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U2451 U4490 U1401

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

G2643

G2608 G4275 G4514 G1425

C2647

C2612 U4279 U4518 C1429

U2648

U2613 A4280 A4519 U1430

L3 protein

W242

Q150 W257 W257 G246

L4 protein

G62

G54 G68 G68 E73

S63

S55 T69 T69 Q74

G64

G56 G70 G70 E75

R65

K57 R71 R71 R76

 G66

NP A72 A72 G78

Q67

NP V73 V73 L79

L22 protein

R128

R88 R131 R131 K184

A129

A89 A132 A132 A186

G131

G91 G134 G134 A188

R132

R92 R135 R135 H189

L37E protein

T1

XX M1 M1 XX

L37E protein

G2

XX T2 T2 XX

A3

XX K3 K3 XX

G4

XX  G4  G4 XX

T5

XX  T5 T5 XX

P6

XX S6 S6 XX

S7

XX  S7  S7 XX

Q8

XX F8 F8 XX

The residues that define the junction pocket (wrapped 5.8 A-12.6 A) were determined as those residues in the 50S ribosomal subunit that are located within 5.8-12.6 angstroms with respect to the tylosin atoms , for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. collc, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequences of the H. marismortui protein and the corresponding sequences of aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicus J, humanok or human mitochondria. The 10 sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters. NP means that no homologous residues were identified. XX means that no homologous protein was identified in these species.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

5 d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein; Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS24 for the L22 protein; Genbank accession number P32410

10 for the L37E protein.

i) Genbank accession number CAA26460 for the L3 protein; Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein; There is no Genbank accession number for the L37E protein.

j) Genbank accession number P21531 for the L3 protein; Genbank accession number JC4277 for the protein

15 L4; Genbank accession number P24049 for the L22 protein; Genbank accession number CAA47102 for the L37E protein.

k) Genbank NP 000958 access number for the L3 protein; Genbanks accession number P36578, NP 000959, S39803, and T09551 for the L4 protein; Genbank XP accession number 051279 for the L22 protein; Genbank accession number NP 000988 for the L37E protein.

20 l) Genbank accession number P09001 for the L3 protein; Genbank XP accession number 049502 for the L4 protein; Genbank XP accession number 051279 for the L22 protein; There is no Genbank accession number for the L37E protein.

Table 15A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of a sparsomycin binding pocket (wrapped 5.8 angstroms). Table 13A identifies the

25 corresponding residues that define at least a portion of the sparsomycin binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. In addition, Table 15B identifies residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of a sparsomycin binding pocket (envelop 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

30 Table 15A. Residues that define the sparsomycin binding pocket (5.8 A envelope) The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosomal unit that are located within 5, 8 angstroms of the sparsomycin atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparison

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A2486h, v

A2451 A4118 A4357 A1268

C2487v

C2452 C4119 C4358 C1269

A2488m, v

A2453 U4120 U4359 A1270

G2540m, v

G2505 G4172 G4411 G1322

U2541v

U2506 U4173 U4412 U1323

C2542

C2507 C4174 C4413 U1324

C2608v

C2573 C4240 C4479 C1390

U2619h, v

U2584 U4251 U4490 U1401

U2620h, v

U2585 U4252 U4491 U1402

C2636

C2601 C4268 C4507 C1418

A2637h, v

A2602 A4269 A4508 A1419

5 of sequences of the structure of H. marismortui 23S rRNA with the corresponding genomic DNA sequences encoding the 23S homologous rRNA (E. colic, Rattus norvegicusd, humanoe, human mitochondriaf. The sequence alignments were determined with the MegAlign program ( DNASTAR, Madison, Wisconsin, USA), for which default parameters were used.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 10 (1): 13-27,36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

15 f) Genbank access number 13683

h) Indicates the residue that is within the hydrogen bonding distance (3.5 A) of the ligand, for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

m) The residue A2488 is in fact 6.0 A; the residue G2540 is really 5.85 A.

v) The residues in contact of van der Waals with the antibiotic are defined as those residues within the rmin of

20 any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the torque potential of Lennard-Jones (see: Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

25 Table 15B. Residues that define the sparsomycin binding pocket (5.8 A-12.6 A envelope) The residues that define the union pocket (5.8 A-12.6 A envelope) were determined as those residues in the 50S ribosomal unit that are located within 5.8-12.6 angstroms with respect to the sparsomycin atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the marismortuig protein and the corresponding sequences of the sequences of

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

G2073

G2032 C3603 C3847 G1025

G2102

G2061 G3632 G3876 G1054

A2103

A2062 A3633 A3877 A1055

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

C2106

C2065 C3636 C3880 C1058

G2284

G2251 G3917 G4156 G1145

G2285

G2252 G3918 G4157 G1146

G2286

G2253 G3919 G4158 G1147

U2473

U2438 U4105 U4344 U1255

A2474

A2439 A4106 A4345 A1256

G2482

G2447 G4114 G4353 G1264

A2485

A2450 A4117 A4356 A1267

G2498

C2463 C4130 C4369 U1280

U2527

U2492 U4159 U4398 U1309

U2528

U2493 U4160 U4399 U1310

G2529

G2494 G4161 G4400 A1311

C2530

G2495 A4162 A4401 C1312

A2532

A2497 C4164 C4403 A1314

C2534

C2499 U4166 U4405 C1316

U2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

A2538

A2503 A4170 A4409 A1320

U2539

U2504 U4171 U4410 U1321

G2543

G2508 G4175 G4414 G1325

G2588

G2553 G4220 G4459 G1370

U2589

U2554 U4221 U4460 U1371

U2590

U2555 U4222 U4461 U1372

U2607

A2572 A4239 A4478 A1389

G2609

G2574 G4241 G4480 G1391

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

G2617

G2582 G4249 G4488 A1399

G2618

G2583 G4250 G4489 G1400

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

G2627

G2592 G4259 G4498 G1409

A2635

A2600 A4267 A4506 C1417

G2638

G2603 G4270 G4509 G1420

G2639

U2604 G4271 G4510 G1421

U2640

U2605 U4272 U4511 U1422

23S ARNr

 G2643

G2608 G4275 G4514 G1425

L3 protein

W242

Q150 W257 W257 G246

5 aligned proteins encoding homologous proteins in E. colii, Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined by the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

10 b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

15 g) Genbank accession number AAA86859 for the L3 protein.

i) Genbank accession number CAA26460 for the L3 protein.

j) Genbank accession number P21531 for the L3 protein.

k) Genbank NP 000958 access number for the L3 protein.

l) Genbank accession number P09001 for the L3 protein.

20 Table 16A identifies residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of a virginiamycin binding pocket (wrapped 5.8 angstroms). In addition, Table 16A identifies the corresponding residues that define at least one portion of the virginiamycin M binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. In addition, Table 16B identifies residues in the H. marismortui 50S ribosomal subunit that together define a broader portion of

25 a virginiamycin binding pocket (wrapped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 16A. Residues that define the virginiamycin M binding pocket (5.8 A envelope) The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosomal unit that are located within 5 , 8 angstroms of the virginiamycin M atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A2100v

A2059 A3630 A3874 A1052

G2102h, v

G2061 G3632 G3876 G1054

A2103h, v

A2062 A3633 A3877 A1055

C2104h, v

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

G2482v

G2447 G4114 G4353 G1264

A2486v

A2451 A4118 A4357 A1268

C2487v

C2452 C4119 C4358 C1269

U2535v

U2500 U4167 U4406 U1317

A2538h, v

A2503 A4170 A4409 A1320

U2539v

U2504 U4171 U4410 U1321

G2540v

G2505 G4172 G4411 G1322

U2541v

U2506 U4173 U4412 U1323

U2620v

U4252 U2585 U4491 U1402

5 sequence comparison of the structure of H. marismortui 23S rRNA with the corresponding genomic DNA sequences encoding the 23S homologous rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 10 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank access number J01695 d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

15 h) Indicates the residue that is within the hydrogen bonding distance (3.5 A) of the ligand, for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones torque (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.).

20 Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen, WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 16B. Waste defining the virginiamycin M binding pocket (wrapped 5.8 A-12.6 A)

H. Marismortui residue

Corresp waste. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A631

A574 G1282 G1363 NP�

A632

A575 C1283 C1364 NP

C633

U576 G1284 G1365 NP

U1838

U1782 U3369 U3613 C832

G2073

G2032 C3603 C3847 G1025

A2095

A2054 G3625 G3869 A1047

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 G1049

C2098

G2057 A3628 A3872 A1050

G2099

A2058 G3629 G3873 G1051

A2101

A2060 A3631 A3875 A1053

C2106

C2065 C3636 C3880 C1058

U2107

C2066 U3637 U3881 U1059

G2284

G2251 G3917 G4156 G1145

U2473

U2438 U4105 U4344 U1255

C2475

C2440 C4107 C4346 C1257

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

A2479

G2444 A4111 A4350 A1261

G2480

G2445 G4112 G4351 G1262

G2481

G2446 G4113 G4352 G1263

U2484

U2449 U4116 U4355 U1266

A2485

A2450 A4117 A4356 A1267

A2488

A2453 U4120 U4359 A1270

G2489

G2454 G4121 G4360 G1271

A2532

A2497 C4164 C4403 A1314

C2534

C2499 U4166 U4405 C1316

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

C2542

C2507 C4174 C4413 U1324

U2607

A2572 A4239 A4478 A1389

C2608

C2573 C4240 C4479 C1390

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

C2636

C2601 C4268 C4507 C1418

A2637

A2602 A4269 A4508 A1419

G2643

G2608 G4275 G4514 G1425

C2644

U2609 U4276 U4515 U1426

U2645

C2610 U4277 U4516 U1427

G2646

C2611 U4278 U4517 U1428

L3

W242

Q150 W257 W257 G246

The residues that define the junction pocket (envelope of 5.8 A-12.6 A) were determined as those residues in the 50S ribosomal unit that are located within 5.8-12.6 angstroms of the virginiamycin M atoms , for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding sequences of the aligned proteins encoding homologous proteins in E. coli, Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein.

i) Genbank accession number CAA26460 for the L3 protein.

j) Genbank accession number P21531 for the L3 protein.

k) Genbank NP 000958 access number for the L3 protein.

l) Genbank accession number P09001 for the L3 protein.

Table 17A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of a spiramycin binding pocket (wrapped 5.8 angstroms). In addition, Table 17A identifies the corresponding residues that define at least a portion of the spiramycin binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. On the other hand, Table 17B identifies the residues in the H. marismortui 50S ribosomal subunit that together define a wider portion of a spiramycin binding pocket (enveloped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 17A. Residues that define the spiramycin binding pocket (5.8 A envelope) The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosome unit that are located within 5, 8 angstroms of the spiramycin atoms, for which the MIDASa program was used. The conserved residues were determined by comparison between

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C839v

U746 G1495 G1574 NP�

C2098

G2057 A3628 A3872 A1050

G2099h, v

A2058 G3629 G3873 G1051

A2100v

A2059 A3630 A3874 A1052

A2103h, v

A2062 A3633 A3877 A1055

A2538v

A2503 A4170 A4409 A1320

U2539v

U2504 U4171 U4410 U1321

G2540h, v

G2505 G4172 G4411 G1322

U2620m, v

U2585 U4252 U4491 U1402

C2644

U2609 U4276 U4515 U1426

G2646v

C2611 U4278 U4517 U1428

L4

S63v

S55 T69 T69 Q74

G64v

G56 G70 G70 E75

R65v

K57 R71 R71 R76

L22

M130m, v

K90 H133 H133 V187

5 proposed secondary structures H. marismortuib, E. colib, and Rattus norvegicusc, humanod, and humanase mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA), for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 10 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

15 f) Genbank access number 13683

g) Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS24 for the L22 protein.

h) Indicates a residue that is within the hydrogen bonding distance (3.5 angstroms) of the ligand; for example see Jorgensen, WL, Nguyen, TB. J. Comput. Chem. (1993) 14: 195-205.

20 i) Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein.

j) Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein

k) Genbank accession numbers P36578, NP 000959, S39803, and T09551 for the L4 protein; Genbank XP access number 25 051279 for the L22 protein.

l) Genbank XP access number 049502 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

m) The residue U2620 is actually at 5.83 A; M130 is actually at 5.87 A.

v) residues in contact with van der Waals with the antibiotic are defined as those residues within the rmin of

30 any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see: Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

35 Table 17B. Waste defining the spiramycin binding pocket (wrapped 5.8 A-12.6 Al)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A841

G748 G1497 G1576 NP�

A843

A750 A1499 A1578 NP

A844

A751 A1500 A1579 NP

U845

A752 A1501 A1580 NP

A846

A753 U1502 U1581 NP

A841

G748 G1497 G1576 NP

A766

A675 A1424 A1502 NP�

A767

A676 A1425 A1503 NP

U835

U744 U1491 U1570 NP

G836

G745 G1492 G1571 NP

U837

NP A1493 A1572 NP

C838

NP C1494 C1573 NP

U840

U747 U1496 U1575 NP

A882

A789 A1538 A1617 NP

U883

U790 U1539 U1618 NP

C884

C791 C1540 C1619 NP

U1359

U1255 U2192 U2327 U351

G1837

U1781 A3368 A3612 C831

U1838

U1782 U3369 U3612 C832

C2056

A2015 A3587 A3831 A1008

U2057

U2016 C3588 C3832 U1009

G2073

G2032 C3603 C3847 G1025

A2095

A2054 G3625 G3869 A1047

G2097

G2056 A3627 A3871 G1049

A2101

A2060 A3631 A3875 A1053

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3625 C3879 C1057

A2474

A2439 A4106 A4345 A1256

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

2478

C2443 C4110 C4349 U1260

A2479

G2444 A4111 A4350 A1261

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

A2485

A2450 A4117 A4356 A1267

A2486

A2451 A4118 A4357 A1268

C2487

C2452 C4119 C4358 C1269

A2488

A2453 U4120 U4359 A1270

U2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

C2542

C2507 C4174 C4413 U1324

U2607

A2572 A4239 A4478 A1389

C2608

C2573 C4240 C4479 C1390

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

C2614

C2579 C4246 C4485 C1396

U2615

U2580 U4247 U4486 U1397

G2616

G2581 G4248 G4487 G1398

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2621

U2586 U4253 U4492 C1403

G2643

G2608 G4275 G4514 G1425

U2645

C2610 U4277 U4516 U1427

C2647

C2612 U4279 U4518 C1429

U2648

U2613 A4280 A4519 U1430

L3 protein

W242

Q150 W257 W257 G246

L4 protein

S60

V52 S66 S66 G71

F61

T53 W67 W67 F72

G62

G54 G68 G68 E73

G66

NP A72 A72 G78

Q67

NP V73 V73 L79

A68

NP A74 A74 A80

H69

K58 R75 R75 D81

L22 protein

A129

A89 A132 A132 A186

G131

G91 G134 G134 A188

L22 protein

R132

R92 R135 R135 H189

The residues that define the junction pocket (envelope of 5.8 A-12.6 A) were determined as those residues in the 50S ribosomal unit that are located within 5.8-12.6 angstroms of the spiriramycin atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui g protein and the corresponding sequences of the aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicusjj humanok or humanoid mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein; Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS24 for the L22 protein.

h) Indicates a residue that is within the hydrogen bonding distance (3.5 angstroms) of the ligand; for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

i) Genbank accession number CAA26460 for the L3 protein; Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein.

j) Genbank accession number P21531 for the L3 protein; Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein.

k) Genbank NP 000958 access number for the L3 protein; Genbanks accession number P36578, NP 000959, S39803, and T09551 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

l) Genbank accession number P09001 for the L3 protein; Genbank XP accession number 049502 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 18A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of an erythromycin binding pocket (wrapped 5.8 angstroms). In addition, Table 18A identifies the corresponding residues that define at least a portion of the erythromycin binding pocket in

E. coli, Rattus, human, and large subunit of human mitochondria. On the other hand, Table 18B identifies residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of an erythromycin binding pocket (enveloped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 18A Waste that defines the erythromycin binding pocket (5.8 A shell)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

C839v

U746 G1495 G1574 NP�

A841v

G748 G1497 G1576 NP

C2098m, v

G2057 A3628 A3872 A1050

G2099h, v

A2058 G3629 G3873 G1051

A2100v

A2059 A3630 A3874 A1052

A2103v

A2062 A3633 A3877 A1055

A2538v

A2503 A4170 A4409 A1320

G2540v

G2505 G4172 G4411 G1322

C2644v

U2609 U4276 U4515 U1426

U2645v

C2610 U4277 U4516 U1427

G2646v

C2611 U4278 U4517 U1428

L4

G64v

G56 G70 G70 E75

L22

M130m, v

K90 H133 H133 V187

The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosomal unit that are located within 5.8 angstroms of the erythromycin atoms, for which the MIDASa program was used . The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding sequences of aligned proteins encoding homologous proteins in E. colii,

10 Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

15 b) Genbank accession number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

20 g) Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS24 for the L22 protein.

h) Indicates a residue that is within the hydrogen bonding distance (3.5 A) of the ligand; for example, see: Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

i) Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein.

j) Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein k) Genbank accession number P36578, NP 000959, S39803, and T09551 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

l) Genbank XP access number 049502 for the L4 protein; Genbank XP accession number 051279 for the 5 L22 protein.

m) C2098 residue is actually at 5.99 A; M130 is at 5.90 A.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.).

10 Lennard-Jones ss are taken from the OPLS-AA force camp (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen, WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 18B Waste that also defines the erythromycin binding pocket

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C633

U576 G1284 G1365 NP�

U835

U744 U1491 U1570 NP

G836

G745 G1492 G1571 NP

U837

NP A1493 A1572 NP

C838

NP C1494 C1573 NP

U840

U747 U1496 U1575 NP

A843

A750 A1499 A1578 NP

A844

A751 A1500 A1579 NP

U845

A752 A1501 A1580 NP

A846

A753 U1502 U1581 NP

C847

U754P C1503 C1582 NP

U1359

U1255 U2191 U2327 U351

G1837

U1781 A3368 A3612 C831

U1838

U1782 U3369 U3613 C832

C2056

A2015 A3587 A3831 A1008

U2057

U2016 C3588 C3832 U1009

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 G1049

A2101

A2060 A3631 A3875 A1053

G2102

G2061 G3632 G3876 G1054

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

G2482

G2447 G4114 G4353 G1264

A2486

A2451 A4118 A4357 A1268

C2487

C2452 C4119 C4358 C1269

2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

U2539

U2504 U4171 U4410 U1321

U2541

U2506 U4173 U4412 U1323

C2542

C2507 C4174 C4413 U1324

G2543

G2508 04175 G4414 G1325

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

G2613

G2578 G4245 G4484 U1395

C2614

C2579 C4246 C4485 C1396

U2615

U2580 U4247 U4486 U1397

G2616

G2581 G4248 G4487 G1398

G2617

G2582 G4249 G4488 A1399

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2620

U2585 U4252 U4491 U1402

U2621

U2586 U4253 U4492 C1403

A2622

A2587 A4254 A4493 A1404

G2643

G2608 G4275 G4514 G1425

C2647

C2612 U4279 U4518 C1429

U2648

U2613 A4280 A4519 U1430

L3 protein

W242

Q150 W257 W257 G246

L4 protein

G62

G54 G68 G68 E73

S63

S55 T69 T69 Q74

R65

K57 R71 R71 R76

Q67

NP� A72 A72 G78

L22 protein

A129

A89 A132 A132 A186

G131

G91 G134 G134 A188

L22 protein

R132

R92 R135 R135 H189

The residues that define the junction pocket (5.8 A-12.6 A) were determined as those residues in the 50S ribosomal unit that are located within 5.8-12.6 angstroms of the erythromycin atoms, for which was used the MIDASa program. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. coilc, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding sequences of aligned proteins encoding homologous proteins in E. colii, Rattus norvegicusj, humanok or mitochondria human. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein; Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS24 for the L22 protein.

h) Indicates a residue that is within hydrogen bonding distance (3.5 A) of the ligand; for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

i) Genbank accession number CAA26460 for the L3 protein; Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein.

j) Genbank accession number P21531 for the L3 protein; Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein.

k) Genbank NP 000958 access number for the L3 protein; Genbanks accession number P36578, NP 000959, S39803, and T09551 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

l) Genbank accession number P09001 for the L3 protein; Genbank XP accession number 049502 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 19A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of an azithromycin binding pocket (wrapped 5.8 angstroms). In addition, Table 19A identifies the corresponding residues that define at least one portion of the azithromycin binding pocket in E. coli, Rattus, human, and large subunit of human mitochondria. On the other hand, Table 19B identifies residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of an azithromycin binding pocket (enveloped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 19A. Waste that defines the azithromycin binding pocket (5.8 A shell)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C839v

U746 G1495 G1574 NP�

G2099h, v

A2058 G3629 G3873 G1051

A2100v

A2059 A3630 A3874 A1052

A2103v

A2062 A3633 A3877 A1055

A2538v

A2503 A4170 A4409 A1320

G2540v

G2505 G4172 G4411 G1322

U2541m, v

U2506 U4173 U4412 U1323

C2644v

U2609 U4276 U4515 U1426

U2645v

C2610 U4277 U4516 U1427

C2646v

C2611 U4278 U4517 U1428

L22 protein

M129m, x

K90 H133 H133 V187

The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosome unit that are located within 5.8 angstroms with respect to azithromycin atoms for which the program was used MIDASa. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the case of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding sequences of the aligned proteins encoding proteins

10 homologs in E. colii, Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) for which default parameters were used. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

15 b) Genbank accession number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

20 g) Genbank access number R5HS24 for the L22 protein.

h) Indicates a residue that is within the distance of hydrogen bonding (3.5 A) of the ligand; for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

i) Genbank accession number CAA26465 for the L22 protein.

j) Genbank accession number P24049 for the L22 protein k) Genbank accession number XP 051279 for the L22 protein.

l) Genbank XP accession number 051279 for the L22 protein.

m) The residue U2541 is actually at 5.93 A; M130 is at 5.95 A.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 19B Waste that also defines the azithromycin binding pocket

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

C633

U576 G1284 G1365 NP�

U835

U744 U1491 U1570 NP

G836

G745 G1492 G1571 NP

U837

NP A1493 A1572 NP

C838

NP C1494 C1573 NP

U840

U747 U1496 U1575 NP

A841

G748 G1497 G1576 NP

A843

A750 A1499 A1578 NP

A844

A751 A1500 A1579 NP

U845

A752 A1501 A1580 NP

A846

A753 U1502 U1581 NP

C847

U754 C1503 C1582 NP

U1359

U1255 U2191 U2327 U351

G1837

U1781 A3368 A3612 C831

U1838

U1782 U3369 U3613 C832

C2056

A2015 A3587 A3831 A1008

U2057

U2016 C3588 C3832 U1009

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 G1049

C2098

G2057 A3628 A3872 A1050

A2101

A2060 A3631 A3875 A1053

G2102

G2061 G3632 G3876 G1054

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

C2476

U2441 C4108 C4347 C1258

C2477

C2442 A4109 A4348 C1259

U2478

C2443 C4110 C4349 U1260

G2481

G2446 G4113 G4352 G1263

G2482

G2447 G4114 G4353 G1264

A2486

A2451 A4118 A4357 A1268

C2487

C2452 C4119 C4358 C1269

U2535

U2500 U4167 U4406 U1317

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

U2539

U2504 U4171 U4410 U1321

C2542

C2507 C4174 C4413 U1324

G2543

G2508 G4175 G4414 G1325

U2607

A2572 A4239 A4478 A1389

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

G2613

G2578 G4245 G4484 U1395

C2614

C2579 C4246 C4485 C1396

G2616

G2581 G4248 G4487 G1398

G2617

G2582 G4249 G4488 A1399

G2618

G2583 G4250 G4489 G1400

U2619

U2584 U4251 U4490 U1401

U2620

U2585 U4252 U4491 U1402

U2621

U2586 U4253 U4492 C1403

U2615

U2580 U4247 U4486 U1397

A2622

A2587 A4254 A4493 A1404

G2643

G2608 G4275 G4514 G1425

C2647

C2612 U4279 U4518 C1429

U2648

U2613 A4280 A4519 U1430

L3 protein

W232

S139 NP NP S236

W242

Q150 W257 W257 G246

L4 protein

G62

G54 G68 G68 E73

S63

S55 T69 T69 Q74

G64

G56 G70 G70 E75

R65

K57 R71 R71 R76

G66

NP A72 A72 G78

Q67

NP� V73 V73 L79

L22 protein

A129

A89 A132 A132 A186

G131

G91 G134 G134 A188

L22 protein

R132

R92 R135 R135 H189

The residues that define the junction pocket (5.8 A-12.6 A envelope) were determined as those residues in the 50S ribosomal unit that are located within 5.8-12.6 angstroms of azithromycin atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding aligned protein sequences encoding homologous proteins in E. colii, Rattus norvegicusj, humanok, or human mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein; Genbank accession number P12735 for the L4 protein; Genbank accession number R5HS24 for the L22 protein.

h) Indicates a residue that is within hydrogen bonding distance (3.5 A) of the ligand; for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

i) Genbank accession number CAA26460 for the L3 protein; Genbank accession number CAA26461 for the L4 protein; Genbank accession number CAA26465 for the L22 protein.

j) Genbank accession number P21531 for the L3 protein; Genbank accession number JC4277 for the L4 protein; Genbank accession number P24049 for the L22 protein.

k) Genbank NP 000958 access number for the L3 protein; Genbanks accession number P36578, NP 000959, S39803, and T09551 for the L4 protein; Genbank XP accession number 051279 for the L22 protein.

l) The residues in contact of van der Waals with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs (see : Computer Simulation of Liquids; Allen, MP and Tildesley, DJ; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 20A identifies the residues in the H. marismortui 50S ribosomal subunit that together define at least a portion of a linezolide junction pocket (wrapped 5.8 angstroms). In addition, Table 20A identifies the corresponding residues that define at least a portion of the linezolide junction pocket in E. coli, Rattus, human, and large subunit of human mitochondria. On the other hand, Table 20B identifies the residues in the H. marismortui 50S ribosomal subunit that together define a larger portion of a linezolide junction pocket (enveloped 5.8-12.6 angstroms). Unconserved residues have been identified as previously described with respect to Tables 5A and 5B.

Table 20A Residues that define the linezolide joint pocket (5.8 A envelope) The residues that define the union pocket (5.8 A envelope) were determined as those residues in the 50S ribosome unit that are located within 5, 8 angstroms with respect to linezolide atoms, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A2100v

A2059 A3630 A3874 A1052

G2102v

G2061 G3632 G3876 G1054

A2103v

A2062 A3633 A3877 A1055

G2482v

G2447 G4114 G4353 G1264

A2486v

A2451 A4118 A4357 A1268

C2487h, v

C2452 C4119 C4358 C1269

A2488v

A2453 U4120 U4359 A1270

U2535v

U2500 U4167 U4406 U1317

A2538v

A2503 A4170 A4409 A1320

U2539h, v

U2504 U4171 U4410 U1321

G2540h, v

G2505 G4172 G4411 G1322

U2541h, v

U2506 U4173 U4412 U1323

C2542v

C2507 C4174 C4413 U1324

U2619v

U2584 U4251 U4490 U1401

U2620v

U2585 U4252 U4491 U1402

5 sequence comparison of the structure of H. marismortui 23S rRNA with the corresponding genomic DNA sequences encoding the 23S homologous rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 10 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

15 f) Genbank access number 13683

h) Indicates a residue that is within the hydrogen bonding distance (3.5 angstroms) of the ligand; for example, see Jorgensen, WL, Nguyen, TB. (1993) J. Comput. Chem. 14: 195-205.

v) Van der Waals contact residues with the antibiotic are defined as those residues within rmin of any atom in the antibiotic, where rmin = 21 / 6s and is the minimum in the potential of Lennard-Jones pairs

20 (see: Computer Simulation of Liquids; Allen, M. P. and Tildesley, D. J .; Oxford University Press: New York, 1992, 11.). Lennard-Jones ss are taken from the OPLS-AA force field (see, for example Jorgensen, WL; Maxwell, DS and Tirado-Rives, JJ Am. Chem. Soc., 1996, 118, 11225-11236 and Jorgensen , WL BOSS, Version 4.3; Yale University: New Haven, CT, 2002).

Table 20B Waste that also defines the linezolide joint pocket (wrapped 5.8 A- 12.6 A)

H. Marismortui residue

Residue correresp. in E. coli Corresponding residue in Rattus Corresponding residue in human Corresponding residue in human mitochondria

23S ARNr

A631

A574 G1282 G1363 NP�

A632

A575 C1283 C1364 NP

C633

U576 G1284 G1365 NP

G2073

G2032 C3603 C3847 G1025

G2094

G2053 G3624 G3868 G1046

A2095

A2054 G3625 G3869 A1047

A2096

C2055 A3626 A3870 C1048

G2097

G2056 A3627 A3871 G1049

C2098

G2057 A3628 A3872 A1050

G2099

A2058 G3629 G3873 G1051

A2101

A2060 A3631 A3875 A1053

C2104

C2063 C3634 C3878 C1056

C2105

C2064 C3635 C3879 C1057

A2474

A2439 A4106 A4345 A1256

G2480

G2445 G4112 G4351 G1262

G2481

G2446 G4113 G4352 G1263

A2485

A2450 A4117 A4356 A1267

G2489

G2454 G4121 G4360 G1271

U2528

U2493 U4160 U4399 U1310

G2529

G2494 G4161 G4400 A1311

C2534

C2499 U4166 U4405 C1316

C2536

C2501 C4168 C4407 C1318

G2537

G2502 G4169 G4408 G1319

G2543

G2508 G4175 G4414 G1325

G2588

G2553 G4220 G4459 G1370

U2607

A2572 A4239 A4478 A1389

C2608

C2573 C4240 C4479 C1390

G2609

G2574 G4241 G4480 G1391

U2610

C2575 U4242 U4481 U1392

G2611

G2576 G4243 G4482 G1393

A2612

A2577 A4244 A4483 A1394

G2616

G2581 G4248 G4487 G1398

G2617

G2582 G4249 G4488 A1399

G2618

G2583 G4250 G4489 G1400

U2621

U2586 U4253 U4492 C1403

A2637

A2602 A4269 A4508 A1419

U2645

C2610 U4277 U4516 U1427

G2646

C2611 U4278 U4517 U1428

L3 protein

W242

Q150 W257 W257 G246

The residues that define the junction pocket (envelope of 5.8 A-12.6 A) were determined as those residues in the 50S ribosome unit that are located within 5.8-12.6 angstroms with respect to the atoms of linezolido, for which the MIDASa program was used. The residues conserved in 23S rRNA were determined by comparing sequences of the structure of H. marismortui 23S rRNA with the corresponding sequences of genomic DNA encoding the homologous 23S rRNA (E. colic, Rattus norvegicusd, humanoe or mitochondria human). In the cases of ribosomal proteins, comparisons were also made between the structural sequence of the H. marismortui protein and the corresponding sequences of aligned protein sequences encoding homologous proteins in E. coliii, Rattus norvegicusj, humanok or human mitochondria. The sequence alignments were determined with the MegAlign program (DNASTAR, Madison, Wisconsin, USA) with default parameters. �NP means that no homologous residues were identified.

a) T. E. Ferrin, C.C. Huang, L. E. Jarvis, and R. Langridge (1988) "The MIDAS Display System" J. Mol. Graphics, 6 (1): 13-27, 36-37.

b) Genbank access number AF034619

c) Genbank accession number J01695

d) Genbank access number 2624399

e) Genbank access number M11167

f) Genbank access number 13683

g) Genbank accession number AAA86859 for the L3 protein.

i) Genbank accession number CAA26460 for the L3 protein.

j) Genbank accession number P21531 for the L3 protein.

k) Genbank NP 000958 access number for the L3 protein.

l) Genbank accession number P09001 for the L3 protein.

The person skilled in the art, when in possession of the aforementioned target site or others, can use the process of rational drug design to identify molecules that potentially bind to one or more of the target sites and / or inhibit activity ribosomica On the other hand, taking into account which residues that define the target site are conserved among pathogens but are not conserved among host species, the expert can design new inhibitors of the specific protein synthesis for the species. It is clear that one skilled in the art can take advantage of regions that are not conserved between E. coli and rat or human in order to provide target regions for a rational drug design. By way of example, Figure 33 shows certain regions of the polypeptide exit tunnel that are conserved between E. coli and rat or human (indicated in red) and the regions of the polypeptide exit tunnel that are not conserved between E. coli and rat or human (indicated in blue). Figures 33 (A) and 33 (B) provide enlarged views of a large ribosomal subunit when cut in the middle along the polypeptide exit tunnel. Figure 33 (C) is provided to orient the reader towards the view in Figure 33 (A) with respect to the large ribosomal subunit. Figure 33 (D) is provided to orient the reader in view in Figure 33B in relation to the large ribosomal subunit. In addition, the expert, when in possession of mutations that prevent or reduce antibiotic activity (i.e., are related to antibiotic resistance) can use this information to model the relevant antibiotic binding product that can then be used as a basis for the rational design of drugs in order to identify small molecules that overcome drug resistance. It is contemplated that it is possible to use a variety of computer modeling procedures, for example homology modeling protocols, to provide a model of an objective site of drug resistance by applying site-directed mutagenesis of nucleotides and / or amino acids and then, using an adequate minimization of the appropriate energy and refinement protocols.

C. Identification of candidate molecules

It is considered that it is possible to completely disseminate de novo candidate molecules that inhibit protein biosynthesis, or that they can be based on a pre-existing inhibitor of protein biosynthesis. Any of these approaches can be facilitated by the selection by computer discard of databases and libraries of small molecules for chemical entities, agents, ligands, or compounds that can be bound in whole, or in part, to the ribosomes and the ribosomal subunits, more preferably, to large ribosomal subunits, and more preferably even 50S ribosomal subunits. In this systematic selection, the quality of the fit or fit of said entities or compounds to or to the union sites can be judged either by the complementarity of the form or by the estimated energy of the interaction (Meng et al (1992) J. Comp Chem. 13: 505-524).

The design of molecules that bind to, or inhibit the functional activity of ribosomes or ribosomal subunits according to this invention generally involves the consideration of two factors. First, the molecule must be able to physically and structurally associate with the large ribosomal subunit. Non-covalent molecular interactions important in the association of ribosomes and ribosomal subunits with the molecule include hydrogen bonds, van der Waals interactions and hydrophobic interactions. Second, the molecule must be able to adopt a conformation that allows it to associate with ribosome ribosome subunits, more preferably, with large ribosomal subunits, and more preferably even with the 50S ribosomal unit. While it is possible that certain portions of the molecule do not participate directly in this association with a ribosome or with ribosomal subunits, such portions may still influence the overall conformation of the molecule. This, in turn, can have a significant impact on binding affinities, therapeutic efficacy, drug qualities, and potency. Such conformational requirements include the overall three-dimensional structure and the orientation of the chemical entity or molecule in relation to all or a portion of the active site or other region of ribosome ribosomes or subunits, or the spacing or separation between functional groups of a molecule comprising several chemical entities that interact directly with ribosomes or ribosomal subunits, more preferably, with large ribosomal subunits, and more preferably even with the 50S ribosomal unit.

The potential, predicted, inhibitory or binding effect of a molecule on ribosomes and ribosomal subunits can be analyzed before its actual synthesis before testing, by using computer modeling techniques. If the theoretical structure of the given molecule suggests insufficient interaction and association between it and ribosome ribosomes or subunits, the synthesis and the test of the molecule are obvious. However, if computer modeling indicates strong interaction, the molecule can then be synthesized and its ability to interact with ribosomes or ribosomal subunits and its ability to inhibit protein synthesis is tested. In this way, the synthesis of inoperative molecules can be avoided. In some cases, inactive molecules predicted in modeling are synthesized and then tested to develop a SAR (structure-activity relationship, relationship between structure and activity) for molecules that interact with a specific region of the ribosome or ribosomal subunit , more preferably, of the large ribosomal subunit, and even more preferably of the 50S ribosomal unit. As used herein, the term "SAR" will collectively refer to the structure-activity / structure property relationships relative to the relationship (s) between the activity and the properties of a compound and its structure chemistry.

d. De novo design

A person skilled in the art can use one of several methods to identify chemical parts or entities, compounds, or other agents depending on their ability to associate with a preselected target site within a ribosome or a ribosomal subunit. This process can begin by visual inspection or by computer-assisted modeling of, for example, the target site on the computer screen based on the atomic coordinates of the 50S ribosome unit and / or its complexes with other analogs and antibiotics, registered in the RCSB Protein Data Bank with PDB access numbers; ID 1FFK, 1JJ2, 1FFZ, 1FG0, 1K73, 1KC8, 1K8A, 1KD1, or 1K9M, and / or content on Disk ND 1. In one embodiment, computer modeling programs that calculate as different are used in the compound design Molecules interact with the various sites of the ribosome, the ribosomal subunit, or a fragment thereof. It is then possible to position selected chemical parts or entities, selected compounds, or selected agents, in a variety of orientations, or docked, within at least a portion of the target site of a ribosome or a ribosomal subunit, more preferably, of a large ribosomal subunit, and even more preferably, of a 50S ribosomal subunit. Chemical structure databases of, for example, Cambridge Crystallographic Data Center (Cambridge, United Kingdom) and Chemical Abstracts Service (Columbus, OH) are available. Coupling can be done using software such as Cerius, Quanta or Sybyl, followed by energy minimization and molecular dynamics with standard molecular mechanical force fields, such as OPLSAA, CHARMM or AMBER.

Some specialized computer programs can also help in the process of selecting chemical entities. They include without limitation:

(one)

GRID (Goodford, P. J., "A Computational Procedure for Determining Energetically Pro Binding Sites on Biologically Important Macromolecules" (1985) J. Med. Chem. 28, 849-857). A software such as GRID, which is a program that determines probable sites of interaction between ribosomal subunits with various characteristics of functional groups and the macromolecular surface, can be used to analyze surface sites in order to determine similarly inhibitory protein structures or molecules. . GRID calculations, with suitable inhibitor groups on molecules (for example, protonated primary amines) such as the probe, are used to identify potential critical sites around accessible positions at adequate energy contour levels. GRID is provided by Oxford University, Oxford, UK.

(2)

MCSS (Miranker, A. and M. Karplus (1991) "Functionality Maps of Binding Sites: A Multiple Copy Simultaneous Search Method." Proteins: Structure, Function and Genetics 11: 29-34). The MCSS is provided by Molecular Simulations, Burlington, MA.

(3)

AUTODOCK (Goodsell, D. S. and A. J. Olsen (1990) "Automated Docking of Substrates to Proteins by Simulated Annealing" Proteins: Structure, Function, and Genetics 8: 195-202). The AUTODOCK is provided by Scripps Research Institute, La Jolla, CA.

(4)

DOCK (Kuntz, I. D. et al. (1982) "A Geometric Approach to Macromolecule-Ligand Interactions" J. Mol. Biol. 161: 269-288). The DOCK program can be used to analyze an active site or a ligand binding site and suggests ligands with complementary ester properties. DOCK is provided by the University of California, San Francisco, CA.

(5)

ALADDIN (Van Drie et al. (1989) "ALADDIN: An Integrated Tool of Computer Assisted Molecular Design and Pharmacophore Recognition From Geometric, Steric and Substructure Searching of Three-Dimensional Structures"

J.

 Comp-Aided Mol. Des. 3: 225).

(6)

CLIX (Davie and Lawrence (1992) "CLIX: A Search Algorithm for Funding Novel Ligands Capable of Binding Proteins of Known Three-Dimensional Structure" Proteins 12: 31-41).

(7)

GROUPBUILD (Rotstein and Murcko (1993) "GroupBuild: A Fragment-Based Method for De Novo Drug Design" J. Med. Chem 36: 1700).

(8)

GROW (Moon and Howe (1991) "Computer Design of Bioactive Molecules: A Method for Receptor-Based De Novo Ligand Design" Proteins 11: 314).

Once suitable parts or entities, compounds or agents, chemicals have been selected, they can be assembled in the form of a single molecule. The assembly can be carried out by visual inspection and / or computer modeling and computational analysis of the spatial relationship of the parts or entities, compounds or agents, chemicals, with respect to each other in three-dimensional space. This could be followed by model building using software such as Quanta or Sybyl.

Useful programs to help the person skilled in the art in the interconnection of entities, compounds or agents, individual chemicals, include without limitation:

(one)

CAVEAT (Bartlett, PA et al. (1989) "CAVEAT: A Program to Facilitate the Structure-Derived Design of Biologically Active Molecules." Intermolecular Recognition in Chemical and Biological Problems ", Special Pub., Royal Chem. Soc. 78: 82 -196) and (Bacon et al. (1992) J. Mol. Biol. 225: 849-858) CAVEAT uses databases of cyclic compounds that can act as "separators" to connect any number of chemical fragments already positioned in the active site.This allows an expert in the art to quickly generate hundreds of possible ways to interconnect fragments that are already known or suspected of being necessary for a close union.The CAVEAT is provided by the University of California, Berkeley , CA.

(2)

3D database systems such as MACCS-3D (MDL Information Systems, San Leandro, (CA). This area has been reviewed in: Martin, YC, (1992) "3D Database Searching in Drug Design", J Med. Chem. 35: 2145-2154.

(3)

 HOOK (provided by Molecular Simulations, Burlington, MA.).

Instead of constructing a molecule of interest in stages at the rate of a chemical entity at a time as described above, it is possible to dissect the molecule of interest as a whole using an empty active site or optionally with the inclusion of some or more portions of one or more more known inhibitors. The software that allows to implement these methods includes:

(one)

LUDI (Bohm, H.-J. (1992) "The Computer Program LUDI: A New Method for the De Novo Design of Enzyme Inhibitors" J. Comp. Aid. Molec. Design 6: 61-78). The LUDI program can determine a list of interaction sites in which to place both hydrogen bonds and hydrophobic fragments. The LUDI uses

then a library of approximately 600 linkers to connect up to four sites of interaction in fragments. Next, smaller "bridging" groups such as -CH2- and -COO- are used to connect these fragments. For example, for the DHFR enzyme, fragments of key functional groups in methotrexate, which is a well known inhibitor, has been reproduced by LUDI. See also, Rotstein and Murcko, (1992) J. Med. Chem. 36: 1700-1710. The LUDI is provided by Biosym Technologies, San Diego, CA.

(2)

LEGEND (Nishibata, Y. and A. Itai (1991) Tetrahedron 47, 8985). The LEGEND is provided by Molecular Simulations, Burlington, MA.

(3)

 LeapFrog (provided by Tripos Associates, St. Louis, MO.).

(4)

 Aladdin (provided by Daylight Chemical Information Systems, Irvine, CA)

It is also possible to employ other molecular modeling techniques in accordance with the present invention. See, for example Cohen, N. C. et al. (1990) "Molecular Modeling of software and Methods for Medicinal Chemistry, J. Med. Chem. 33: 883-894. See also, Navia, MA and MA Murcko (1992)" The Use of Structural Information in Drug Design ", Curr. Opinions in Structural Biology 2: 202-210; and Jorgensen (1998) "BOSS- Biochemical and Organic Simulation System" in the Encyclopedia of Computational Chemistry (PVR Schleyer, ed.) Wiley & Sonstra., Athens, USA 5: 3281-3285 ).

It is considered that during the modeling it may be necessary to introduce into the molecule of interest chemical parts that may be beneficial for a molecule to be administered as a pharmaceutical product. For example, it may be possible to introduce or omit from the molecule of interest, chemical parts that may not directly influence the linkage of the molecule to the target area but that contribute, for example, to the overall solubility of the molecule in a pharmaceutically acceptable vehicle or carrier, to the bioavailability of the molecule and / or to the toxicity of the molecule. Considerations and methods can be found to optimize the pharmacology of the molecule of interest, for example in: "Goodman and Gilman's The Pharmacological Basis of Therapeutics" Eighth Edition (Goodman, Gilman, Rall, Nies, & Taylor (eds.)). Pergaman Press (1985); Jorgensen & Duffy (2000) Bioorg. Med. Chem. Lett. 10: 1155-1158.

On the other hand, the computer program "Qik Prop" can be used to obtain rapid predictions in terms of physically significant descriptions and in terms of the pharmaceutically relevant properties of an organic molecule of interest. It is possible to use a "five rule" probability scheme to estimate the oral absorption of the newly synthesized compounds (Lipinski et al. (1997) Adv. Drug Deliv. Rev. 23: 3).

Appropriate programs for the selection and design of drug agents include:

(one)

 DISC (Abbot Laboratories, Abbot Park, I11.).

(2)

 Catalyst (Bio-CAD Corp., Mountain View, CA).

(3)

 Chem DBS-3D (Chemical Design Ltd., Oxford, R.U.).

On the other hand, the skilled person can use the available information about how to design therapeutically active compounds and pharmaceutically useful, suitable compounds, and use this information in the design of the novel inventive inhibitors of protein synthesis. See, for example, Lipinski et al. (1997) Ad. Drug Deliv. Reviews 23: 3-25; Van de Waterbeemd et al. (1996) Quantitative Structure-Activity Relationships 15: 480490; and Cruciani et al. (2000), Theochem-J. Mol. Struct. 503: 17-30.

The entry of the coordinates of the proteins and RNAs of the ribosome or of the ribosome subunit into the computer programs set forth above results in the calculation of the most likely structure of the macromolecule, including the overall atomic coordinates of a ribosome, of a ribosomal subunit or a fragment thereof. It is possible to combine and refine these structures by means of additional calculations in which said programs are used to determine the three-dimensional structure, probable or real, of the ribosome, ribosomal unit or binding sites, of the ligands.

and. Modification of existing molecules

Instead of designing completely de novo molecules of interest, it is considered that it is possible to use pre-existing molecules or proteins thereof as a starting point for the design of a new candidate. It is considered that many of the approaches useful for designing de novo molecules can also be useful for modifying existing molecules.

It is considered that knowledge of a spatial relationship between an inhibitor of protein biosynthesis, for example, an antibiotic, and of their respective binding sites within a ribosome, allows the design of modified inhibitors that may have better binding properties. , for example, a greater affinity of binding and / or specific character, with respect to the molecule from which it has been derived. Alternatively, knowledge of inhibitor contact sites within a ribosome allows the synthesis of a new molecule that contains, for example, a portion of a first molecule (for example, an antibiotic or analogue or derivatives thereof) that is links to the contact site and another contributing portion that provides additional functionality.

It is considered that it is possible to design a variety of modified molecules (for example, modified antibiotics) by means of the atomic coordinates provided herein. For example, it is considered that knowing the spatial relationship of one or more of the antibiotics with respect to the large ribosomal subunit it is possible to generate new molecules based on antibiotics. The atomic coordinates of each antibiotic with respect to the large ribosomal subunit provide information about which portions of the ribosome or ribosomal subunit and of the antibiotic come into contact with each other. Consequently, from this information the person skilled in the art can not only identify the contact locations within the ribosomes that can be used for de novo drug design, as discussed above, but can also identify portions of a antibiotic that act as a ribosome binding domain.

On the basis of the information provided herein, the person skilled in the art can easily identify and produce hybrid antibiotics comprising a ribosome binding domain of a first antibiotic or an analogue.

or derivative thereof and a ribosome binding domain of a second, different antibiotic, or an analog or a derivative of the second. The resulting hybrid antibiotics preferably bind to each of the respective contact locations within the ribosomal subunit at the same time. The atomic coordinates provided herein allow the person skilled in the art to identify candidate antibiotics that can be used as templates in the synthesis of a hybrid, and also provide the necessary information to produce binding chemicals such that each ribosome binding domain It is oriented correctly with respect to its respective contact site. As a result, it is contemplated that one skilled in the art can produce a hybrid antibiotic that binds to a ribosome or ribosomal subunit with a higher affinity and / or has an inhibitory activity of protein synthesis, which are greater than of any of the individual template antibiotics used to generate the hybrid. Alternatively, the hybrid antibiotic may overcome the resistance of phenotypes that may have developed against any of the template antibiotics. For example, the proximity of the site occupied by the discarded part of carbomycin to the site occupied by anisomycin suggests that a hybrid compound that includes portions of both carbomycin and anisomycin may be an effective inhibitor of protein synthesis.

In addition, the atomic coordinates provided herein allow the person skilled in the art to use the information necessary to identify a ribosome binding domain and disseminate other types of protein synthesis inhibitors. For example, with an understanding of the ribosome contact region and the surrounding environment, the person skilled in the art can provide new molecules, a portion of which is based on the antibiotic binding region (binding domain) and another portion of which (the effector domain) can be designed as a novel space-filling domain that sterically inhibits or interrupts the biosynthesis of the proteins within the ribosomes or the secretion through the polypeptide exit tunnel. For example, the person skilled in the art can combine the ribosome binding region of the antibiotic, tylosin or an analogue or derivative thereof, which is attached to one side of the exit of the polypeptide tunnel near the peptidyltransferase site, with, by For example, a novel chemical part not present in the antibiotics identified to date that is bulky enough to block the polypeptide exit tunnel. However, it is contemplated that one skilled in the art can take advantage of one or more of the many antibiotic contact regions described herein, to design completely new binding and effector domains.

It is preferable that the inhibitors resulting from the synthesis of proteins have a molecular weight not exceeding approximately 1,500, more preferably, not exceeding approximately 1,000, more preferably, not exceeding 750 and, more preferably, still not exceeding about 500. Protein synthesis inhibitors preferably have a molecular weight in the range of about 250 to about 1,500, and more preferably, in the range of about 500 to about 1,200. In addition, protein synthesis inhibitors have a minimum inhibitor concentration preferably not more than 50 mM, more preferably, not more than 10 mm, and more preferably, not more than 1 mM to inhibit 50% activity ( IC50) in a biological test, for example, in an in vitro translation test, for example, an E. coli translation test. Protein synthesis inhibitors preferably have an IC50 in the range of about 0.001 mM to about 50 mM, or in the range of about 0.01 mM to about 10 mM, or in the range of about 0.1 mM to about 1 mM

On the other hand, the present invention allows the person skilled in the art to design molecules, for example, inhibitors of the synthesis of selective proteins adapted to be more potent with respect to the ribosomes of a target organism, for example, a pathogenic agent such as a microbe, and less potent, that is, less toxic, for the ribosomes of a non-target organism, for example, a host organism such as a human being. In addition, the invention allows the person skilled in the art to use the atomic coordinates and structures of the large ribosomal subunit and their complexes with protein synthesis inhibitors to disseminate modifications in the starting compounds, such as an antibiotic, that will bind more closely to an objective ribosome (for example, the 50S ribosomal unit of bacteria) and less strongly to a non-objective ribosome (for example, the 60S human ribosome subunit or to the human mitochondrial ribosome).

The structure of a complex between the large large ribosomal subunit and the starting compound (for example, tylosin or other protein synthesis inhibitor) can also be used to guide the modification of said compound in order to produce new compounds having other desirable properties for industrial and other purposes (for example, in the form of pharmaceuticals, herbicides or insecticides), such as chemical stability, solubility or membrane permeability.

A variety of antibiotics binds to the large ribosomal subunit and undoes protein synthesis, and include members of the antibiotic families that include, without limitation, chloramphenicols, macrolides, lincosamides, streptogramins, oxazolidinone althomycins, nucleotide analogs, thioestreptones (including the family of micrococcines), peptides, glutarimides, trichothecenes, TAN-1057, pleuromutilins, hygromycins, betaines, everninomycins, boxazomycins and fusidans.

Members of the chloramphenicol family include, for example, chloramphenicol and iodofenfenicol. Members of the macrolide family include, for example, Biaxin (Clarithromycin), Zithromax (Azithromycin; azalide), Ketek (Telithromycin, cetolide), ABT-773, tylosin, spiramycin I, spiramycin II, spiramycin III, erythromycin A, carbomycin A, telithromycin methmicin, narbomycin, Lankamycin, Oleandomycin, Megalomycin, Chalcomycin, Niddamycin, Leucomycin, Angolamicin, Picromycin, and Relomycin. Members of the licosamide family include, for example, Clindamycin and Lincomycin. Members of the streptogramin family include, for example, streptogramin A, streptogramin B, Ostreogricin G, Sinercida, Virginiamicin S1, Virginiamycin S2, Virginiamicin S3, Virginiamycin S4, Vernamicin B, Vernamicin C, Patricin A, and Patricin B. A member of the altiomycin family, includes, for example, altiomycin. Members of the oxazolidinone family include, for example, Linezolid, Eperezolide and DuP721. Members of the family of nucleotide analogs include, for example, Esparsomycin, Puromycin, Anisomycin and Blasticidine S. Members of the thiostreptone family include, for example, Thiostreptone, Siomycin, Sporangiomycin, Micrococcin M1, Micrococcin P and Thiopeptin . Members of the peptide family include, for example, Viomycin, Capreomycin IA, Capreomycin IB, Capreomycin IIA, and Capreomycin IIB. Members of the glutarimide family include, for example, Cycloheximide, Streptovitacin, Streptimidone, Inactone, Actiphenol. Members of the trichothene family include, for example, Tricodermin, Tricodermol, Tricodermone, Vomitoxin, T-2 toxin, Tricothecin, Nivalenol and Verrucarina A. Tan-1057 includes Tan-1057A, Tan-1057B, Tan-1057C, and Tan-1057D. Pleuromutilins include, for example, Pleuromutilin, Tiamulin, Azamiline and Valnemulin. Hygromycin include the antibiotics Hygromycin

A. Betaines include, for example, the family of natural betaine products and VCR4219. Eveminomycins include, for example, Ziracin, Avilamycin, Evernimycin and Curamycin. Boxazomycins include, for example, Boxazomycin A and B. Fusidans include, for example, fusidic acid and 17S, diethylene glycol acid diethylene glycol hydrate.

The inhibitors can be diffused into, or soaked with, the stabilized crystals of the large ribosomal subunit as described in Examples 4 and 5 so as to form a complex with the large ribosomal subunit to collect X-ray diffraction data. As an alternative, the inhibitors can be co-crystallized with the large ribosomal subunit by mixing the inhibitor with the large ribosomal subunit before its precipitation by a high salt content.

From the ribosome structure of H. marismortui, the ribosome structure of a non-target organism (for example, the human 60S ribosome subunit) can be constructed by homology modeling, that is, by changing the structure of the residues to a target site of interest for residues in the same positions of the non-target ribosome. This can be achieved by the computational elimination of the ribosome side chains of the known structure and their replacement by the side chains of the unknown structure placed in sterically plausible positions. In this way, it can be understood how the shapes of the target sites differ within the targeted and non-targeted ribosomes. This process, therefore, provides information concerning how a molecule that binds to the target site can be chemically altered in order to produce molecules that bind strongly and specifically to the target ribosomes, but those that at the same time are prevented Join the non-objective ribosome. Similarly, knowledge of the portions of the bound molecules that face the solvent, allow the introduction of other functional groups for additional pharmaceutical purposes. The homology structure modeling process can also be used to understand the mechanisms by which mutant ribosomes become resistant to the effects of pharmaceuticals or pesticides, such as herbicides or insecticides. In addition, with the knowledge of the portions of the ribosomal subunit involved in drug resistance, the person skilled in the art can design new molecules that overcome the problem of drug resistance. The use of homologous structure modeling to disseminate molecules that bind more closely to the objective ribosome than to the non-objective ribosome has wide diffusion applicability. The methods described herein can be used to control any target organism, for example, a pathogenic agent, by designing molecules that inhibit large ribosomal subunits of the target organisms and that at the same time do not inhibit the 50S or 60S ribosomal subunit of the non-target organism, for example, a host, in the same amplitude or not at all. Molecules identified or prepared by the methods of the present invention can be used to control specific organisms, despite having little or no adverse effect on the non-targeted organism. Therefore, the molecules identified or developed using the methods of the present invention can be designed in such a way that their administration kills the target organisms or inhibits some aspect of the biological functions of the target organisms and at the same time does not have a similar effect. in the non-targeted organism. Adverse effects of the agent on the target organisms may include, without limitation, the death of the target organism; the deceleration of growth rates; slow down or eliminate the passage from one growth phase to another (for example, by extending the larval growth stage); the deceleration or elimination of reproduction, the decrease or prevention of mating, the decrease or elimination of the production of offspring, the limitation or elimination of increases in body weight of the targeted organisms; the reduction or elimination of capacity and behaviors related to food; and the alteration or destruction of cellular, tissue and / or organ functions.

The novel agents covered by the present invention can be useful as herbicides, pesticides (for example, insecticides, nematicides, rodenticides, etc.), acaricides, or antimicrobial agents (for example, antifungal, antibacterial, antiprotozoal, etc.) with respect to specific target organisms. For example, the novel agents can be targeted to parasitic nematodes of plants and animals, prokaryotic organisms (disease-causing microbes), and eukaryotic multicellular pests. Specific examples of multicellular pests include, without limitation, insects, fungi, bacteria, nematodes, mites and ticks, protozoal pathogens, hepatic animal parasitic trematodes, and the like.

Herbicides, pesticides, acaricides, and antimicrobial agents that inhibit protein synthesis through interaction with ribosomes are known to those skilled in the art. Some examples are set out below. These known agents can be modified to obtain novel agents through the use of computer modeling techniques and knowledge of the structure of ribosomes and ribosomal subunits and the structure of ribosome / ribosomal subunit / agent complexes.

Ketolide ABT-773 binds ribosomes more tightly than erythromycin in S. pneumoniae and is capable of defeating macrolide resistance in bacteria (Capobianco et al (2000) Antimicrob Agents Chemother 44 (6): 1562/67). The tools and methodologies of the present invention can be used to obtain erythromycin derivatives that bind to ribosomes or ribosomal subunits more strongly than that with which they bind to ribosomes and ribosomal subunits of non-target animals. The target bacteria can be any infectious bacteria, in particular S. pneumoniae, and even more particularly S. pneumoniae resistant to erythromycin. Non-object animals can be any animal, particularly mammals, and even more particularly, humans.

Examples of antibiotics that are inhibitors of protein synthesis include without limitation: puromycin, cycloheximide, chloramphenicol, tetracycline, and streptomycin (Heldt (1996) Plant Biochemistry and Molecular Biology

21.2: 458-464). Puromycin, as discussed above, binds as an analogue of an aminoacyl-tRNA to the An site and binds to the chains of nascent peptides, and avoids additional stages of elongation in prokaryotes and eukaryotes. Cycloheximide inhibits peptidyltransferase in eukaryotic ribosomes. Chloramphenicol inhibits peptidyltransferase in prokaryotic ribosomes. Tetracycline binds to the 30S subunit and inhibits the binding of aminoacyl-tRNA to prokaryotic ribosomes much more than to eukaryotes. Streptomycin interacts with 30S ribosomes which results in an incorrect recognition of mRNA sequences and therefore inhibits initiation in prokaryotic ribosomes. U.S. Patent UU. N D 5,801,153 discloses antibiotics against pathogens. Aminoglycosides are examples of antibacterial antibiotics that appear to inhibit protein synthesis. However, there is a limitation for its use due to its cytotoxic and nephrotoxic properties. Amicacin sulfate, framicetin sulfate, gentamicin sulfate, kanamycin sulfate, neomycin sulfate, netilmicin sulfate, paromomycin sulfate, sissomycin sulfate, tobramycin, vancomycin hydrochloride, and viomycin sulfate are members of the family of the aminoglycosides. The tools and methodologies of the present invention can be used to obtain derivatives of any antibiotic of choice so as to inhibit the synthesis of proteins of the target organisms to a greater extent with which they inhibit the synthesis of proteins of non-target organisms, such as Humans.

Examples of objective and non-objective organisms include without limitation those provided in Table 21.

Table 21

Examples of classes of molecules that can be identified and / or developed by the methods of the invention and applicable target / non-target organisms

Type of molecule

Target organizations Non-target organizations

Herbicides

Monocot plants

Herbicides

Lawn Soy, potato, coffee

Insecticides

Flies, mites Honeybees

Pesticides

Ticks Deer

Pesticides

Lice Birds

Acaricides

Parasitic faces (scabies) Dogs

Antimicrobial agents (antibacterial agents)

Streptococcus pneumoniae Humans

Antimicrobial agents (antibacterial agents)

Clostridium difficile Escherichia coli

Antimicrobial agents (antifungal agents)

Erysiphe graminis Barley

Antimicrobial agents (antiprotozoal agents)

Toxoplasma gondii Animals

Poisons (rodenticides)

Rats Dogs, cats, humans

It is considered that the tools and methodologies of the present invention can be used to obtain inhibitors of the synthesis of target insect proteins, such as capsule worms and mosquitoes, to a greater extent with which they inhibit the synthesis of non-targeted insect proteins, such like the beetles of the family Coccinellidae (for example, ladybugs) and Apis mellifera (bees). Other target insects include, without limitation, insects selected from the orders Coleoptera (beetles), Diptera (flies, mosquitoes), hymenoptera (wasps, ants, sawflies), lepidoptera (butterflies and moths), Mallophaga (lice), Homoptera ( whiteflies, aphids, hemipteros (bedbugs), Orthroptera (lobsters, cockroaches), Thysanoptera (trips), Dermaptera (earwigs), Isoptera, Anopluros, Siphonaptera and Trichoptera (caddis flies).

In addition, it is considered that the tools and methodologies of the present invention can be used to obtain inhibitors of the synthesis of target plant proteins that inhibit the synthesis of target plant proteins more than inhibit the synthesis of plant and animal proteins non objective. The target plants can be any species of unwanted plants, in particular particular weeds, and even more particularly harmful weeds. Whether a particular plant is a weed will depend on the context in which it is growing. For example, unwanted Zea mays (corn) plants that grow in a Glycine max (soybean) field could be considered unwanted weeds. Examples of weeds that are likely target plants include without limitation: Allium vineale (wild garlic), Bromus tectorum (hairy bromine), Triticum cylindricum (articulated goat grass), Amaranthus spp. (porcine weeds), Chenopodium album, Avena fatua (wild oats), B. secalinus, Echinochloa crus-galli (crow's feet), Alopecurus myosuroides,) Setaria faberii (giant foxtail), Xanthium strumarium (common cadillo), Ambrosia artemisiifolia (common ambrosia), and Ipomoea spp. (bells). Non-target organisms can be from any plant, particularly any desirable plant, and even more particularly any crop plant. Non-target organisms can also be any animals, particularly mammals, and even and more particularly in humans. In a preferred embodiment, the tools and methodologies of the present invention can be used to produce protein synthesis inhibitors that kill or injure one or more species of harmful weeds, but which do not damage plants and non-object animals.

Bacteria of interest include without limitation: Staphylococcus aureus, Streptococcus pyogenes, Streptococcus agalactiae, Streptococcus bovis, Streptococcus pneumoniae, Moraxella catarrhalis, Neisseria gonorrhoeae, Neisseria meningitides, Bacillus anthracis, Corynebacterium diphtheriae, Listeria monocytogenes, Erysipelothrix rhusiopathiae, Clostridium perfringens, Clostridium tetani, Clostridium difficile, Escherichia coli, Proteus mirabilis, Psuedomonas aeruginosa, Klebsiella pneumoniae, Haemophilus influenzae, Haemophilus ducreyi, Yersinia pestis, Yersinia enterocolitica, Francisella tularensis, Pasteurella multocida, Vibrio cholerae, Flavobacterium meningosepticum, Pseudomonas mallei, Pseudomonas pseudomallei, Campylobacter jejuni, Campylobacter fetus , Fusobacterium nucleatum, Calymmatobacterium granulomatis, Streptobacillus moniliformis, Legionella pneumophila, Mycobacterium aviumintracellulare, Mycobacterium tuberculosis, Mycobacterium leprae, Treponema pallidum, Treponema pertenu and, Borrelia burgdorferi, Borrelia recurrentis, Actinomyces isrealii, Nocardia asteroides, Ureaplasma urealyticum, Mycoplasma pneumoniae, Chlamydia psittaci, Chlamydia trachomatis, Chlamydia pnemoniae, Pneumocystis carinii, Coccidioides immitis, Histoplasma capsulatum, Blastomyces dermatitidis, Paracoccidioides brasiliensis, Sporothrix schenckii, Cryptococcus neoformans.

Once a candidate molecule has been designed or selected by the methods described above, the affinity with which this molecule can bind to the ribosome or ribosomal subunit can be tested and optimized by computational evaluation and / or by biological activity assays. after synthesizing the compound. Candidate molecules can interact with ribosomes or ribosomal subunits in more than one conformation each of which has a similar overall binding energy. In these cases, the bond deformation energy can be considered as the difference between the energy of the free molecule and the average energy of the conformations observed when the molecule binds to the ribosomes

or to ribosomal subunits, more preferably, to large ribosomal subunits, and more preferably even to 50S ribosomal units.

A molecule designed or selected for binding to a ribosome or ribosomal subunit can also be optimized by computer so that in its bound state it preferably lacks a repulsive electrostatic interaction with respect to the target region. Such non-complementary interactions (eg electrostatic) include repulsive charge-charge, dipole-dipole and charge-dipole interactions. Specifically, the sum of all electrostatic interactions between the inhibitor and the enzyme, when the inhibitor is bound to the ribosome or ribosomal subunit, preferably makes a neutral or favorable contribution to enthalpy of binding. It is also possible to design weak binding compounds by these methods in order to determine SAR.

In the art there are specific computer programs that can evaluate a strain energy and the electrostatic interaction of a compound. Examples of suitable programs include: Gaussian 92, revision C (M. J. Frisch, Gaussian, Inc., Pittsburgh, PA.); AMBER, version 4.0 (P. A. Kollman, University of California at San Francisco, CA); QUANTA / CHARMM (Molecular Simulations, Inc., Burlington, MA); OPLS-AA ("OPLS Force Fields." WL Jorgensen. Encyclopedia of Computational Chemistry, Schleyer, Ed .; Wiley: New York, 1998; Vol. 3, pp 1986-1989) and Insight II / Discover (Biosysm Technologies Inc., San Diego, CA). These programs can be implemented, for example, using a Silicon Graphics workstation, IRIS 4D / 35 or an IBM RISC / 6000 model 550 workstation. Other hardware systems and software packages are known to experts.

Once a molecule of interest has been selected or designed, as described above, substitutions can then be made in some of its atoms or lateral groups in order to improve or modify its binding properties. In general, the initial substitutions are conservative, that is, the replacement group will approximate the same size, shape, hydrophobicity and charge as the original group. Of course, it should be understood that the components known in the art for altering the conformation should be avoided. Such substituted chemical compounds can then be analyzed to establish the efficiency of their adjustment to the ribosome or ribosomal subunit by the same computational methods described in detail, in the foregoing.

In addition, ligands related to real, complex or mimetic ribosomes can be crystallized and analyzed by X-ray diffraction. The coordinates of the diffraction pattern are similarly used to calculate the three-dimensional interaction of a ligand and the ribosome, ribosome subunit, or a mimetic, in order to confirm the ligand bind to, or change the conformation of, a particular site in the ribosome or ribosomal subunit, or where the mimetic has a three-dimensional structure similar to that of a ribosome, ribosome subunit or a fragment thereof.

3. Synthesis of leader molecules

A leader molecule of the present invention can be without limitation, at least one selected from a lipid, a nucleic acid, peptide, inorganic or organic small molecule, chemical compound, element, saccharide, isotope, carbohydrates, imaging agent, lipoprotein , glycoprotein, enzyme, analytical probe, and an antibody or fragment thereof, any combination of any of the foregoing, and any chemical or variant modification of any of the foregoing. In addition, a leader molecule may optionally comprise a detectable label. Such labels include without limitation enzymatic labels, radioisotopes or radioactive compounds or elements, fluorescent compounds or metals, chemiluminescent compounds and bioluminescent compounds. Well-known methods can be used for fixing such a detectable label to a leading molecule.

Useful methods for synthesizing leading molecules such as lipids, nucleic acids, peptides, inorganic or organic small molecules, chemical compounds, elements, saccharides, isotopes, carbohydrates, imaging agents, lipoproteins, glycoproteins, enzymes, analytical probes, antibodies and Antibody fragments are well known in the art. Such methods include the traditional approach to the synthesis of such a leading molecule, such as a single peptide defined at the same time, as well as the combined synthesis of multiple leader molecules in one or more containers. These multiple leader molecules may include one or more variants of a previously identified leader molecule. The methods for the combined synthesis of multiple leader molecules are particularly useful in the preparation of combinatorial libraries, which can be used in systematic selection techniques known in the art.

By way of example, it is well known in the art that it is possible to synthesize multiple peptides and oligonucleotides simultaneously. Leading molecules that are small peptides up to 50 amino acids in length can be synthesized using standard solid phase procedures for peptide synthesis, for example by procedures similar to those described in Merrifield (1963) J. Am. Chem. Soc., 85: 2149. For example, during the synthesis, N-protected amino acids having protected side chains are added step by step to a growing chain of polypeptides linked by their C-terminal end to an insoluble polymeric support, for example, beads of polystyrene. Peptides are synthesized by linking an amino group of an N-deprotected amino acid to a carboxy group of an N-protected amino acid that has been activated by reacting it with a reagent such as dicyclohexylcarbodiimide. The fixation of a free amino group to the activated carboxyl leads to the formation of the peptide bond. The N- protecting groups (most commonly used include Boc which is labile in acidic medium and Fmoc which is labile in basic medium.

Simply put, the N-a-protected C-terminal amino acid binds first to polystyrene beads. Then the protective group N-a is removed. The deprotected amino group is coupled to the activated α-carboxylate group of the following N-α-protected amino acid. The process is repeated until the desired peptide is synthesized. The resulting peptides unfold from the insoluble polymeric support and from the side chains of unprotected amino acids. Longer peptides, for example longer than 50 amino acids, are typically derived by condensation of protected peptide fragments. The details of the appropriate chemical part, resins, protective groups, protected amino acids and reagents, are well known in the art and therefore are not set forth in detail herein. See for example, Atherton et al. (1963) Solid Pass Peptide Synthesis: A Practical Approach (IRL Press), and Bodanszky (1993) Peptide Chemistry, A Practical Textbook, 2nd Ed. Springer-Verlag, and Fields et al. (1990) Int. J. Peptide Protein Res. 35: 161-214.

Purification of the resulting peptide is carried out using conventional procedures, such as preparative HPLC, for example, gel permeation, partition and / or ion exchange chromatography. The choice of appropriate matrices and buffers is well known in the art and therefore are not described in detail herein.

It is considered that a synthetic peptide according to the invention may comprise naturally occurring amino acids, unnatural amino acids, and / or amino acids that have specific characteristics, such as, for example, amino acids that are positively charged, negatively charged, hydrophobic, hydrophilic, or aromatic As used herein, the term "naturally occurring amino acids" refers to the L isomers of the amino acids normally found in proteins. The predominant naturally occurring amino acids are glycine, alanine, valine, leucine, isoleucine, serine, methionine, threonine, phenylalanine, tyrosine, tryptophan, cysteine, proline, histidine, aspartic acid, asparagine, glutamic acid, glutamine, arginine, and lysine. Unless specifically indicated, all amino acids referred to herein are in the L-form. In addition, as used herein, the term "unnatural amino acids" refers to amino acids that are not naturally found in proteins. For example, selenomethionine.

Amino acids that are "positively charged" include any amino acid that has a positively charged side chain under normal physiological conditions. Examples of positively charged naturally occurring amino acids include for example arginine, lysine, and histidine. Conversely, "negatively charged" amino acids include any amino acid that has negatively charged side chains under normal physiological conditions. Examples of naturally occurring negative-charged amino acids that occur in nature include, for example, aspartic acid and glutamic acid.

As used herein, the term "hydrophobic amino acid" includes any amino acid having a non-polar, non-charged side chain, which is relatively insoluble in water. Examples of naturally occurring hydrophobic amino acids include, for example, alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan, and methionine. In addition, as used herein, the term "hydrophilic amino acid" refers to any of the amino acids having a polar side chain without a charge that is relatively soluble in water. Examples of naturally occurring hydrophilic amino acids include, for example, serine, threonine, tyrosine, asparagine, glutamine and cysteine.

Finally, and as used herein, the term "aromatic" refers to amino acid residues whose side chains have a delocalized conjugated system. Examples of aromatic residues include for example phenylalanine, tryptophan, and tyrosine.

With respect to the production of small non-peptide organic molecules that act as a ligand in the present invention, these molecules can be synthesized using standard organic chemistry procedures well known and well documented in the patent and other literature.

Many of the known methods useful in the synthesis of leader molecules of the present invention can be automated, or in some other way can be implemented on a commercial scale. As such, once a leading molecule has been identified as having a commercial potential, it is easy to produce massive amounts of said molecule.

4. Characterization of molecules

Molecules designed, selected and / or optimized by methods described above, once produced, can be characterized using a variety of assays known to those skilled in the art to determine if the compounds have biological activity. For example, the molecules can be characterized by conventional tests, including but not limited to, the tests described in the following, to determine if they have a predicted activity, a binding activity and / or a specificity of binding.

In addition, a high performance systematic selection can be used to accelerate the analysis using such assays. As a result, it may be possible to systematically select new molecules rapidly based on their ability to interact with a ribosome or ribosomal subunit by means of the tools and methods of the present invention. General methodologies for performing the high performance systematic selection are described in for example Devlin (1998) High Throughput Screening, Marcel Dekker; and in the US patent. UU. N.D. 5,763,263. High performance assays may use one or more different test techniques, including without limitation, those described below.

(one)

Surface union studies. A variety of binding assays can be useful in detecting new molecules in terms of their binding activity. One approach includes the SPR (resonant surface plasmon, surface plasmon resonance) that can be used to evaluate the binding properties to molecules of interest with respect to a ribosome, ribosomal subunit or a fragment thereof. The SPR methodologies measure the interaction between two or more real-time macromolecules through the generation of a surface plasmon of quantum mechanics. One device, (the BIAcore Biosensor RTM from Pharmacia Biosensor, Piscatawy, NJ) provides a focused beam of polychromatic light to the interface between a gold film (provided in the form of a disposable biosensor chip) and an intermediate compartment that can be User regulated. A "hydrogel" 100 nm thick composed of carboxylated dextran that provides a matrix for the covalent immobilization of the analytes of interest, is attached to the gold film. When the focused light interacts with the free electron cloud of the gold film, plasmon resonance is reinforced. The resulting reflected light is spectrally depleted at wavelengths that developed optimally in resonance. By separating the polychromatic light reflected in its component wavelengths (by means of a prism), and determining the frequencies that are depleted, the BIAcore establishes an optical interface that accurately reports on the behavior of surface plasmon resonance generated. If the plasmon resonance is designed as previously described (and therefore the depletion spectrum) it is sensitive to the mass in the evanescent field (corresponding approximately to the thickness of the hydrogel). If one of the components of an interacting pair is immobilized in the hydrogel, and the interaction comparator is provided through the intermediate compartment, the interaction between the two components can be measured in real time based on the accumulation of mass in the evanescent field and in its corresponding effects of plasmon resonance measured by the depletion spectrum. This system allows quick and sensitive measurement in real time of molecular interactions and without the need to label any of the components.

(2)

Immunodiagnostics and immunoassays. It is a group of techniques that can be used for the measurement of specific biochemical substances, usually at low concentrations in complex mixtures such as biological fluids, which depend on the specificity and high affinity shown by antibodies properly prepared and selected as to its complementary antigens. A substance to be measured must necessarily be antigenic - whether an immunogenic macromolecule or a small haptenic molecule. For each sample a limited, known amount of specific antibody is added and the fraction of the antigen that is combined with it is estimated, often expressed as a relationship between bound and free, for which a form of the radioisotope-labeled antigen is used as an indicator (radioimmunoassay), fluorescent molecule (fluoroimmunoassay), stable free radical l (spin immunoassay), enzyme (enzyme immunoassay), or other easily distinguishable label.

Antibodies can be labeled in a variety of ways, including: enzyme-linked immunosorbent assay (ELISA, enzyme-linked immunosorbent assay); radioimmunoassay (RIA, radioimmuno assay); fluorescent immunoassay (FIA); chemiluminescent immunoassay (CLIA, chemiluminescent immunoassay), and labeling the antibody with colloidal gold particles (immunooro).

Common test formats include the sandwich assay, the competitive or competitive assay, the latex agglutination assay, the homogeneous assay, the microtiter plate format and the microparticle based assay.

(3)

Enzyme linked immunosorbent assay (ELISA). ELISA is an immunochemical technique that avoids the risks of radiochemical products and the cost of fluorescence detection systems. Instead, the assay uses enzymes as indicators. ELISA is a form of qualitative immunoassay based on the use of antibodies (or antigens) that are attached to the surface of an insoluble vehicle, which is then used to "capture" the relevant antigen (or antibody) in the test solution. The antibody antigen complex is then detected by measuring the activity of a suitable enzyme that has previously been covalently bound to the antigen (antibody).

General methods and compositions are described in for example Crowther (1995) ELISA - Theory and Practice (Methods in Molecular Biology), human Press; Challacombe and Kemeny, (1998) ELISA and Other Solid Phase Immunoassays - Theoretical and Practical Aspects, John Wiley; Kemeny, (1991) A Practical Guide to ELISA, Pergamon Press; Ishikawa, (1991) Ultrasensitive and Rapid Enzyme Immunoassay (Laboratory Techniques in Biochemistry and Molecular Biology) Elsevier.

(4)

Colorimetric tests Colorimetry is any method of quantitative chemical analysis in which the amount

The amount of a compound is determined by comparing the color produced by the reaction of a reagent with both standard and test amounts of the compound, often using a colorimeter. A colorimeter is a device for measuring color intensity or differences in color intensity, whether visually or photoelectrically. Standard colorimetric assays to establish the enzymatic activity of beta-galactosidase are well known to those skilled in the art (see, for example, Norton et al (1985) Mol Cell Biol. 5: 281-290). A colorimetric assay can be carried out in whole cell lysates using Onitrophenyl-beta-D-galactopyranoside (ONPG, Sigma) as a substrate in a standard colorimetric beta-galactosidase assay (Sambrook et al (1989) Molecular Cloning - A Laboratory Manual. , Cold Spring Harbor Laboratory Press). There are also automated colorimetric assays available for the detection of betagalactosidase activity, as described in US Pat. UU. N D 5,733,720.

(5)

 Immunofluorescence assays. Immunofluorescence immunofluorescence microscopy assays are a technique in which an antigen or antibody is made fluorescent by conjugation with a fluorescent dye and then allowed to react with the antibody or complementary antigen in a tissue section or smear. The location of the antigen or antibody can then be determined by observing the fluorescence by microscopy under ultraviolet light.

A general description of immunofluorescent techniques appears, for example, in Knapp et al. (1978) Immunofluorescence and Related Staining Techniques, Elsevier; Allan, (1999) Protein Localization by Fluorescent Microscopy -A Practical Approach (The Practical Approach Series) Oxford University Press; Caul, (1993) Immunofluorescence Antigen Detection Techniques in Diagnostic Microbiology, Cambridge University Press. For detailed explanations of immunofluorescent techniques applicable to the present invention, see for example US Pat. UU. N.D. 5,912,176; U.S. Patent UU. N.D. 5,869,264; U.S. Patent UU. N. 5,866,319; and U.S. patent UU. N.D. 5,861,259.

(6)

 Fluorescence polarization. Fluorescence polarization (FP) is a measurement technique that can be easily applied to protein-protein and protein-ligand interactions in order to derive IC50 and Kd from the association reaction between two molecules. In this technique one of the molecules of interest is conjugated with a fluorophore. This is generally the smallest molecule in the system (in this case, the molecule of interest). The sample mixture, which contains both the ligand-probe conjugate and the ribosome, ribosomal subunit or fragment thereof, is excited with vertically polarized light. The light is absorbed by the fluorophors of the probe, and re-emitted shortly after. The degree of polarization of the emitted light is measured. The polarization of the emitted light depends on several factors, but especially on the viscosity of the solution and the apparent molecular weight of the fluorophore. With appropriate controls, changes in the degree of polarization of the emitted light depend only on changes in the apparent molecular weight of the fluorophore, which in turn depends on whether the probe-ligand conjugate is free in solution, or is bound to a receiver. FP-based binding assays have a number of important advantages, including the measurement of IC50 and Kd under true homogeneous equilibrium conditions, analysis speed and suitability for automation, and the ability to systematically select in cloudy suspensions and solutions colored.

(7)

 Protein synthesis. It is considered that, in addition to the characterization by the biochemical tests just described, the molecule of interest can also be characterized as a modulator (for example, an inducer of protein synthesis or an inhibitor of protein synthesis) of functional activity of the ribosome or ribosomal subunit.

Protein synthesis inhibitors can be tested at a cellular level. For example, the molecules of interest can be tested against organisms, microorganisms, by culturing the microorganism of interest in a medium that is contained without the molecule of interest. Growth inhibition may be an indication that the molecule may be acting as an inhibitor of protein synthesis.

On the other hand, it is possible to carry out more specific tests regarding the inhibition of protein synthesis by administering the compound to a complete organism, tissue, organ, organelle, cell, cell or subcellular extract, or to a preparation of purified ribosome and observing its pharmacological and inhibitory properties by determining for example its inhibition constant (IC50) to inhibit protein synthesis. The incorporation of 3H leucine or 35S methionine, or similar experiments, can be carried out to investigate the activity of protein synthesis.

A change in the amount or rate of protein synthesis in the cell in the presence of a molecule of interest indicates that the molecule is an inducer of protein synthesis. A decrease in the speed or amount of protein synthesis indicates that the molecule is an inhibitor of protein synthesis.

In addition, the antibacterial activity of the compounds of the present invention against bacterial pathogens can be demonstrated by the ability of the compound to inhibit the growth or development of defined strains of human pathogens. For this purpose, it is possible to assemble a panel of bacterial strains that include a variety of target pathogenic species, some of which contain resistance mechanisms that have been characterized. The use of such an organism panel allows to determine the relationships between structure and activity not only in terms of power and spectrum, but also with a view to obviating resistance mechanisms. The tests can be carried out in microtiter trays according to conventional methodologies published by: The National Committee for Clinical Laboratory Standards (NCCLS) guidelines (NCCLS. M7-A5-Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard -Fifth Edition. NCCLS Document M100- S12 / M7 (ISBN 1-56238-394-9).

H. Drug formulation and administration

It is considered that once identified, the active molecules of the invention can be incorporated into any suitable carrier before use. More specifically, the dose of the active molecule, the mode of administration and the use of the appropriate carrier will depend on the target organism and not the target of interest.

It is considered that, with respect to mammalian receptors, the compounds of interest can be administered by any conventional method known and / or used in the art. Therefore, as appropriate, administration may be oral or parenteral, including intravenous and intraperitoneal administration routes. In addition, administration may be by periodic bolus injections, or it may be made more continuous by intravenous or intraperitoneal administration from a reservoir that is external (for example, an intravenous bag). In certain embodiments, the compounds of the invention may be of type or quality for therapeutic purposes. That is, certain embodiments meet the standards of purity and quality control required for administration to humans. Veterinary applications also fall within the meaning provided herein.

Formulations, both for veterinary use and for human medical use, of drugs according to the present invention typically include such drugs in association with a pharmaceutically acceptable vehicle and optionally other therapeutic ingredient (s). The vehicle (s) should be "acceptable" in the sense of being compatible with the other ingredients of the formulations and not harmful to the recipient thereof. The intention is that the pharmaceutically acceptable vehicles, in this sense, include any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic agents and absorption retarders, and the like, compatible with pharmaceutical administration. The use of such media and agents for pharmaceutically active substances is known in the art. Except insofar as any conventional agent or medium is incompatible with the active compound, their use in the compositions is considered. Complementary active compounds (identified or designed according to the invention and / or known in the art) can also be incorporated into the compositions. The formulations can be conveniently presented in unit dosage form and can be prepared by any of the methods well known in the pharmacy / microbiology technique. Usually, some formulations are prepared by placing the drug in association with a liquid vehicle or with a finely divided solid vehicle, or both, and then, if necessary, the product is configured with the desired formulation.

A pharmaceutical composition of the invention should be formulated so as to be compatible with its intended route of administration. Examples of routes of administration include oral or parenteral, for example, intravenous, intradermal, inhalation, transdermal (topical), transmucosal and rectal. The solutions or suspensions used for parenteral, intradermal, or subcutaneous application may include the following components: a sterile diluent such as water for injections, saline solution, fixed oils, polyethylene glycols, glycerin, propylene glycol or other synthetic solvents, antibacterial agents such as benzylparabenos methyl or alcohol; antioxidant agents such as ascorbic acid or sodium bisulfite; chelating agents such as ethylenediaminetetraacetic acid; buffers such as acetates, citrates or phosphates and agents for the adjustment of tonicity such as sodium chloride or dextrose. The pH can be adjusted with acids or bases, such as sodium hydroxide or hydrochloric acid.

Solutions useful for oral or parenteral administration can be prepared by any of the methods well known in the pharmaceutical art, described, for example, in Remingron's Pharmaceutical Sciences (Gennaro, A., ed.), Mack Pub., (1990). Formulations for parenteral administration may also include glycocholate for oral administration, methoxysalicylate for rectal administration, or cutric acid for vaginal administration. Parenteral preparation can be enclosed in ampoules, disposable syringes or multi-dose vials made of glass or plastic. Suppositories for rectal administration can also be prepared by mixing the drug with a non-irritating excipient such as cocoa butter, other glycerides, or other compositions that are solid at room temperature and liquid at body temperatures. The formulations may also include, for example, polyalkylene glycols such as polyethylene glycol, oils of vegetable origin, hydrogenated naphthalenes, and the like. Formulations for direct administration may include glycerol and other high viscosity compositions. Other parenteral carriers potentially useful for these drugs include ethylene-vinyl acetate copolymer particles, osmotic pumps, implantable infusion systems, and liposomes. Formulations for administration by inhalation may contain as excipients, for example, lactose, or they may be aqueous solutions containing, for example, polyoxyethylene-9-lauryl ether, glycocholate and deoxycholate, or oily solutions for administration in the form of nasal drops, or as a gel for intranasal application. Retention enemas may also be used for rectal administration.

The formulations of the present invention suitable for oral administration may be in the form of discrete units such as capsules, gelatin capsules, sachets, tablets, troches, or lozenges, each containing a predetermined amount of the drug; in the form of a powder or granules; in the form of a suspension in solution or an aqueous liquid or a non-aqueous liquid; or in the form of an emulsion or an emulsion of oil in water-or water-in-oil. The drug can also be administered in the form of a bolus, electuary or paste. A tablet can be prepared by compression or molding of the drug, optionally with one or more accessory ingredients. The tablets may be prepared by compressing, in a suitable machine, the drug in a free-flowing form such as a powder or granules, optionally mixed with a binder, lubricant, inert diluent, active surface agent or dispersing agent. The molding tablets can be prepared by molding, in a suitable machine, a mixture of the powdered drug and the suitable carrier moistened with an inert liquid diluent.

Oral compositions generally include an inert diluent or an edible vehicle. For the purpose of oral therapeutic administration, the active compound may be incorporated with excipients. Oral compositions prepared using a fluid vehicle for use as a mouthwash include the compound in the fluid vehicle and are applied orally and are rinsed and expectorated or swallowed. Pharmaceutically compatible binding agents, and / or adjuvant materials may be included as part of the composition. The tablets, pills, capsules, troches and the like may contain any of the following ingredients, or compounds of a similar nature: a binder such as microcrystalline cellulose, gum tragacanth or gelatin; an excipient such as starch or lactose; a disintegrating agent such as algic acid, Primogel, or corn starch; a lubricant such as magnesium stearate or Sterotes; a sliding agent such as colloidal silicon dioxide; a sweetening agent such as sucrose or saccharin; or a flavoring agent such as peppermint, methyl salicylate, or orange flavor.

Pharmaceutical compositions suitable for injections include sterile aqueous solutions (when soluble in water) or sterile dispersions and powders for the extemporaneous preparation of sterile injectable dispersion solutions. For intravenous administration, suitable vehicles include physiological saline solution, bacteriostatic water, Cremophor ELTM (BASF, Parsippany, NJ) or phosphate buffered saline (PBS). In all cases, the composition must be sterile and must be fluid to the extent that there is easy injectability. It should be stable under manufacturing and storage conditions and should be preserved against the contaminating action of microorganisms such as bacteria and fungi. The vehicle may be a solvent or dispersion medium containing, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyethylene glycol, and the like), and suitable mixtures thereof. The proper fluidity can be maintained, for example, by the use of a coating such as lecithin, by the required maintenance of the particles in the case of a dispersion and by the use of surfactants. The prevention of the action of microorganisms can be achieved by various antibacterial and antifungal agents, for example, parabens, chlorobutanol, phenol, ascorbic acid, thimerosal, and the like. In many cases, it will be preferable to include isotonic agents, for example, sugars, polyalcohols such as mannitol, sorbitol, sodium chloride in the composition. Prolonged absorption of the injectable compositions can be caused by including in the composition an agent that delays absorption, for example, aluminum monostearate and gelatin.

Injectable sterile solutions can be prepared by incorporating the active compound in the required amount in an appropriate solvent with one or a combination of ingredients listed above, as required, followed by filtration sterilization. Generally, dispersions are prepared by incorporating the active compound in a sterile vehicle containing a basic dispersion medium and the other required ingredients from those listed above. In the case of sterile powders for the preparation of sterile injectable solutions, the preparation methods include vacuum drying and freeze drying that produces a powder of the active ingredient plus any additional desired ingredient from a sterile solution thereof previously filtered.

Formulations suitable for intra-articular administration may be in the form of a sterile aqueous preparation of the drug that may be in microcrystalline form, for example, in the form of an aqueous microcrystalline suspension. Liposomal formulations or biodegradable polymer systems can also be used to present the drug for both intra-articular and ophthalmic administration.

Formulations suitable for topical administration, including eye treatment, include semi-liquid or liquid preparations, such as liniments, lotions, gels, applications, or emulsions of oil in water or water in oil such as creams, ointments or pastes, solutions or suspensions such as drops. Formulations for topical administration to the surface of the skin can be prepared by dispersing the drug with a dermatologically acceptable vehicle such as a lotion, cream, ointment or soap. They are particularly useful carriers capable of forming a film layer on the skin to locate the application and inhibit its removal. For topical administration to internal tissue surfaces, the agent may be dispersed in a liquid tissue adhesive or to another known substance to improve adsorption to a tissue surface. For example, it is possible to use hydroxypropylcellulose or fibrinogen / thrombin solutions advantageously. Alternatively, tissue coating solutions, such as formulations containing pectin, can be used.

For inhalation, powder inhalation treatments (spray or spray formulations) supplied with an aerosol can, a nebulizer or an atomizer can be used. Such formulations may be in the form of a fine powder for pulmonary administration of a powder inhalation device or of self-propelled powder dispensing formulations. In the case of the self-propeller solution and spray formulations, the effect can be achieved either by choosing a valve that has the desired spray characteristics (i.e., that has the ability to produce a spray having particles of a desired size ) or by incorporating the active ingredient as a suspended powder in the form of particles of a controlled size. For administration by inhalation, the compounds may also be administered in the form of an aerosol spray from a pressurized container or dispenser containing a suitable propellant, for example, a gas such as carbon dioxide, or a nebulizer.

Systemic administration can also be through transmucosal or transdermal means. For transmucosal or transdermal administration, penetrating agents suitable for the barrier to be permeated are used in the formulation. Such penetrating agents are generally known in the art, and include, for example, for transmucosal administration, detergents, bile salts, and derivatives of the phsidic acid. Transmucosal administration can be achieved through the use of nasal sprays or suppositories. For transdermal administration, active compounds are typically formulated in the form of ointments, ointments, gels, or creams as is generally known in the art.

The active compounds can be prepared with vehicles that will protect the compound against rapid removal from the body, such as a controlled release formulation, including implants and microencapsulated delivery systems. Biodegradable, biocompatible polymers, such as ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid can be used. The methods for the preparation of such formulations will be apparent to those skilled in the art. The materials are also commercially provided by Alza Corporation and Nova Pharmaceuticals, Inc. Liposomal suspensions can also be used as pharmaceutically acceptable carriers. They can be prepared according to methods known to those skilled in the art, for example, as described in US Pat. UU.

N. 4,522,811. Microsomes and microparticles can also be used.

The oral or parenteral compositions can be formulated in unit dosage form for ease of administration and dosage uniformity. The expression "unit dosage form" refers to physically discrete units suitable as unit dosages for the subject to be treated; each unit containing a predetermined amount of active compound calculated to produce the desired therapeutic effect in association with the required pharmaceutical vehicle. The specification for the unit dosage forms of the invention are imposed by, and are directly a function of, the unique characteristics of the active compound and the particular therapeutic effect that it is desired to obtain, and of the limitations inherent in the technique of the formation of the active compounds for the treatment of individuals.

As noted above, the drugs identified or designed according to the invention can be formulated in the form of pharmaceutical compositions by mixing with pharmaceutically acceptable non-toxic excipients and vehicles. Such compositions can be prepared for parenteral administration, particularly in the form of liquid solutions or suspensions; for oral administration, particularly in the form of tablets or capsules; or for intranasal route, particularly in the form of powders, nasal drops or aerosols. When adhesion to the surface of a tissue is desired, the composition may include the drug dispersed in a fibrinogen -trombin or other bioadhesive composition. The drug can then be painted, sprayed or otherwise applied to the desired tissue surface. Alternatively, the drugs can be formulated for parenteral administration or oral administration to humans or other mammals for example, in therapeutically effective amounts, for example, amounts that provide appropriate concentrations of the drug to the target tissue for a time sufficient to induce the effect. wanted.

When the active compound is to be used as part of a transplant procedure, it can be provided to the living tissue or organ to be transplanted before removal of tissue or organ from the donor. The drug can be provided to the donor guest. Alternatively, or by way of example, once removed from the donor, the living tissue or organ can be placed in a preservation solution containing the active compound. In all cases, the active compound can be administered directly to the desired tissue, such as by injection into the tissue, or it can be provided systematically, either by oral or parenteral administration, for which any of the methods and formulations described herein and / or known in the art.

When the drug comprises part of a solution for the preservation of an organ or tissue, any commercially available preservation solution can be advantageously used. For example, solutions known in the art include Collins' solution, Wisconsin's solution, Belzer solution, Eurocollins solution and lactated Ringer solution.

The effective concentration of the compounds to be delivered in a therapeutic composition will vary depending on a number of factors, including the desired final dosage of the compound to be administered and the route of administration. It is also likely that the preferred dosage to be administered depends on variables such as the type and extent of the disease or indication to be treated, the general health status of the particular patient, the relative biological efficacy of the delivered compound, the formulation of the drug, the presence and types of excipients in the formulation, and the route of administration. In general terms, the drugs of this invention can be provided to an individual for which typical unit doses deduced from the previously described studies in mammals such as nonhuman primates and rodents are used.

When the active compounds are nucleic acid molecules, the nucleic acid can be inserted into vectors and used as gene therapy vectors. Gene therapy vectors can be administered to a subject by, for example, intravenous injection, local administration (see, U.S. Patent No. 5,328,470) or by stereotactic injection (see, for example, Chen et al. (1994) Proc. Natl. Acad. Sci., USA 91: 3054-3057). The pharmaceutical preparation of the gene therapy vector may include the gene therapy vector in an acceptable diluent, or it may comprise a slow-release matrix into which the vehicle for gene administration is inserted. Alternatively, when the entire gene delivery vector can be produced intact from recombinant cells, for example, retroviral vectors, the pharmaceutical preparation may include one or more cells that produce the gene delivery system.

When an active compound of the invention is intended for administration to a plant host, the invention can be applied directly to the plant environment, for example, to the surface of the leaves, buds, roots or floral parts. Alternatively, the present invention can be used as a seed coating. The determination of an effective amount of the present invention required for a plant in plant is within the skill of the technique and will depend on factors such as plant species, planting method, and soil type. It is considered that the compositions containing drugs of the invention can be prepared by formulating such drugs with adjuvants, diluents, vehicles, etc., in order to provide compositions in the form of filings / solids in the form of divided particles, granules, tablets , wettable powders, powder, aqueous suspensions or dispersions, and emulsions. It is also considered to use such drugs in encapsulated form, for example, the drugs can be encapsulated within polymers, gelatin, lipids or other formulation aids such as emulsifiers, surfactants, antifoaming agents and antifreeze agents, and wetting agents that can be incorporated into such compositions, especially if such compositions were stored for any period of time before use. The application of compositions containing drugs of the invention as active agent can be carried out by conventional techniques. When an active compound is intended for administration to an insect host, standard methods such as without limitation, air dispersion are considered.

The active compound identified or designed by a method of the invention also includes the precursors of the active compounds. The term "precursor" refers to a pharmacologically inactive (or partially inactive) derivative of a genitator molecule that requires biotransformation, be it spontaneous or enzymatic, within the body to release the active compounds. Precursors are variations or derivatives of the compounds of the invention that have groups that are cleavable under metabolic conditions. The precursors become the active compounds of the invention that are pharmaceutically active in vivo, when they undergo solvolysis under physiological conditions or undergo enzymatic degradation. Precursor forms often offer advantages of solubility, tissue compatibility, or delayed release in the mammalian organism (see Bundgard, Design of Prodrugs, pp. 7-9,21-24, Elsevier, Amsterdam (1985); and Silverman, The Organic Chemistry of Drug Design and Drug Action pp. 352-401, Academic Press, San Diego, CA (1992).

The active compound identified or designed by the methods described herein can be administered to individuals to treat disorders (prophylactically or therapeutically). In conjunction with such treatment, the pharmacogenomics (that is, the study of the relationship between the genotype of an individual and the response of said individual to an extraphonic compound or drug) can be taken into account. Differences in the metabolism of therapeutic agents can lead to serious toxicities or a therapeutic failure due to the alteration of the relationship between the dose and the concentration of the pharmacologically active drug in the blood. Therefore, a doctor or clinician may consider the application of the knowledge obtained in the relevant pharmacogenomic studies to decide if a drug is administered, as well as the adaptation of the dose and / or therapeutic regimen of the drug treatment.

With respect to mammals, it is considered that the effective dose of an inducer or inhibitor of protein synthesis will be in the range of about 0.01 to about 50 mg / kg, preferably about 0.1 to about 10 mg / kg of body weight, administered in a single or in multiple doses. Typically, the inducer or inhibitor can be administered to a human recipient in need of a treatment in a daily dose range of about 1 to about 2000 mg per patient.

In the light of the foregoing general statement, the specific examples presented below are illustrative only and are not intended to limit the scope of the invention. Other generic and specific configurations will be apparent to those skilled in the art.

III.

 Examples

TO.

 Example 1: Preparation of 50S ribosomal subunit crystals

H. marismortui was cultured as previously described (Ban et al. (1998) supra) in a slightly modified version of the ATCC 1230 culture medium, which was supplemented with 4.3 g of yeast extract, 5.1 g of tris, and 3.4 g of glucose per liter. Bacteria were grown at 37 ° C to an OD550 nm between 1.0 and 2.2. They were harvested by centrifugation, and stored at - 80 ° C until use. The cells were broken up by a French press. Ribosomes were prepared from lysates by centrifugation, and subunits were isolated in sucrose gradients as previously described (Shevack et al (1985) FEBS Lett 184: 68-71.

The crystals were prepared and stabilized as follows:

1. Reverse extraction

(one)

Mix 1 mg of subunits in a 50S ribosomal subunit preparation (30 mg / ml in 1.2 M KCl, 0.5 M NH4Cl, 20 mM MgCl2, 10 mM TRis, 1 mN CdCl2, 5 mM Tris, pH 7.5 ) with 1/2 volume of 30% 6000 PEG (300 g of PEG, 700 ml of H2O to prepare 1 liter of 30% PEG; filter through a 0.2 mm filter). Leave on ice for 1 to 2 hours.

(2)

centrifuge in an Eppendorf tube for approximately 30 seconds using a desktop centrifuge.

(3)

Remove the supernatant material and add 100 ml of RE- buffer (7% PEG 6000, 1.2 M KCl, 0.5 N NH 4 Cl, 100 mM KAc, 30 mM MgCl 2, 10 mM Tris, 10 mM MES (pH 7.5 ), and 21 mM CdCl).

(4)

Resuspend the pellets at room temperature by mixing with a P200 pipette set in 50 ml. The resuspended material should look a bit cloudy.

 (5)

 Wrap the Eppendorf tube in aluminum foil and leave for equilibration at room temperature for 30

-

 60 minutes. The solution will be saturated with 50S ribosomal subunits.

(6)

Centrifuge the precipitate for 2 minutes in a tabletop centrifuge at room temperature, and transfer the supernatant material to a new Eppendorf tube. A small pellet should be found in the tube used for centrifugation. Store the supernatant at room temperature,

(7)

Place 8-10 ml of supernatant material in the sample cavity of a seated fall tray (Charles-Cena). Spread planting material an hour later taken from a stock of planting material. The stock of planting material is prepared by placing previously cultured crystals in the stabilization buffer solution A (see below), and then subjecting them to an energy vortex. To smear the planting material, a human hair cleaned with water and ethanol and then dried, is passed through the solution that has been subjected to vertices, and then it is touched in the new crystallization drop. The drops should look cloudy. Deposits in the seated trays contain 1,000 ml of a solution containing 8% PEG 6000, 1.2 M KCl, 0.5 M NH4Cl, 100 mM KAc, 6.5 mM HAc (result: pH 5.8 ), 30 mM MgCl2, 1 mM CdCl2.

(8)

After a day, see if the planting was successful and, if so, let the crystals grow for three weeks.

2. Stabilization protocol

When the crystals have finished growing (after approximately 3 weeks), open each sitting chamber by making only a single cut (slit) from the center to the edge of the cavity. Through this narrow slit, add 10 ml of buffer A (1.2M KCl, 0.5M BH4Cl, 30mM MgCl2, 10% PEG 6000, 1mM CdCl2, 100mM KAc, to each drop in each tank. 10 mM Tris (titrated at a final pH of 6.1), 30 mM MONTH) at room temperature and 45 μl of buffer C (0.667 MES, 0.333 M Tris).

Place the trays in a plastic box with a lid, and place in an incubator at 16 C for approximately one day, and then reduce the temperature of the incubator to 12 DC for another day. Next, place the plastic box in a polystyrene container with a lid, and place it in a cold room for another day. The crystals can be stored in this way for a long time, but they must be subject to a more tampon change before use

Carry out the following transition series using buffer A (see above) and buffer B (1.7 M NaCl, 0.5 M NH 4 Cl, 30 mM MgCl 2, 1 mM CdCl 2, 12% PEG 6000, 20% EG , 100 mM KAC (titrated at final pH 5.8 with HAC) so as to obtain final ratios between buffer B and buffer A of: 1/16, 1/8, 1/4, 1/2, 3/4 All solutions should be kept at a cold room temperature All the following manipulations of the drops will be carried out through the narrow slit:

(one)

 add 40 1l of "1/16" to the drop, and leave for 15 minutes.

(2)

 Add 40 1l of "1/8" to the drop, and leave for 30-60 minutes.

(3)

 Remove 40 1l of the drop (and discard it in the tank), add 40 1l of "1/4", and leave for 30-60 minutes.

(4)

 Remove 40 1l of the drop (and discard it in the tank), add 40 1l of "1/2", and leave for 15 minutes.

(5)

 Remove 40 1l of the drop (and discard it in the tank), add 40 1l of "3/4", and leave for 15 minutes.

(6)

 Remove 40 1l of the drop (and discard it in the tank), add 40 1l of buffer B, and leave for 15 minutes.

(7)

Remove 60-80 1l of the drop (and discard it in the tank), add 60-80 1l of buffer B, and replace the deposits with 500 1l of buffer B.

B.

 Example 2: Determination of the crystalline structure of the 50S ribosomal unit, with the initial refinement

All data, except the two native data sets, were obtained in the National Synchrotron Light Source (Brookhaven) from crystals frozen at 100 K, using X12b and X25 beam lines and recorded using a MAR image plate 345 mm For each heavy atom derivative, anomalous diffraction data were collected at the wavelength corresponding to the anomalous peak dispersion. The beam size was 100 x 100 mm for most data collections in X25 and 200 x 200 mm in the X12b light line. The crystals were aligned along the long axis of the cell unit (570 A) so that oscillations of 1.00 could be used to collect reflections up to a maximum of 2.7 A of resolution at the edge of the MAR detector. In the X12b beam of light the crystal to detector distances varied between 450.0 mm and 550.0 mm depending on the wavelength, crystal quality, and beam divergence, and was chosen in such a way that the data of Maximum resolution could be collected while avoiding overlapping points. In the X25 beam of light the detector was positioned on a platform rigid to 480 mm which allowed the collection of data at 3.2 A for iridium and osmium derivatives with the wavelength set at the anomalous edge. The native data was collected at a resolution of 2.4 Angstrom in the ID19 structural biology beam line of the Advanced Photon Source (Argonne), for which a CCD detector was used. Datasets were processed using DENZO and SCALEPACK (Otwinowski, (1993) Data Collection and Processing).

Phasing based on heavy atoms was extended to a resolution of 3.2 Angstrom for which MIR phases calculated for two different isomorphic data groups (MIR1 and MIR2, Table 1) were combined with SAD phases (single derivative anomalous dispersion, dispersion simple derivative anomalies). The two best derivatives were osmium pentamine and iridium hexamine, each of which contained a large number of binding sites (Table 1). Several other derivatives with fewer sites improved map quality even more. The whole phase was carried out using the maximum probability method implemented in CNS (Brunger et al (1998) supra.) With the exception of the derivative Ta6Br12, which was refined in SHARP (de La Fortelle, (1997) Meth. Enzymol. 276 : 472-494) represented as spherically averaged electron density (Table 1). The phases were improved and expanded from 3.3 A to 2.4 A by solvent flipping (Abrahams et al. (1996) supra), and models were constructed from the data set.

C. Example 3: Preparation of crystals of 50S / Puromycin ribosomal subunit complex and collection of X-ray diffraction data

Crystals of 50S ribosomal subunits were cultured and stabilized as described above. CCdA-p-puromycin (see Figure 9A) was a generous gift from Michael Yarus (Welch et al. (1995) supra). The oligonucleotides of amino-N-acylated mini-helices (see Figure 9B) were synthesized by Dharmacon. After deprotection, the oligonucleotides were briefly heated to 100 ° C and rapidly cooled on ice for re-tempering. The crystals of 50S ribosomal subunits were stabilized and then left to soak for 24 hours in stabilization buffer plus 100 mM CCdA-p-puromycin or amino-N-acylated mini-helices before liquid propane cryoprification and ray diffraction data collection - X. The phases were calculated by density modification (CNS) starting with the best experimental phases and using 2Fo (analog) -Fo (native) for the amplitudes, from 60.0 to 3.2 A. (The native amplitudes were of the native data set 1 plus isomorphic, with the exception of those amplitudes that were present only in the most complete native data set 2. The calculated amplitudes 2Fo-Fo that were less than twice the corresponding calculated and were replaced by Fo ( analogical)). The maps were then calculated using modified density phases and amplitudes 2Fo (analog) -Fo (native).

D. Example 4: Antibiotic binding sites located in the polypeptide exit tunnel near the center of peptidyltransferase

The electron density maps derived from crystalline complexes of the large subunit of H. marismortui formed complexes with the three antibiotics tylosin, carbomycin A and anisomycin with a resolution of approximately 3.0 A. All these antibiotics bind to the ribosomes in the region between the peptidyltransferase center as defined by the Yarus inhibitor, CCdA-puromycin and the tips of the L22 and L4 proteins at the point that they form a small hole at the exit of the polypeptide exit tunnel. The general location of this important antibiotic binding site is shown in Figure 19. Tylosin and carbomycin A appear to work by blocking the outflow of recently synthesized polypeptides. Anisomycin appears to block the An site.

The vast majority of the interactions between these antibiotics and the ribosome are through the rRNA that defines the An site, and the surface of the tunnel between the peptidyltransferase center and the L22 protein. Since these antibiotics do not bind identically, there will be many additional ways in which small molecule compounds can be designed to bind in this region, by using the tools and methodologies of the present invention. For example, by connecting each of the components of each of the different antibiotics that bind to non-overlapping sites, it will be possible to create new hybrid antibiotics (see Example 6). In addition, based on new principles of small molecule RNA interaction shown by these antibiotic complexes, it is possible to disseminate new, completely novel molecules that bind to the same sites in the ribosome, as well as other potential targeted RNAs.

E. Example 5: Design and test of hybrid antibiotics

Many antibiotics that target ribosomes, more particularly large ribosomal subunits, and destroy protein synthesis are complex molecules that are effectively concatenations of simpler substructures, of which at least it interacts with a discrete part of the ribosome. When the compound in question includes several interactive substructures, its binding site is effectively the sum of the subsites that come into contact and are coupled with each of said substructures. It has been found that many of the antibiotics that target the large ribosomal subunit bind to the ribosomal subunit at sites that are close to each other. Therefore there is the possibility of synthesizing new antibiotics in which a part that binds to ribosome, of a first known antibiotic, is chemically linked to a part that joins the ribosome, of a second known antibiotic that interacts with an adjacent subsite . The resulting new compound is therefore a chimera or a hybrid of the two antibiotics from which it is derived.

Chimeric antibiotics can be designed using information about the structures of the antibiotic / ribosome complexes discussed above herein. These structures allow the identification of the binding subsites of antibiotics in the ribosome, and the specification of the chemical entities that interact with them. Equipped with this knowledge, experts in the technique of organic synthesis can synthesize compounds that link the substructures of interest to each other in ways that will allow them to interact with their respective subsites at the same time. It is likely that any compound thus conceived to function in the intended manner and in the manner indicated inhibits cell growth and if it does, the synthesis of proteins in vivo. At the very least, it should block protein synthesis in in vitro test systems. More information about the ribosomal interactions of such a compound can be obtained by determining the structure of the complex that forms with the ribosome, thanks to the methods described in Chapter D above herein.

For example, as a result of the work described herein, it has been found that the discarded portion of carbomycin binds to the large ribosomal subunit at a site in close proximity to the junction site of an anisomycin portion. Thanks to the use of this information and the software packages described above, the expert can design a hybrid antibiotic comprising the relevant ribosome binding portions of carbomycin and anisomycin linked by means of a suitable chemical linker.

Figure 34 shows the design of hybrid antibiotics given by way of example. In this Figure, a portion of the sparsomycin antibiotic is attached to a portion of the chloramphenicol antibiotic so as to produce a sparsocloranfenicol hybrid antibiotic. In a first antibiotic sparsocloranfenicol hybrid, hybrid A, n = 1 in the union region (1). In a second antibiotic sparsocloranfenicol hybrid, hybrid B, n = 2 in the junction region (2). The portion of the chloramphenicol molecule was chosen as a result of the structural studies described in Schlunzen et al. (2001) Nature 413: 814-821. Again, the hybrid antibiotic was designed to allow each hybrid antibiotic component to bind to its respective binding site in the large ribosomal subunit.

Figure 35 shows the design of another hybrid antibiotic given by way of example. In this figure, a portion of the antibiotic sparsomycin is attached to a portion of the antibiotic anisomycin so as to produce a sparsoanisomycin hybrid antibiotic. The portions of each antibiotic were chosen as a result of the structural studies of the present that showed how each of the antibiotics binds to the large ribosomal subunit. The hybrid antibiotic was designed to allow each component of the hybrid antibiotic to simultaneously join its respective binding site in the large ribosomal subunit.

These hybrid molecules, once designed, can be synthesized and purified by means of conventional organic chemistry procedures and conventional purification schemes. Once synthesized and purified, the hybrid molecule of interest can be systematically selected to establish its bioactivity and determine, for example, the value of the IC50 for each molecule of interest. These systematic selections may include, for example, the cultivation of microorganisms on or within culture media, is supplemented with the hybrid molecule or lacking it. Any reduction in the number of microorganisms or in the magnitude of the colonies in the presence of the hybrid molecule would be indicative of their bioactivity. In addition, the hybrid molecule could be tested in a cell-free transcription / translation coupled test or in a translation system in the presence of one or more labeled amino acids or using a non-radioactive indicator system. Any reduction in the level of labeled amino acids incorporated in proteins in cell-free systems that include the hybrid molecule in relation to cell-free systems lacking the hybrid molecule, would be indicative that the hybrid molecule acts as a functional inhibitor. of protein synthesis. It is considered that the hybrid molecule could then be iteratively refined as set forth hereinbefore to improve its bioactivity and bioavailability.

F. Example 6: Synthesis of esparsocloranfenicol hybrids

The synthesis of the sparsocloranfenicol hybrids (1 and 2) is shown in Figures 36 to 39, and is described in stages as follows.

i. Acid Synthesis 6

The acid 6 in Figure 36 was obtained by the procedures described in Ottenheijm et al., J. Org. Chem. 1981, 46, 3273-3283, starting with compound 4 and passing through the intermediate compound aldehyde 5.

5. ii. Amine Synthesis 7

The amine 7 in Figure 36 was synthesized as shown in Figure 37. A few amounts of L-cysteine (1.00 g, 5.41 mmol), of dimethoxytrityl chloride (DMT) (2.12 g, 5, 95 mmol), and triethylamine (1.20 ml, 8.56 mmol) were combined in 25 ml of 80% aqueous acetic acid and allowed to react at room temperature for 2 hours. The solvents were evaporated in vacuo, and the residue was taken up in 60 ml of ethyl acetate (EtOAc). The solution was washed with saturated aqueous sodium bicarbonate (NaHCO3) (3 x 50 ml), and brine (2 x 50 ml). The organic layer was dried with sodium sulfate (Na2SO4), and evaporated so as to provide the DMT thioether of L-cysteine 10 (2.13 g, 5.03 mmol, 93%) with a purity suitable for use In the next reaction.

Protected amino acid 10 (5.41 mmol) was dissolved in 20 ml of anhydrous tetrahydrofuran (THF), and sodium borhydride (0.70 g, 18.2 mmol) was added portionwise. Boron etherate trifluoride (3.03 ml, 23.9 mmol) was added dropwise, and the mixture was allowed to stir overnight at room temperature. The excess boron trifluoride etherate was destroyed with ethanol, and the mixture was filtered. The filtrate was evaporated, and the residue was dissolved chloroform (60 ml), and washed with saturated aqueous NaHCO3 (40 ml). The aqueous layer was extracted with chloroform (20 ml), and the combined organic phase was washed with brine and dried over Na2SO4. The solvent was evaporated in order to provide expected amino alcohol amino 11 (1.75 g, 85%) as a white solid n of purity suitable for use in the next step.

An amount of amino alcohol 11 (1.70 g, 4.15 mmol) was dissolved in 5 ml of acetonitrile and 5 ml of methylene chloride, and the solution was cooled to 0 ° C. Tert-butyldimethylsilyl chloride (1 M, 5.00 ml, 5.00 mmol) was added, followed by 1.8-diazabicyclo �5.4.0� undec-7-ene (DBU) (0.67 ml, 4, 40 mmol), and the mixture was allowed to stir overnight at room temperature. The reaction was quenched by stirring with saturated aqueous NaHCO3 for 20 minutes. Additional methylene chloride was added and the layers were separated. The organic layer was washed with NaHCO3, water and then with brine. The organic phase was dried (Na2SO4) and evaporated. The residue was chromatographed on silica gel using 0.2% methanol / chloroform to obtain amine 7 (0.100 g, 5%), 1 H-NMR (500 MHz, CDCl 3) � 7.41 (d, J = 2 Hz, 2H), 7.41-7.19 (m, 8H), 6.82-6.79 (m, 4H), 3.79 (s, 6H), 3,443.41 (m, 1H ), 3.35-3.32 (m, 1H), 2.69-2.67 (m, 1H), 2.36-2.33 (m, 1H), 2, 9-215 (m, 1H ), 0.85 (s, 9H), 0.05-0.00 (m, 6H).

iii. Synthesis of Amide 8

Amide 8 in Figure 36 was synthesized as follows. Amounts of acid 6 (0.044 g, 0.221 mmol), amine 7 (0.070 g, 0.134 mmol), and 1-hydroxybenzotriazole (HOBt) (0.030 g, 0.21 mmol) were dissolved in 2 ml of dimethyl formamide (DMF) . 1,3-Dicyclohexylcarbodiimide (DCC) (0.056 g, 0.272 mmol) was added, and the mixture was stirred for 24 hours at room temperature. The mixture was evaporated and chromatographed on silica gel using 0-4% methanol / chloroform to provide amide 8 (0.070 grams, 75%), 1 H-NMR (500 MHz, CDCl 3) � 7.50-7 , 15 (m, 12H), 6.80 (d, J = 2 Hz, 4H), 5.91 (broad d, J = 5 Hz, 1H), 4.20-4.11 (m, 1H), 3.74 ( s, 6H), 3.66-3.63 (m, 1H), 3.52-3.49 (m, 1H), 2.49-2.44 (m, 2H), 2.36 (s, 3H), 0.84 (s, 9H), 0.01 (s, 3H), -0.01 (s, 3H).

iv. Synthesis of thiol 9

The thiol9 in Figure 36 was synthesized as follows. An amount of amide 8 (0.060 g, 0.086 mmol) was dissolved in THF at 0 ° C. Boron trifluoride etherate (0.30 ml, 2.37 mmol) was added dropwise, and stirring was continued for 20 minutes. The reaction mixture was quenched with ethanol (5 ml, stirring for 5 min), and the solvents evaporated. The residue was triturated with chloroform, and the solids collected by centrifugation were washed with chloroform. The solids were dried to provide thiol 9 (0.015 g, 63%), 1H-NMR (500 MHz, CDCl3) � 7.31 (d, J = 9 Hz, 1H), 7.11 (d, J = 9 Hz, 1H), 3.93-3.90 (m, 1H), 3.59-3.50 (m, 3H), 3.22 to 3.20 (m, 1H), 2.68-2.64 (m, 1H), 2.56-2.53 (m, 1H), 2.23 (s, 3H).

v. Synthesis of Bromoacetamide 13

Bromoacetamide13 in Figure 37 was synthesized as follows. The free base of (1R, 2R) -chloramphenicol 12 was converted into the bis t-butyldimethylsilylether derivative as described in Orsini et al., Organic Preparations and Procedures International 1989, 21, 505-508. The bis-silyl ether resulting from (1R, 2R) -chloramphenicol (0.56 g, 1.30 mmol), bromoacetic acid (0.146 g, 1.00 mmol and HOBt (0.163 g, 1.20 mmol) dissolved in 6 ml of THF DCC (0.330 g 1.6 mmol) was added, and the mixture was stirred overnight at room temperature The precipitate was filtered, and the filtrate was evaporated.The residue was taken up in 60 ml of EtOAc and extracted with 5% aqueous NaHCO3 (30 ml) and brine (2 x 40 ml) The organic layer was dried over Na2SO4 and evaporated.The residue was chromatographed on silica using 01:07 EtOAc / hexane at 1: 5 EtOAc / hexane as eluents so as to provide bromoacetamide 13 (0.480 g, 86%) as an oil, 1H-RMN (500 MHz, CDCl3) � 8.32 (d, J = 5 Hz, 2H), 7.62 (d, J = 5 Hz, 2H), 7.21 (broad d, J = 5 Hz, 1H), 5.36 (s, 1H), 4.11 to 4.6 (m, 1H), 4.01-3.85 (m, 4H), 3.75-3.68 (m, 2H), 3.39-3.31 (m, 1H), 1.10 (s, 9H), 1, 05 (s, 9H), 0.24 (s, 3H), 0.22 (s, 3H), 0.21 (s, 3H), 0.00 (s, 3H).

saw. Sulfide Synthesis 14

Sulfide 14 in Figure 38 was synthesized as follows. An amount of thiol 9 0.090 g, 0.315 mmol) was suspended in 3 ml of THF / 0.5 ml of DMF and DBU (49 ml, 0.315 mmol). A solution of bromoacetamide 13 (0.16 g, 0.284 mmol) in 2 ml of THF was added at 0 ° C, and the mixture was heated to room temperature and then stirred at room temperature for 12 hours. The reaction was poured into saturated brine (20 ml) and chloroform (CHCl3) (40 ml), and the layers were separated. The organic layer was washed with saturated brine (2 x 20 ml) and dried over Na2SO4. The solvents were evaporated, and the residue was chromatographed on silica (pipette column) using an elution gradient from 1:20 methanol / chloroform to 1:10 methanol / chloroform to produce sulfide 14 (0.093 g, 43%) as a white solid 1H-RMN (500 MHz, 01:01 chloroform-d / methanol-d4 (CDCl3 / CD3OD)) � 8.15 (d, J = 6 Hz, 2H), 7.55 (d, J = 5 Hz, 1H), 7.53 (s, 1H), 7.47 (d, J = 6 Hz, 2H), 7.37 (d, J = 9 Hz, 1H), 7.14 (d, J = 9 Hz, 1H), 5.20 (wide s, 1H), 4.09-4.02 (m, 1H), 3.98-3.92 (m, 1H), 3.70-3.53 ( m, 5H), 3.33-3.12 (m, 4H), 2.67-2.61 (m, 2H), 2.31 (s, 3H), 0.91 (s, 9H), 0 , 90 (s, 9H), 0.08 (s, 3H), 0.06 (s, 3H), -0.06 (s, 3H), -0.14 (s, 3H).

vii. Sulphoxide Synthesis 16

Sulfoxide 16 in Figure 38 was synthesized as follows. An amount of sulfide 14 (0.060 g, 0.078 mmol) was dissolved in 2 ml of THF, and the mixture was cooled to 0 0 C. Diisopropylethylamine (i-Pr2NEt) (18 1, 0.10 mmol) was added, followed by chloromethyl methyl ether (33.2 ml, 0.39 mmol), and stirring was continued for 1 hour at 0 ° C. The cooling bath was removed and the reaction was left under stirring for 12 hours at room temperature. The reaction was quenched with saturated aqueous NaHCO3, and partitioned with chloroform (40 ml). The aqueous layer was extracted with chloroform (20 ml), and the combined organic phase was dried over Na2SO4, and evaporated. The crude material was chromatographed on silica using 0-4% methanol / chloroform as eluent. Two fractions were obtained. Fraction 1 (10.8 mg) contained a minor impurity with an Rf = 0.63 (12% methanol / chloroform) and a major component with an Rf = 0.58. Fraction 2 (16 mg) contained only one compound with Rf = 0.50. (The starting alcohol had an Rf = 0.36). These fractions appeared to contain compounds that were completely protected in the hydroxyl group of interest, and partially overprotected (MOM ethers ethers in the uracil portion of the molecules).

The material from fraction 2 above (0.016 g, 0.020 mmol) was dissolved in 1 ml of carbon tetrachloride (CCl4). To this solution was added a mixture containing titanium (IV) isopropoxide (Ti (Oi-Pr) 4) (0.6 ml, 0.002 mmol), BINAP (0.001 g, 0.004 mmol, water (0.72 1l, 0, 02 mmol) in 1 ml CCl4), and the resulting mixture was stirred at 20 ° C (see Komatsu et al., J. Org. Chem. 1993, 58, 4529-4533 for a detailed description of this oxidation protocol). tert-butyl hydroperoxide added in toluene (1 M, 60 ml, 0.06 mmol) and stirring was continued for 72 hours. At this time, the TLC analysis indicated the conversion of � 75% to a new product of lower Rf (Rf = 0.39, CHCl3 / MeOH 8: 1). The reaction was partitioned between saturated NaHCO3 (10 ml and CHCl3 (20 ml). The aqueous layer was washed with CHCl3 (2 x 10 ml) and the combined organic phase was dried over Na2SO4 and the solvent was evaporated. The crude material was evaporated. chromatographed on silica gel eluting with 0-5% methanol (MeOH) in CHCl 3. Two fractions were obtained: fraction 1 contained unreacted starting material (4.1 mg), while fraction 2 contained sulfoxide 16 (7.7 mg, 0.036 mmol).

viii Synthesis of the sparsocloranfenicol A1 hybrid

The sparsocloramaphenicol A hybrid in Figure 38 (complete structure in Figure 34) was synthesized as follows. An amount of sulfoxide 16 (0.0075 g 0.0091 mmol) was dissolved in 0.1% hydrochloric acid (HCl) in MeOH (2 mL) and Dowex � 50 (HB form (0.2 g) was added. The mixture was subjected at stirring at 50 ° C for 3 hours and the TLC analysis showed a complete conversion to a product with a unique lower value of Rf (Rf = 0.25, CHCl3 / MeOH 25: 3) .The Dowex � 50 was removed by filtration and the solvent was evaporated in vacuo.

The crude product was dissolved in THF (0.5 ml) and MeOH (0.5 ml), a solution of H2SiF6 (20 wt / vol in H2O) (120 ml, 0.15 mmol) was added and the mixture was stir for 3 hours at room temperature. The TLC analysis showed a conversion of approximately 50% in a baseline product within 3 hours, therefore stirring continued overnight (� 20 hours) during which almost a quantitative formation of the product was observed from baseline. The reaction was partitioned between CHCl3 (1 ml) and water (H2O) (1.5 ml), the layers were separated and the aqueous layer was extracted once with CHCl3 / MeOH 2: 1 (1.5 ml). TLC analysis (CHCl3 / MeOH 7: 1) showed that the reference product was exclusively in the aqueous layer, while TLC (CHCl3 / MeOH / H2O 2: 1: 0.1) showed the presence of compounds closely related (Rfs 0.71 and 0.67). The aqueous layer was concentrated and chromatographed on silica (pipette column) using CHCl3 / MeOH / H2O 2: 1: 0.1 so as to obtain a diastereomeric mixture of 1 (0.0017 g, 34%) as a white solid, 1H-NMR, partial (500 MHz, deuterium oxide (D2O) / CD3OD) � 8.15 (d ap, J = 8.5 Hz, 2H), 7.57 (d ap, J = 8, 5 Hz, 2H), 17.42 (d ap, J = 15.5 Hz, majority diastereoisomer), 7.39 (app d, J = 15.5 Hz, minor diastereoisomer); 1H�, �6.99 (app d, J = 16 Hz, minor diastereoisomer), 6.96 (app d, J = 15.5 Hz, main diastereoisomer); 1H�, �2.53 (s, majority diastereoisomer), 2.51 (s, minor diastereoisomer); 3H). HRMS calculated for C22H27N5 O10S (M, monodeuterized) B: Found, 555.15.

ix. Sulfide Synthesis 15

Sulfide 15 of Figure 38 was synthesized as follows. An amount of sulfide 14 (0.02 g, 0.026 mmol) was dissolved in THF (0.3 ml) and MeOH (0.3 ml), a solution of H2SiF6 (20 wt / vol in H2O) (200 ml, 0.25 mmol) and the mixture was stirred for 20 hours at 20 ° C. The reaction was partitioned between CHCl3 (1 ml) and H2O (0.5 ml), and the layers were separated. The aqueous layer was concentrated and chromatographed on silica (pipette column) using CHCl3 / MeOH / H2O 2.5: 1: 0.1 so as to provide compound 15 (0.0097 g, 70%) as a white solid, 1 H-NMR (500 MHz, D2O / CD3OD) � 8.21 (d, J = 9 Hz, 2H), 7.67 (d, J = 9 Hz, 2H), 7.44 (d, J = 15.5 Hz, 1H), 7 , 15 (d, J = 15.5 Hz, 1H), 5.17 (d, J = 3Hz, 1H), 4.27 (m, 1H), 4.12 (m, 1H), 3.85 ( m, 2H), 3.65 (m, 3H), 3.20-3.39 (m, 4H), 2.60 (m, 2H), 2.43 (s, 3H). HRMS calcd for C22H27N5O9S (M B H) B: 538.16. Found: 538.12

x. Bromopropionamide Synthesis 17

Bromopropionamide 17 of Figure 39 was prepared with the same protocol described for the synthesis of bromoacetamide 13, except that 3-bromopropionic acid was used instead of bromoacetic acid. Bromopropionamide 17 was purified on a silica gel column using 01:08 EtOAc / hexane to EtOAc / hexane 1: 4 as eluents so as to provide bromopropionamide 17 (45%) as an oily foam, 1 H-NMR (500 MHz, CDCl3) (contaminated with � 30% acrylamide impurity) � 8.18 (d apparatus, J = 9 Hz, 2H), 7.47 (d ap, J = 9, 2H), 5.24 (d ap , J = 4.5 Hz, 1H), 4.04 (m, 1H), 3.52-3.67 (m, 4H), 2.69 (m, 2H), 0.95 (s, 9H) , 0.92 (s, 9H), 0.08 (s, 3H), 0.06 (s, 3H), -0.05 (s, 3H), - 0.10 (s, 3H).

xi. Sulfide Synthesis 18

Sulfide 18 of Figure 39 was prepared from thiol 9 and bromopropionamide 17 using the protocol described for the synthesis of sulfide 14. The resulting product was purified on a silica gel column eluting with 10: 1 CHCl3 / MeOH at 08 : 01 CHCl3 / MeOH to provide sulfide 18 (7%) as a white solid.

xii. Synthesis of sparsocloranfenicol B 2 hybrid

The sparsocloranfenicol hybrid B 2 in Figure 39 (complete structure in Figure 34) was synthesized as follows. An amount of sulfide 18 (0.02 g, 0.026 mmol) was dissolved in CCl4 (1.28 ml) containing Ti (Oi-Pr) 4 (0.77 ml, 0.0026 mmol), BINAP (0.0015 g, 0.0052 mmol), water (0.92 ml, 0.026 mmol), and the solution was stirred at 20 ° C. Tert-butyl hydroperoxide in toluene (1 M, 71.3 ml, 0.07 mmol) was added and stirring was continued at 20 ° C for 72 hours. The reaction was partitioned between saturated NaHCO3 (10 ml) and CHCl3 (20 ml). The aqueous layer was washed with CHCl3 (2x10 ml), the combined organic layer was dried over Na2SO4 and the solvent was evaporated. The crude material was chromatographed on silica eluting with 0-10% MeOH in CHCl3. Two fractions were obtained: fraction 1 contained unreacted sulfide 18 (0.0055 g) and fraction 2 contained the desired sulfoxide (0.008 g, 38%).

The sulfoxide product in fraction 2 above (0.008 g 0.01 mmol) was dissolved in THF (0.5 ml) and MeOH (0.5 ml), a solution of H2SiF6 (20 w / v in water) was added (120 ml, 0.15 mmol) and the mixture was stirred at room temperature overnight (� 20 hours) during which an almost quantitative formation of a baseline product was observed. The reaction was partitioned between CHCl3 (1 ml) and H2O (1.5 ml), the layers were separated and the aqueous layer was extracted once with CHCl3 / MeOH 2: 1 (1.5 ml). TLC analysis (CHCl3 / MeOH 7: 1) showed that the reference product was exclusively in the aqueous layer, while TLC based on CHCl3 / MeOH / H2O 2: 1: 0.1 showed the presence of compounds closely associated (RFS 0.52 and 0.47). The aqueous layer was concentrated and chromatographed on silica (pipette column) using CHCl3 / MeOH / H2O 2: 1: 0.1 resulting in a diastereomeric mixture of 2 (0.0036 g, 69%) as a white solid) , 1H-NMR, partial (500 MHz, D2O / CD3OD) � �8.20 (d app, J = 8.8 Hz), 8.13 (d device, J = 8.8 Hz); 2H�, �7.58 (d apparatus, J = 9 Hz), 7.54 (d ap, J = 9 Hz); 2H), �7.34 (d app, J = 15 Hz), 7.24 (d ap, J = 15 Hz); 1H�, �7.01 (d app, J = 15 Hz), 7.00 (d app d, J = 15 Hz); 1H�, 2.76 (m, 2H), 2.34 (wide s, 3H). HRMS calculated for C23H29N5O10S M B H) B: 568.17. Found: 568.14.

Example 7: synthesis of the sparsoanisomycin hybrid

The synthesis of the sparsoanisomycin 3 hybrid is shown in Figures 40 and 41, and is described in stages as follows:

i. Phenol Synthesis 21

Phenol 21 of Figure 40 was synthesized as follows. An amount of anisomycin 19 (0.5 g 1.88 mmol) was suspended in 15 ml of acetonitrile. An amount of 9-fluorenylmethyl succinimide (0.762 g 2.26 mmol) was added to the suspension. After the addition, all the solid dissolved. After stirring for 16 hours at room temperature, water was added to the reaction solution, and extracted with ethyl acetate. The organic compounds were combined and dried over sodium sulfate. After filtration and concentration, purification was performed by column chromatography using 3: 1 hexanes: ethyl acetate as the mobile phase. This allowed to obtain 9fluorenylmethyl carbamate 20 in the form of foam with quantitative yields (0.916 g).

Compound 20 (0.929 g 1.90 mmol) was dissolved in 1.5 ml of dried 1-methyl-2-pyrrolidinone. Imidazole (0.776 g 11.4 mmol) was added and then tert-butyldimethylsilyl chloride (0.859 g 5.70 mmol) to the solution. After stirring at room temperature for 1.5 hours, the reaction solution was applied directly to a silica column. The column was eluted with a 6: 1 solution of hexanes: ethyl acetate resulting in the desired silylated product (1.04 g, 91%) as a white foam.

The above silylated product (1.04 g 1.73 mmol) was dissolved in 15 ml of dry methylene chloride. The solution was cooled to -10 0C. A 1.0 M solution of boron tribromide (17.3 ml) was added. After 2 hours, 4 ml of the boron tribromide solution was added. After an additional 2 hours at -10 ° C, the reaction solution was poured into an aqueous solution at 0 ° C saturated with sodium bicarbonate. The mixture was stirred vigorously for 30 minutes. Then, it was separated and extracted with methylene chloride. The organic layers were combined and dried over sodium sulfate. Purification was performed using column chromatography eluting with 4: 1 hexanes / ethyl acetate and then 2: 1 hexanes / ethyl acetate after the starting material was outside the phenol column so as to provide 21 ( 0.748 g 74%) in the form of a white foam, 1H-NMR (300 MHz, CDCl3) � 7.80 (d, J = 7.4 Hz, 1H), 7.74 (d, J = 7.4 Hz , 1H), 7.63 (m, 2H), 7.36 (m, 4 H), 7.06 (d, J = 8.0 Hz, 1H), 6.79 (d, J = 8.3 Hz, 1H), 6.74 (s, 2H), 4.84 (m, 1H), 4.69 (d, J = 4.9 Hz, 1H), 4.50 (m, 2H), 4, 34 (m, 2H), 4.02 (m, 1H), 3.90 (m, 1H), 3.35 (m, 3H), 2.84 (m, 1H), 2.51 (m, 1H ), 2.01 (s, 3H), 0.87 and 0.83 (two s, 9H), 0.06 and -0.01 (two s, 6H).

ii. Synthesis of methylthiomethylphenyl ether 23

Methylthiomethylphenyl ether 23 of Figure 40 was prepared as follows. An amount of phenol 21 (0.748 mg 1.27 mmol) was dissolved in 10 ml of THF and 0.2 ml of piperidine was added. After stirring at room temperature for 2 hours, water was added and the mixture was extracted with ethyl acetate. The organic layers were combined and dried over sodium sulfate and filtered. The filtrate was concentrated and dried in a vacuum pump for several hours. The crude yellowish foam was dissolved in 10 ml of dry THF and triethylamine (0.35 ml, 2.54 mmol) was added. After stirring for a few minutes, di-tert-butylcarbonate (1.11 g 5.08 mmol) was added. The reaction was stirred at room temperature for 20 hours. Water was added and the mixture was extracted with ethyl acetate and the organic layers were combined and dried over sodium sulfate. Purification was performed by column chromatography using 4: 1 hexanes / ethyl acetate. The desired compound protected by N-BOC 22 was isolated in the form of a white foam (0.348 g 59%, two steps).

The compound 22 protected by N-BOC (0.202 g 0.434 mmol) was dissolved in 2.0 ml of dry THF under an argon atmosphere and cooled to 0 ° C. Sodium hydride (0.018 g of a 60% dispersion, 0.46 mmol was added and the reaction continued under stirring at 0 ° C for fifteen minutes. Sodium iodide (0.071 g 0.48 mmol), hexamethylphosphoramide (0.45 ml, 2.6 mmol), and chloromethylthiomethyl ether (36 ml, 0.48 mmol), in this order After stirring for five minutes, the ice bath was removed and the reaction was heated to room temperature. For another 45 minutes, water was added to the reaction mixture and extracted with ethyl acetate.The organic compounds were combined and dried over sodium sulfate Purification was performed using column chromatography eluting with 6: 1 hexanes / ethyl acetate in order to provide methylthiomethylphenyl ether 23 (0.203 g, 89%) as an oil, 1 H-NMR (300 MHz, CDCl 3) � 7.07 (m, 2H), 6.86 (d, J = 8.5 Hz, 2H), 5.11 (s, 2H), 4.84 (m, 1H), 4.33 (m, 1H), 3.91 (m, 1H), 3.35 (m, 3H), 2.83 (m, 1H), 2.24 (s, 3H), 2.06 (s, 3H), 1.48 (s, 9H), 0.81 (s, 9H), -0.02 (s, 6H).

iii. Synthesis of Chloromethylphenyl Ether24

Chloromethylphenyl ether 24 of Figure 40 was synthesized as follows. An amount of methylthiomethylphenyl ether 23 (0.032 g 0.061 mmol) was dissolved in 1 ml of dry methylene chloride containing 0.050 g of molecular sieves of 3 1m and then cooled to 0 ° C. Diisopropylethylamine (15 ml, 0.085 mmol) and sulfuryl chloride (6.5 ml, 0.079 mmol) were added to the reaction mixture. The reaction mixture was then stirred for two minutes and cyclohexene (12 ml, 0.12 mmol) was added. Five minutes after the addition of cyclohexene, the reaction mixture was allowed to warm to room temperature and stirred for another twenty minutes. Then, the reaction was quenched with water, extracted with methylene chloride, and dried over sodium sulfate. Purification was carried out using column chromatography eluting with 6: 1 hexanes: ethyl acetate so as to obtain 24-chloromethylphenyl ether (0.021 g 67%) as an oil, 1 H-NMR (300 MHz, CDCl 3) � 7 , 26 (m, 2H), 7.00 (d, J = 8.5 Hz, 2H), 5.87 (s, 2H), 4.84 (m, 1H), 4.35 (m, 1H) , 3.92 (m, 1H), 3.33 (m, 3H), 2.85 (m, 1H), 2.06 (s, 3H), 1.48 (s, 9H), 0.81 ( s, 9H), -0.01 (s, 6H).

iv. Sulphoxide Synthesis 25

The sulfoxide 25 of Figure 41 was synthesized as follows. An amount of thiol 9 (0.022 g 0.078 mmol) was dissolved in THF (0.5 ml), DMF (0.2 ml) and DBU (10.3 ml, 0.066 mmol). A THF solution (0.5 ml) of a 3: 2 mixture of methylthiomethylphenyl 23 ether and chloromethylphenyl 24 ether (0.03 grams 0.023 mmol, based on 24) was added and the resulting mixture was stirred at room temperature overnight . The reaction was poured into saturated brine (10 ml) and CHCl3 (20 ml) and the two layers were separated. The organic layer was washed with saturated brine and dried over Na2SO4. TLC analysis (CHCl3 / MeOH 7: 1) showed a quantitative removal of unreacted thiol 9. The solvent was evaporated, and 1 H-NMR of the residue in CD3OD / CDCl3 revealed a complete disappearance of the methylene chloromethylphenoxy peak at 5, 9 ppm

The above crude material was dissolved in CCl4 (3.5 ml) containing Ti (Oi-Pr) 4 (1.81 ml, 0.006 mmol), BINAP (0.0035 g 0.012 mmol), water (2.17 ml, 0.06 mmol) and the solution was stirred at 20 ° C. Terbutyl hydroperoxide in toluene (1 M, 200 ml, 0.2 mmol) was added and stirring was continued at 20 ° C for 72 hours. The reaction was partitioned between saturated NaHCO3 (10 ml) and CHCl3 (20 ml). The aqueous layer was washed with CHCl3 (2x10 ml), the combined organic layers were dried over Na2SO4 and the solvent was evaporated. The crude product was chromatographed on silica eluting with 0-7% MeOH in CHCl3 so as to obtain sulfoxide 25 (0.0054 g, 30% in two stages) as a yellow-white solid.

v. Synthesis of sparsoanisomycin 3

Sparsoanisomycin 3 of Figure 41 was synthesized as follows. An amount of sulfoxide 25 (0.0035 g, 0.0045 mmol) was dissolved in anhydrous methylene chloride (CH2Cl2) (0.25 ml) at 0 ° C. To this solution was added trifluoroacetic acid (TFA) (0.25 ml) and the mixture was stirred at room temperature for 30 minutes during which a quantitative consumption of 25 was observed. The solvent was evaporated and a brown residue was obtained.

The above brown residue was dissolved in anhydrous THF (0.4 ml) and pyridine (0.2 ml) in a Falcon tube, and stirring was maintained at 0 ° C. Pyridine Hydrochloride (HF�Pyr) (0.1 ml) was added to the reaction, the mixture was heated to 20 ° C and stirred overnight. The reaction was quenched by the addition of aqueous triethylammonium bicarbonate (1 M, 50 ml) and chromatographed on silica (pipette column), first eluting with CHCl3 / MeOH 03: 01-02: 01, then with CHCl3 / MeOH / H2O at 2: 1: 0.1 and sparsoanisomycin 3 (0.0012 g 47%) was obtained as a white solid, 1 H-NMR, partial (500 MHz, D2O / CD3OD) �7.36 (d ap , J = 15.5 Hz), 7.35 (d ap, J = 15.5 Hz); 1H�, 7.25 (d app. J = 8.6 Hz, 2H), �7.10 (app d, J = 8.6 Hz), 7.09 (d device. J = 8.6 Hz) ; 2H�, �7.04 (d device, J = 15 Hz), 7.02 (d app, J = 15 Hz); 1H�, �2.36 (s), 2.35 (s); 3H�, �2.18, (s), 2.17 (s); 3H�. HRMS calculated for C25H32N4O9S (M B H) B: 565.19. Found: 565.16.

G. Example 8: assays to establish translation inhibitory activity

The translation inhibitory activity of the hybrid antibiotics produced in Examples 6 and 7 were tested in extracts of E. coli 30 S and in rabbit reticulocyte lysate assays, as follows:

An variation of the protocol of Promega Technical Bulletin N D 092 was used to evaluate the translation inhibition of an E. coli D30 extract system for circular DNA (part number of Promega L1010). 5ml of S30 extract was incubated with 0.4 mg of BestLuc plasmid DNA, 16 units of ribonuclease inhibitor (Promega part number: N2115), 1 mM nucleotide triphosphates (NTPs), 10 mM MgCl2, 40 mM Tris pH 7.5, for 30 minutes at 37 ° C. After allowing transcription, the antibiotic (in a final concentration of 0.1% DMSO), 7 ml of Promega premix and amino acids in a final concentration of 0.1 mM were added in a final volume of 20 ml. The reactions were incubated for 20 minutes at 37 ° C and 50 ml of luciferase assay reagent (Promise part number: E 1483) was added to stop the reaction. The luminescence of the sample was measured by a Victor2V spectrophotometer (Perkin Elmer).

The rabbit reticulocyte lysate system (treated with nuclease) (Promega part number L4960) was used as the source of the eukaryotic translation and the protocol of the Promega 232 Technical Manual was followed. For example, 10 ml of lysate to 2 ml of 1 mM amino acids and 3 ml of water. 2 ml of antibiotic in 1% DMSO was added followed by the addition of 0.6 mg of luciferase mRNA (Promega part number L4561) and 16 ribonuclease inhibitor units (Promega part number N2115) in one volume 20 ml final The reaction was incubated at 24.5 ° C for 45 minutes and 40 ml of luciferase assay reagent (Promise part number: E1483) was added to stop the reaction. The luminescence of the sample was read by a Victor2V spectrophotometer (Perkin Elmer).

The resulting IC50 data for each antibiotic hybrid vs. Anisomycin, sparsomycin and chloramphenicol have been summarized in Table 22.

TABLE 22

Compound

IC50 (! M)

E. coli

Rabbit reticulocytes

Sparsocloranfenicol hybrid A1

29.9 0.82

Sparsocloranfenicol Hybrid B2

4.1 0.36

Deoxiesparsocloranfenicol 15

14 6.4

Sparsoanisomycin 3

26.7 0.07

Anisomycin

> 100 0.19

Sparsomycin

0.21 0.14

Chloramphenicol

12.3 > 100

The data in Table 22 indicate that all hybrid antibiotics produced in Examples 6 and 7 are capable of inhibiting translation in two different translation assays. On the other hand, under certain test conditions, at least one of the hybrid antibiotics (Sparsoanisomycin) was a more potent protein synthesis inhibitor than any of the antibiotics on which it was based.

H. Example 9: Union of macrolide antibiotics to the large ribosomal subunit

The crystalline structures of the large ribosomal subunit of H. marismortui formed complexes with five macrolides, carbomycin A, spiramycin, tylosin, azithomycin, and erythromycin showed that these antibiotics bind in the polypeptide exit tunnel immediately adjacent to the peptidyltransferase center. Tylosin, carbomycin A and spiramycin are 16-member macrolides. Azithotromycin is a 15-member macrolide. Erythromycin is a 14-member macrolide.

Their respective binding sites and masses suggest that they inhibit protein synthesis by blocking the passage of nascent polypeptides through the peptide exit tunnel. The one taken at position C5 of each of the lactone rings extends through the exit tunnel to the peptidyltransferase site. The carbomycin A disaccharide isobutyrate extension overlaps the substrate binding site A. A reversible covalent bond appears to be formed between the ethylaldehyde substituent at the C6 of the lactone ring and the N6 of A2103 (A2602, E. coli) at 23 RNAR.

Crystals containing large ribosomal subunits of H. marismortui with carbomycin A, spiramycin, tylosin, azithromycin, and bound erythromycin were obtained by adding stabilizing buffers containing these antibiotics to preformed crystals of large ribosomal subunits. The structures of these antibiotics complexed with these crystals were determined with a resolution of 3.0 A (3.5 A for erythromycin) from Fourier difference maps initially calculated using observed diffraction amplitudes of the native and complex crystals � Fo (complex) -Fo (native) �, and then the coordinates of the entire complex were refined.

The ribosomes were purified and crystallized as previously described (Ban et al., (2000). Antibiotics were selected for testing based on their known activity against H. marismortui (Sanz et al., (1993) Can J. Microbiol. 39: 311-317) and its availability. Carbomycin A was provided by Pfizer. Tylosin was purchased from Sigma. A mixture of spiramycins I, II and II, which only differ in part size. attached to C3, Sigma was also purchased Azithromycin was generously provided by Dale L. Boger, Department of Chemistry, The Scripps Research Institute, La Jolla, CA. Erythromycin was provided by Sigma.

Crystals containing large ribosomal subunits of H. marismortui complexed with carbomycin A, spiramycin, tylosin, azithromycin and erythromycin were obtained by immersion of crystals of large, pre-formed subunits in stabilizing buffers containing antibiotics. The antibiotics were solubilized in dimethylsulfoxide (DMSO), and then added to the standard stabilization buffer (Ban et al., (2000) supra) to a final concentration of 1.0 mM (and final DMSO of 1 to 4% ), and then incubated at 4 ° C for 24 hours before the cryovitrification of crystals in liquid propane. Initial X-ray diffraction data were collected for carbomycin A, spiramycin and tylosin antibiotics in beam lines X25, X12b or X12c at the Brookhaven National Laboratory. The data has been reduced by Denzo or HKL2000 software and scaled with SCALEPACK (Otwinowski (1997) Processing of Xray Diffraction Data Collected in Oscillation Mode, "Methods in Enzymology 276 (A): 307-326). The electronic density corresponding to these macrolides It was first seen on Fourier maps of difference Fo (antibiotic) -Fo (native) with a resolution of 4.0 A. Higher resolution data was collected in the beam line ID 19 in the Advanced Photon Source, Argonne National Laboratory The start of the macrolide models and their topology and parameter files was made by connecting and modifying the lactone and sugar rings of various related structures of small molecules (Woo et al. (1996) Tetrahedron 52 ( 11): 3857-3872; Jones et al. (1982) Journal of Antibiotics 35 (4): 420-5; Stephenson et al. (1997) Pharmaceutical Sci., 86: 1239) using standard sterochemical geometry and xplo-2d software (Kleywegt et al. (1998) Acta Cryst, D54: 1119-1131) and O (Jones et al. (1991) Acta Cryst. A47: 110-119). For the mixture of spiramycins, only spiramycin I was used in the determination of the structures. The antibiotic models were initially adapted to the electronic density maps of Fo-Fo difference. The structures of the antibiotics complexed with these crystals were determined with a resolution of 3.0 A with respect to the Fourier difference maps calculated using initially observed diffraction amplitudes of the native and complex crystals, �Fo (complex) -Fo ( native) �, and then the coordinates of the entire complex were refined. The structures of these complexes were refined by CNS (Brunger et al. (1998) Acta Cryst. D54: 905-921) for refinement for rigid body, energy minimization, and refinement of factor B of the entire structure of the native ribosome that includes the antibiotic, and by manual modifications of only the immediate area surrounding the bound antibiotic. Remote non-isomorphic differences with respect to the antibiotic were ignored. The refining process was repeated iteratively in the model that contained the antibiotic. The covalent structures of the macrolides and a conformational change that involves A2103 explained the main features in the difference electron density maps.

Based on these studies, the ethylaldehyde group at the C6 position of the lactone ring of each of the 16-member macrolides appears to form a covalent bond with the N6 of A2103. Not only the N6 of A2103 and the C6 ethylaldehyde of the macrolides are juxtaposed, but they are also linked by the continuous electronic density, which, with a resolution of 3.0 A, is indicative of the chemical bonds. In contrast, for both 15-member and 14-member macrolides (both lacking aldehyde groups) no such continuous electron density was observed. The only models for the macrolideribosome complex that satisfactorily adjust to the observed electron density are those in which the N6 of A2103 is covalently bound to the carbon of the aldehyde group of the macrolide. The aldehydes react reversibly with primary amines, and if the resulting carbinolamines become dehydrated, Schiff bases are produced. However, when the exocyclic amine of a nucleotide base is involved, the expected product is a carbinolamine, not a Schiff base (McGhee et al (1977) Biochem 14 (6): 1281-1303). Therefore, the model that best fits out-of-plane electronic density in this region joins them through a carbinolamine, not through a Schiff base.

Based on these studies, it has been found that carbomycin A, spiramycin, tylosin, azithromycin, and erythromycin bind, all of them, at the exit of the peptide tunnel from the large ribosomal subunit, immediately adjacent to the peptidyltransferase center. The five co-crystal structures were superimposed on the basis of ribosomal RNA phosphates only in order to objectively compare their relative binding. The three 16-member macrolides overlap almost on an atom-by-atom basis. In addition, all five lactone rings occupy a very similar position, share a common conformation and orientation, and form similar interactions with the ribosome. In addition, the shared parts, such as the micaminous of the 16-member rings (and the corresponding deosamine of the other macrolides) overlap on almost an atom basis by atom. All the sugar parts adopt the same extended extended conformations with respect to the lactone ring that was observed in the structures of small macrolide molecules.

Based on the resolved structures, it seems that hydrophobic interactions are important in the union of macrolides. One face of the lactone rings of these antibiotics is quite hydrophobic, while the opposite face is of a more hydrophilic character. All of these five macrolides bind to the ribosome with the hydrophobic faces of their lactone rings facing the wall of the exit tunnel and with the hydrophilic faces of their lactone rings exposed to the solution. The binding site of the tunnel wall includes the aromatic face of G2646, which is exposed because the base pair C2098-G2646 is of a helical termination.

In addition, these structures shed light on why the lengths of oligopeptides synthesized by poisoned macrolide ribosomes vary in the way they do. The length of oligopeptides synthesized in the presence of macrolide inhibitors is determined by the degree to which the substituents at the C5 position of the lactone ring penetrate the center of peptidyltransferase. Erythromycin and azithromycin, which have only one monosaccharide in this position, should allow the synthesis of peptides longer than tylosin or spiramycin, which have a disaccharide. These observations are compatible with those derived from biochemical studies that indicate that erythomycin does allow the formation of tetrapeptides, tylosin and spiramycin allow the formation of only dipeptides, and carbomycin A strongly inhibits the formation of even the first peptide bond.

INCORPORATION BY REFERENCE TITLE

The disclosure of each of the patent documents, scientific articles, atomic coordinates (which includes,

5 without limitation, those sets registered in the Research Collaborative for Structural Bioinformatics Protein Data Bank (PDB) with PDB access numbers ID: 1FFK; PDB ID: 1FF2; PDB ID: 1FG0; PDB ID: 1JJ2; PDB ID: 1K73; PDB ID: 1KC8; PDB ID: 1K8A; PDB ID: 1KD1; and PDB ID: 1K9M, and / or contained in Disk N.D 1) referred to herein, is incorporated herein by reference.

Hereby, all materials presented together with this in a compact disk, Disk N.D 1 are

10 incorporated herein by reference. Disk N.D 1 was created on February 6, 2002 and is identified with a content of the following thirty-nine files:

Disc N.D 1: EQUIVALENTS Therefore, the preceding embodiments should be considered in all aspects as illustrative rather than limiting the invention described herein.

Archive:

Magnitude (in bytes) Pages

�DIR>

BINDER

�DIR>

Us:

�1JJ2.RTF

12,742,023 2. 3 409

1JJ2.TXT

8,372,780 -

�ANISOMYCIN.PDB

7,593,128 7 to 13 1617

�azithromycin.pdb

8,088,411 126 to 132 1618

�BLASTICIDIN.PDB

7,594,206 14 to 20 1617

�CARBOMYCIN.PDB

7,592,552 21 to 27 1617

�erythromycin.pdb

7,690,337 133 to 139 1617

 FOLDER TO �DIR>

- -

FOLDER B �DIR>

- -

FOLDER C �DIR>

- -

�linezolid.pdb

8,086,197 28-34 1617

�PDB1FFK.DOC

7,046,656 4 266

PDB1FFK.ENT

5,484,652 - -

�PDB1FFZ.DOC 1

1,219,072 5 46

PDB1FFZ.ENT

937,342 - -

�PDB1FG0.DOC

1,255,728 6 46

PDB1FG0.ENT

942,344 - -

�SPARSOMYCIN.PDB

7,541,230 35-41 1606

�SPIRAMYCIN.PDB

7,592,549 42-48 1617

�TYLOSIN.PDB

7,595,512 49-55 1617

�VIRGINIAMYCIN.PDB

7,591,745 56-62 1 617

FOLDER:

�DIR>

Archive:

Magnitude (in bytes) Pages

�DIR>

BINDER

�DIR>

Us:

�DIR>

�1JJ2 PDB

8,270,586 - -

FOLDERB:

�DIR>

�DIR>

�ANISOMYC.PDB

8,383,598 63-69 1677

�BLASTICI.PDB

8,393,766 70-76 1679

�CARBOMYC.PDB

8,387,042 77-83 1677

�SPARSOMY.PDB

7,593,360 84-90 1617

�SPIRAMYC.PDB

8,402,048 91-97 1680

�TYLOSIN.PDB

8,339,400 98-104 1668

�VIRGINIA.PDB

7,590,514 105-111 1617

FOLDERC:

-

�DIR>

- -

�DIR>

- -

�AZITHROM.PDB

7,989,198 112 - 118 1618

�LINEZOLI.PDB

7,987,583 119-125 1617

These files are printed in the binders mentioned and contained in compact disk N0 1, both presented as part of the application with the preceding description in binder 1.

5 A copy of the compact disc previously referenced as a compact disc is submitted simultaneously with this application for inclusion in the public file of this EPO application. As noted earlier in the description, Figures 1 to 41 include Color Figures. Blank versions and

Black of these drawings are also included in this application as Figures 1 "to 33". A compact disc marked "Figures" of all these figures, in color and black and white, is presented simultaneously with this application to

10 its inclusion in the public file of this application. This description now continues with the list of atomic coordinates taken from compact disc 1 and with a list of sequences.

The next page of this description is page 1 of listing 1JJ2.RTF. SEQUENCE LIST 15 �110> Steitz, Thomas Moore, Peter Ban, Nenad Nissen, Poul Hansen, Jeffrey

�120> Structure of ribosome and protein synthesis inhibitors

�130> RIB-005

�150> US 09 / 922,251

�151> 2001-08-03

�160> 1

�170> PatentIn version 3.0 5 �210> 1

11211> 33

 12212> RNA

13213> Artificial Sequence

20220> 10 �223> oligonucleotide sequence that should form 12 base pairs

�400> 1 ccggcgggcu gguucaaacc ggcccgccgg acc 33

Claims (16)

1. A method for identifying a molecule that binds to a large ribosomal subunit, wherein the method comprises the following steps:

(to)

provide a molecular model comprising one or more target regions selected from the group consisting of at least a portion of: (i) an anisomycin binding site, (ii) an azithromycin binding site, (iii) a binding site of blasticidine, (iv) a carbomycin binding site, (v) an erythromycin binding site, (vi) a linezolid binding site, (vii) a sparsomycin binding site, (viii) a binding site of spiramycin, (ix) a tylosin binding site, and (x) a virginiamycin M binding site, where the molecular model is made

(one)

 from the atomic coordinates corresponding to the great ribosomal subunit of Haloarcula marismortui found in Disk 1 under the file names ANISOMYCIN.PDB, azithromycin.pdb, BLASTICIDIN.PDB, CARBOMYCIN.PDB, erythromycin. pdb, linezolid.pdb, SPARSOMYCIN.PDB, SPIRAMYCIN.PDB, TYLOSIN.PDB, or VIRGINAMYCIN. PDB, or found on Disk 1, under FOLDERB, under the file names ANISOMYC.PDB, BLASTICI.PDB, CARBOMYC. PDB, SPARSOMY.PDB, SPIRAMYC.PDB, TYLOSIN.PDB, VIRGINIA.PDB, or that are located on Disk 1, under FOLDERC, under the file names AZITHROM.PDB, or LINEZOLI.PDB, or

(2)

from the atomic coordinates derived by molecular modeling through the use of coordinates corresponding to the great ribosomal subunit of Haloarcula marismortui found in Disk 1 under the file names ANISOMYCIN.PDB, azithromycin. pdb, BLASTICIDIN.PDB, CARBOMYCIN.PDB, erythromycin.pdb, linezolid.pdb, SPARSOMYCIN.PDB, SPIRAMYCIN.PDB, TYLOSIN.PDB, or VIRGINAMYCIN.PDB,

or that are on Disk 1, under FOLDERB, under the file names ANISOMYC.PDB, BLASTICI.PDB, CARBOMYC.PDB, SPARSOMY.PDB, SPIRAMYC.PDB, TYLOSIN.PDB, VIRGINIA.PDB, or found in Disk 1, under FOLDERC, under the file names AZITHROM.PDB, or LINEZOLI.PDB; Y

(b)

 the use of the molecular model to identify a candidate molecule that can bind to the molecular model.

2.

The method according to claim 1, wherein in step (a) (2) the molecular modeling consists of homology modeling.

3.

The method according to claim 1, wherein in step (a) (2) the molecular modeling consists of the molecular replacement.

Four.

The method according to claim 1, which comprises the additional step of determining whether the candidate molecule modulates ribosomal activity.

5.

The method according to claim 4, comprising the additional step of repeating one or more of the steps (a) and (b) to identify a modified candidate molecule.

6.

The method according to claim 5, which comprises the additional step of determining whether the modified candidate molecule modulates ribosomal activity.

7.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the filename ANISOMYCIN.PDB, or on Disk 1, under FOLDERB, under the filename ANISOMYC.PDB.

8.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the filename azithromycin.pdb or on Disk 1, under FOLDERC, under the filename AZITHROM.PDB.

9.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the filename BLASTICIDIN.PDB or on Disk 1, under FOLDERB, under the filename BLASTICI.PDB.

10.

The method according to claim 1, wherein the atomic coordinates are located on Disk 1 under the filename CARBOMYCIN.PDB or on Disk 1, under FOLDERB, under the filename CARBOMYC.PDB.

eleven.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the file name erythromycin.pdb.

12.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the file name linezolid.pdb or on Disk 1, under FOLDERC, under the file name LINEZOLI.PDB.

13.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the filename SPARSOMYCIN.PDB, or on Disk 1, under FOLDERB, under the filename SPARSOMY.PDB.

14.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the filename SPIRAMYCIN.PDB or on Disk 1, under FOLDERB, under the filename SPIRAMYC.PDB.

fifteen.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the filename TYLOSIN.PDB or on Disk 1, under FOLDERB, under the filename TYLOSIN.PDB,

16.

The method according to claim 1, wherein the atomic coordinates are found on Disk 1 under the file name VIRGINIAMYCIN.PDB or on Disk 1, under FOLDERB, under the file name VIRGINIA.PDB.

Various components of PTC are sought