MXPA99002631A - Fmoc-trinucleotido-fosforamiditos and its use as mutagenic units for the construction of enhanced combinatory libraries with low multiply substitutions - Google Patents

Abstract

The present invention relates to a group of compounds consisting of trinucleotide-phosphoramidites which are protected in the OH 5'with the 9-fluorenylmethoxycarbonyl group (Fmoc) and in the OH 3'with the phosphoramidite group, their synthesis and their use as mutagenic units in the construction of combinatorial libraries enriched with low multiplicity substitutions. This constitutes a valuable tool for the study of the structure-function relationship of proteins and their engineering. By means of this strategy, few amino acid replacements can be generated per protein and prevent destruction of the function.

Description

Fmoc-trinucleotide-phosphoramidites and their use as mutagenic units for the construction of combinatorial libraries enriched with low multiplicity substitutions.

TECHNICAL FIELD OF THE INVENTION The invention object of the present invention relates to a group of compounds consisting of trinucleotide-phosphoramidites which are protected in OH 5 'with the 9-fluorenylmethoxycarbonyl group (Fmoc) and in OH 3' with the phosphoramidite group, their synthesis and its use as mutagenic units in the construction of combinatorial libraries enriched with low multiplicity substitutions.

BACKGROUND OF THE INVENTION The proteins are constituted mainly by 20 different amino acids, and since the DNA is only composed of 4 different nucleotides, each of the amino acids of a protein is encoded by a group of three nucleotides that is called triplet or codon.

It is now known that the function of a protein, such as molecular recognition or catalysis, depends on its characteristic three-dimensional structure ordered by its primary amino acid sequence. For this reason many research groups are interested in studying this type of relationship in order to rationally improve a large number of enzymes that may well have a commercial application.

Protein engineering is the structural modification of these materials through the modification of their coding gene, in order to understand and manipulate the relationships that are established between the amino acid sequence, the final folding of the polypeptide chain and its function.

This work is feasible thanks to advances in recombinant DNA technology and methods of chemical synthesis of DNA, which make it possible to replace, eliminate or add any amino acid in a protein, modifying in its coding gene the respective trinucleotide called codon; It is also known that a large number of proteins can dramatically alter their properties by changing a single, functionally critical amino acid.

Site-directed mutagenesis (codon), through oligonucleotides, is being used extensively in almost all disciplines of biochemistry to explore the relationship between the structure and function of proteins. However, the selective perturbation of individual amino acids requires some understanding at the molecular level of the interacting structures (protein-protein, enzyme-substrate or protein-DNA). That is, you need to have some bases to predict the changes of some amino acid residues that will produce a particular functional consequence in the protein. Therefore, rational attempts in the engineering and evolution of proteins to modify existing proteins are limited to the need to have a high resolution structure of the protein of interest and a good understanding of its mechanism. Unfortunately, there are few systems that meet this requirement [Her es, J. D. 1989].

An alternative method, of a more general nature, both for the understanding of structure-function relationships, and for protein engineering, employs random mutagenesis or also called combinatorial mutagenesis. In this method a random "mix" of mutants (library) is generated, instead of introducing particular premeditated changes. Once the library of mutant genes is created, they are cloned and the proteins that these genes encode are expressed in an appropriate host. The transformed colonies are then selected or monitored, looking for the appearance of a phenotype caused by the properties of the new proteins (e.g., greater thermostability, greater catalytic power, different specificity, etc.).

A large number of combinatorial gene mutagenesis protocols have been described in the literature, being classified in general into chemical, enzymatic and oligo-directed methods [Botstein, D. 1985].

Chemical methods require exposure of the microorganism to mutagenic reagents [Myers, R.M. 1985 and Kadonaga, J.T. 1985] (eg sodium bisulfite, hydroxylamine, nitrous acid, etc.), which cause modifications on the nucleotides that they make up in cellular DNA. In enzymatic methods [Lehtovaara, et al. 1988] it is necessary to make use of polymerases that make mistakes during nucleotide addition (eg the reverse trahscriptase of the avian myeloblastosis virus). In both cases, libraries of mutant genes that present specific nucleotide changes are generated, generating a sequence space in which it is only possible to explore between 24% and 40% of all possible amino acid changes, a direct consequence of the degeneracy of the code. genetic [Sirotkin, K. 1986]. In both methods, the window of mutagenesis, the frequency, the distribution of amino acids, and the level of mutagenesis can only be controlled in a crude manner.

The method of combinatorial mutagenesis directed by oligonucleotides, resorts to the use of libraries of synthetic oligonucleotides, produced in a single experiment through mixtures of the four nucleotides [Sirotkin, K. 1986 and Del Rio, G. 1994]. This type of mutagenesis can be performed "by saturation" or "contamination". In the first of these, each wild nucleotide of the codons to be mutated is replaced by a mixture of the four nucleotides, generating a library of variant codons of size 64n (32n in case of using the NNG / C system, OJO Arkin, A. 1992) where n represents the number of codons to be replaced. Considering a practical transformation efficiency of 107 to 109 colonies, with this methodology it is only possible to analyze a maximum of 5 amino acids per experiment. However, the main drawback of this methodology is the preferential generation of multiple mutants (several changes of codons per gene), which normally generate destruction of the function of the protein.

The second corresponds to the combinatorial mutagenesis by oligonucleotide-directed contamination, currently considered the best method to create variability, is to contaminate each of the wild nucleotide couplings during the chemical synthesis of the oligonucleotide, with a small proportion (a) of the 3 non-wild nucleotides [Hermes, JD 1989]. Although this methodology makes it possible to obtain libraries rich in low multiplicity mutants (eg simple and double mutants) and to explore windows of relatively large mutagenesis, the problem inherent in the degeneracy of the genetic code is still present, that is, the substitution of those amino acids whose codons only vary on a basis with respect to wild ones.

The ideal random mutagenesis protocol must meet many requirements. First, the region of n codons or amino acids to be explored (that is, the window of mutagenesis) must be easily specified. Second, each codon located in the mutagenesis window must have the same probability of being replaced (homogeneous distribution). Third, the substitution of an amino acid for any of the other 19 must occur with the same probability (homogeneous frequency). Fourth, the system must allow defining immutable positions within the window of mutagenesis. Fifth, it should be possible to control the rate or level of mutagenesis (a) in order to be able to adjust the density of mutants desired through the combinatorial analysis theory, and sixth, the mutagenesis efficiency must be high so that most of the clones analyzed correspond to mutant sequences [Hermes, JD 1989; Lehtovaara, P.M. 1988 and Sondek, J. & Shortle, D. 1992].

Several authors agree that the ideal combinatorial mutagenesis method should involve the use of trinucleotide mixtures (codon level mutagenesis) that can be coupled during conventional oligonucleotide synthesis. With this methodology, libraries of variants could be obtained where the probability of finding substitutions for each of the 19 non-wild amino acids is comparable.

During the period 1992-1996, some articles related to the synthesis and application of dimethoxytrityl trinucleotide phosphoramidites appeared in the literature [Sondek, J. & Shortle, D. 1992; Virnekas, B., 1994; Ono, A. 1995; Lyttle, M. H. 1995 and Kayushin, A. L. 1996]. In four of them the trinucleotide mixture was used in combinatorial mutagenesis experiments to saturate several residues. After analyzing a large number of clones per sequence, the result was the expected logic, a high proportion of multiple mutants. Shortle, D. (1992), was the only one that used a trinucleotide contamination strategy, only instead of creating substitutions it generated codon insertions. Likewise, Shortle D. refers to the versatility and advantages of this methodology, as long as the technical problem of having trinucleotide-protected phosphoramidites in the terminal 5 'hydroxyl with a protective group orthogonal to dimethoxytrityl (DMT) is solved. Orthogonally understood as a group of chemical characteristics contrary to those of DMT, that is, this group must be stable to the acidic conditions (2% trichloroacetic acid to dichloromethane) used to remove the DMT group and selectively removable under mild basic conditions. to be stable to the automated conditions of oligonucleotide synthesis (trichloroacetic acid, iodine, acetic anhydride and N-methylimidazole).

The main drawback of the DMT-trinucleotide-phosphoramidites synthesized by Shortle D., was its low reactivity, because comparatively with the standard of the concentrations used in monomers that is 0.1M, its maximum performance per coupling was only 4%, being completely independent of the reaction time.

It should be mentioned that the inventors of the present invention submitted the results obtained from their research works to publication and they were published in the journal: Chemistry & Biology Vol. 5 No. 9 p. 519-527 (1998), with the title: "Combination of DMT-mononucleotide and F or-trinucleotide phosphoramidites in oligonucletide synthesis with an automated automatable codon-level utagenesis method", which according to article 18 of the Industrial Property Law It should not be considered as a prior art for the purpose of determining the patentability of the invention. t i i i? DETAILED DESCRIPTION OF THE INVENTION BRIEF DESCRIPTION OF THE FIGURES Figure 1 shows an example of the synthesis of F oc-trinucleotide-phosphora iditos. Wherein Glb represents 2N-isobutyrylguanine; dCbz represents 6N-benzoyl-deoxycytidine; dT represents thymidine; NMI represents N-methylimidazole; MeOClPNiPr2 represents chloro-N, N-diisopropyl-methylphosphonamide; DIPEA represents N, N-diisopropylethylamine.

Figure 2 shows the strategy of constructing combinatorial libraries enriched with low multiplicity substitutions using Fmoc-trinucleotide-phosphoramidites. Where • -0DMT represents any of the 4 conventional DMT-nucleoside-phosphoramidites; CCO-Fmoc represents an Fmoc-trinucleotide-phosphoramidite or a mixture thereof; TCA, represents a solution of 2% trichloroacetic acid in dichloromethane; DBU represents a 0.1 M solution of 1,8-diazabicyclo [5.4.0] undec-7-ene in acetonitrile.

It has been recognized for several years that methods of site-directed mutagenesis to create diversity in fragments of genes of various sizes suffer from significant disadvantages due to a predisposition introduced by the degeneracy of the genetic code and inherent limitations in the distribution of replacements resulting from substitution. of bases one by one, against three at a time. These limitations are particularly relevant when the goal is to achieve one or a few mutations per gene, that is, a low multiplicity, which is generally achieved in conditions of unsaturation or contamination. In contrast to saturation conditions, the opposite is achieved, that is, several mutations per gene, concentrated in one or in a few contiguous codons.

As mentioned in the previous chapter, some authors agree that the synthesis of combinatorial genetic libraries is feasible with the degeneracy of the fully controlled genetic code, through the use of mixtures of trinucleotides during the chemical synthesis of oligonucleotides.

However, the use of DMT-trinucleotide-phosphoramidites in combination with conventional methods of oligonucleotide synthesis (requiring DMT-nucleoside-phosphoramidite) can only generate combinatorial libraries with multiple codon substitutions or low multiplicity codon insertions.

An alternative to solve the problem of generating combinatorial libraries enriched with low multiplicity substitutions is through an oligonucleotide synthesis strategy that combines DMT-nucleoside-phosphoramidites and trinucleotide-phosphoramidites protected at OH 5 'with a group orthogonal to DMT. Understanding orthogonal to DMT as that group that is stable to the acidic conditions used to remove the DMT group and therefore selectively removable in other conditions.

However, the technical limitation of not having this type of trinucleotide phosphoramidites that have the characteristics of orthogonality to DMT persists and that they also have good coupling efficiency in solid phase synthesis and stability to automated oligonucleotide synthesis conditions ( 2% trichloroacetic acid in CrL, Cl2, iodo, acetic anhydride and N-methylimidazole).

The inventors propose as solution to this problem the obtaining of trinucleotide-phosphoramidites that are protected in OH 5 'with the 9-fluorenylmethoxycarbonyl group (Fmoc) as "orthogonal group to DMT and the phosphoramidite group in OH 3', which they confer specific and favorable characteristics for their use in the method of combinatorial mutagenesis.

The present invention relates to a group of trinucleotide phosphoramidites characterized by being protected with the 9-fluorenylmethoxycarbonyl group (Fmoc) in the OH 5 'and a phosphoramidite group in the OH 3', their synthesis and use as mutagenic units for the construction of combinatorial libraries enriched with low multiplicity substitutions.

Therefore, having a codon-level mutagenesis strategy, which generates libraries with low multiplicity substitutions by variant, constitutes a valuable tool for the study of the structure-function relationship of proteins and their engineering. By means of this strategy, few amino acid replacements can be generated per protein and thus prevent their functional destruction.

One of the main and characteristic aspects of the present invention is the protection of trinucleotides with the 9-fluorenylmethoxycarbonyl group (Fmoc), which has several advantages compared to other protective groups orthogonal to DMT. The Fmoc group is commercially available, has a high regioselectivity for OH 5 'and its removal by soft basic treatment is carried out quickly.

The regioselectivity of the Fmoc group favors obtaining the main product in the OH 5 'position, with high yields and minimizes the generation of undesirable by-products. On the other hand, the characteristic that is quickly removable by a soft basic treatment, gives greater advantages over other groups orthogonal to the DMT group for use in combinatorial strategies in solid phase, because it is not necessary to modify the standard protocols of synthesis recommended by the manufacturer.

The presence of the phosphoramidite group in OH 3 'makes it possible to incorporate said trinucleotide phosphoramidites in automated oligonucleotide synthesis.

Another important aspect of these compounds is their synthesis route, which generally comprises the following steps: a) The synthesis of the Fmoc-mononucleoside, which includes the reaction of the 9-fluorenylmethoxycarbonyl chloride and the corresponding nucleoside base, appropriately protected in the exocyclic amino, in pyridine as solvent; once the reaction is finished, the washing, purification and recovery of the product is carried out; b) The synthesis of the Fmoc-dinucleoside monophosphate, which involves the Fmoc-mononucleoside reaction of part a) previously dried by co-evaporation with anhydrous pyridine and activated with the phosphorylation reagent o-chlorophenyldithriazole phosphate, with the following corresponding nucleoside base, appropriately protected in the exocyclic amino and in the presence of a catalyst for coupling this reaction, N-methylimidazole; once the reaction is completed, the washing, purification and recovery of the product is carried out; The synthesis of the Fmoc-trinucleoside diphosphate, which involves the reaction of the Fmoc-dinucleoside monophosphate of subsection b) previously dried by co-evaporation with anhydrous pyridine and activated with the phosphorylation reagent chlorophenylditriazolfosphate, with the following 3 'terminal nucleoside base defining the trinucleotide . This base must be properly protected. N-Methylimidazole is added to catalyze the coupling reaction. Once the reaction is completed, the washing, purification and recovery of the product is carried out; d) The phosphorylation of the Fmoc-trinucleoside diphosphate, which involves the reaction of the Fmoc-trinucleoside diphosphate of part c) previously dried by coevaporation with anhydrous pyridine and followed by a coevaporation with tetrahydrofuran (THF), with N, N-diisopropylethylamine (D) PEA) and the chlorine N, N-diisopropyl-methylphosphonamide, the reaction is stopped by the addition of pyridine and diluted with an organic solvent, then the washing, purification and recovery of the product is carried out.

As is known to a person skilled in the art, there are a wide variety of methods of washing, purifying and recovering the product that can be used in this process. For example: in the washing the reaction can be diluted with different organic solvents and treated with different aqueous solutions; the purification in turn can be carried out by chromatography or by crystallization and the recovery can be carried out by precipitation on an apolar organic solvent such as hexane or ethyl ether among others.

As can be seen, the described route is a general synthesis route, since the nucleoside bases referred to in steps a), b), and c), can be A, T, C or G, and in addition, it can be a different base in each step or the same base in the three steps. To demonstrate this, the inventors have synthesized trinucleotide-phosphoramidites of different combinations such as: GCT, TTT, AAA, CAG and TGC, in which the added first base (step a) can be any of the four nucleoside bases; the second may be the same as the first or different and the third may be the same as the first or the second or different. Therefore, it is obvious to a person skilled in the art that by this means it is feasible to produce the 64 possible combinations of trinucleotides based on these four nucleoside bases.

Another way to carry out the synthesis of the Fmoc-trinucleotide phosphoramidites with higher yields is to follow the synthesis route described above, but preferably by washing the products obtained in steps a), b), c) and d ) with a concentrated ionic solution. The monomers, dimers and trimers protected with the Fmoc group, exhibit resistance to mild acid conditions, but are labile to mild basic conditions; for this reason it is important that the washing solution be non-basic and preferably ionic.

A central aspect of the present invention is the use of the trinucleotide phosphoramidites as mutagenic units for the construction of libraries enriched with low multiplicity substitutions (few changes of codons per gene), characterized in that it comprises the following steps: a) Sequentially coupling, on a solid support, the nucleoside phosphoramidites protected in the OH 5 'by a conventional group, to form a wild-type sequence corresponding to the adjacent zone 3' of the window of mutagenesis that it is desired to scan in a given gene; b) initiating the first cycle of mutagenesis, by coupling to the wild-type sequence, in suitable proportions, a mixture containing the protected mutagenic units, and the next protected nucleoside-phosphoramidite by a conventional group corresponding to the wild-type sequence; c) the following two couplings are performed using nucleoside-phosphoramidites protected by a conventional group corresponding to the wild-type sequence, to reach the length of the sequence that has been replaced by the addition of the mutagenic units; d) removal of protective groups, both conventional and orthogonal, through basic treatments and mild acid, to complete the first cycle of mutagenesis; e) optionally, repeating steps (b) to (d) as many times as necessary to conclude the number of codons that one wishes to explore; f) after the mutagenesis cycles, assemble the corresponding wild sequence to the adjacent zone 5 'of the mutagenesis window, using nucleoside-phosphoramidites protected by a conventional group.

The couplings of the DMT-nucleoside-phosphoramidites and Fmoc-trinucleotide-phosphoramidites were made in an automatic DNA synthesizer through a normal synthesis protocol suggested by the manufacturer. This protocol includes the following steps for each coupling: 1) hydrolysis of the DMT group with an acid solution (2% dichloroacetic acid in dichloromethane); 2) coupling of the DMT-nucleoside-phosphoramidite in the presence of tetrazole to the growing oligonucleotide in the solid support; 3) blocking by acetylation of the 5 'OH' s that did not react and 4) oxidation of the newly formed triester phosphite to more stable triester phosphate.

An important aspect in the use of the Fmoc-trinucleotide-phosphoramidites for the construction of libraries enriched with low multiplicity substitutions, is that they can be combined with nucleoside-phosphoramidites protected in OH 5 'with a conventional group; these mononucleosides are used to synthesize the wild-type sequence of the corresponding oligonucleotide. A conventional group is defined as a group labile to mild acidic conditions, which protect commercially available nucleoside phosphoramidites, such as, for example, the dimethoxytrityl group (DMT) or monomethoxytrityl (MMT). Therefore, the Fmoc group has orthogonal characteristics to these conventional groups.

Another relevant point in the use of the Fmoc-trinucleotide-phosphoramidites for the construction of libraries enriched with low multiplicity substitutions, is the flexibility of the mutagenesis window, since the number of codons to be explored can be very variable, from a minino of two codons up to the maximum allowed by the particular characteristics of the synthesis team and the economics of the experiment.

The equipment currently available allows the synthesis of oligonucleotides of up to 200 nucleotides, so by deducting 10 base pairs for each of the ends adjacent to the window, we can manage a window of mutagenesis of 180 nucleotides (60 codons).

It is obvious to an expert in the state of the art that as new devices allow the synthesis of larger oligonucleotides, the mutation window may grow in the same proportion.

On the other hand, it should be noted that the use of the Fmoc-trinucleotide-phosphoramidites in the construction of combinatorial libraries enriched with low multiplicity substitutions allows the level of mutagenesis to be controlled by the appropriate management of their proportions with respect to the conventionally protected nucleoside-phosphoramidite. which defines the first nucleotide of the wild-type codon to be substituted.

The flexibility in the handling of these proportions of mutagenesis allows to generate libraries of variants whose population follows a binomial distribution, predictable by means of the following equation: F = - P X (1-P) xl (n-x)! Where: P represents the probability of finding certain types of mutants (single, double or triple, etc.) n is the size of the mutagenesis advantage (expressed as the number of codons to be explored) x the type of mutant (the simple corresponds to the No. 1, double No. 2, etc.). p the level of mutagenesis.

In view of the fact that our method is focused on the production of low multiplicity mutants, generally the p mutagenesis level will be low and will be determined by the aforementioned variables.

Also, an additional advantage that results from combining two alternative chemistries (Fmoc-trinucleotide-phosphoramidites and conventional nucleoside-phosphoramidites) with high coupling efficiency, is the fact that the mutant libraries are built on a single support, making this process of mutation is automatable.

In order to illustrate the obtaining and application of the trinucleotide phosphoramidites in the construction of combinatorial libraries enriched with low multiplicity substitutions, some examples are described below for different trinucleotide phosphoramidites and their use in the synthesis of some oligonucleotides, generated at different levels of mutagenesis.

EXAMPLES In the following examples the present invention is described in order to illustrate it better, but of course without restricting its scope.

Example 1 In this example, the synthesis of 5'-O- (9-fluorenylmethoxycarbonyl-2N-isobutyryldeoxyguanosine-3'-yl- (O-chlorophenyl) -phosphate-5'-yl-4N-benzodeloxycytidine-3'-yl- (0-chlorophenyl) is illustrated. -phosphate- 5 '-thymidine-3' -o-methyl-N, N-diisopropyl phosphoramidite (Fmoc-dGCT-phosphoramidite) For the synthesis of Fmoc-dGCT-phosphoramidite, the following steps are carried out: a) The synthesis of Fmoc-dG is carried out by the reaction of Fmoc-Cl (13 mmol, 3.36 g) and 2N-isobutyryldeoxyguanosine (10 mmol, 3.37 g) in pyridine as solvent; once the reaction was completed, washing was carried out, purification by chromatography and 7.83 g of the product was recovered; b) The synthesis of Fmoc-d (GC) monophosphate, is carried out by the reaction of Fmoc-dG (5 mmol, 2.80 g) previously dried by co-evaporation with anhydrous pyridine and activated with the phosphorylation reagent o-chlorophenyldithriazole phosphate, with 3 molar equivalents of 4N-benzoyldeoxycytidine (15 mmol, 4.97 g) in the presence of N-methylimidazole as a catalyst; once the reaction was completed, the washing was carried out, purification by chromatography and 2.8 g of the product was recovered; c) The synthesis of Fmoc-d (GCT) -diphosphate, is carried out by the reaction of Fmoc-d (GC) -monophosphate (3 mmol, 3.18 g) previously dried by co-evaporation with anhydrous pyridine and activated with the reagent phosphorylation o-chlorophenyldithriazolfosphate; with thymidine (9 mmol, 2.18 g) in the presence of N-methylimidazole; once the reaction is complete, the washing is carried out, purification by chromatography and 2.21 g of the product is recovered; d) The phosphorylation of Fmoc-d (GCT) -diphosphate is carried out by the reaction of (l mol, 1.47 g) of Fmoc-d (GCT) diphosphate previously dried by co-evaporation with anhydrous pyridine (2 x 20 ml) and followed by a coevaporation with 20 ml of tetrahydrofuran (THF), with N, N diisopropylethylamine (4 mmol, 696 μl) (DIPEA) and the chloro N, N diisopropyl-methylphosphonamide (3 mmol, 603 μl), the reaction is stopped by the addition of pyridine and diluted with dichloromethane. After washing with an ionic saturated solution at neutral pH, the reaction is concentrated and purified by column chromatography. The pure product is concentrated and precipitated in hexane at -40dC. The product is stable at -202C for at least 2 years.

Nuclear magnetic resonance analyzes of: H and 31P confirming the identity of the Fmoc-dGCT-phosphoramidite compound were performed on a Varian VXR equipment at 300 and 121 MHz respectively. The reported yield corresponds to the phosphorylation step of the trimer. The results obtained were the following: Performance: 44%; R £ 0.62; * H NMR (CDC13, 300 Mhz): (12.10 (Hl of dG, 1H, ls), 10.68, 10.65, 10.62 and 10.29 (NH of dG, 1H, 3s), 8.89 (H3 of dT, 1H, sa), 8.22-8.06 (H6 of dC, 1H, m), 7.83-7.10 (aromatics, 20H, m), 6.39- 5.89 (Hl ', 3H, m ), . 40-5.31 (H3 'of dG + H3' of dC, 2H, m), 4.66-4.32 (CH, of Fmoc + H4 'of dG + H4' and H5 'of dC + H3', H4 'and H5' of dT, 10H, m), 4.22 (CH of Fmoc + H5 'of dG, 3H, m), 3.56 (CH of isopropyl, 2H, m), 3.36 (MeOP, 3H, m), 3.10-2.10 (CH of isobutyryl + D2 + H2 'from dC + H2' from dT, 7H, m), 1.83 (d7 H7, 3H, m), 1.26 (isopropyl CH3, 12H, m), 1.16 (isobutyryl CH3, 6H , m). 31P NMR (CDC13): (150.60-149.97 (P for phosphoramidite, 1P, m), -6.62 to -7.44 (P for phosphates, 2P, m).

Example 2 In this example, the synthesis of 5'O- (9-fluorenylmethoxycarbonyl-thi idine-3'-yl- (O-chlorophenyl) -phosphate-5'-yl-imidin-3'-yl- (0-chlorophenyl) -phosphate) is illustrated. '-thymidine-3' -0-methyl-N, N-diisopropyl phosphoramidite (Fmoc-dTTT-phospho-amide) For the synthesis of Fmoc-dTTT-phosphoramidite, the same steps as in Example 1 are carried out, but as a base nucleoside, thymidine is used in steps a), b) and e), with the respective monomer, dimer and trimer formed in those steps. Likewise, the reported performance corresponds to the phosphorylation stage of the trimer and the results obtained from the identity analyzes were the following: Yield: 75%; R £ 0.27; 'H NMR (CDC13, 300 Mhz) of the diastereomeric mixture: (Fmoc + o-chlorophenyl + H6 aromatics, 19H, m), 6.24 (Hl', 3H, m), 5.24 (2H3 ', m), 4.60- 4.18 (1H3 '+ CH and CH2 of Fmoc + 6H5' + 3H4 ', 13H, m), 3.62-3.45 (CH of isopropyl, 2H, m), 3.37 and 3.36 (CH3OP, 3H, 2d, J = 13.2 Hz) , 2.65-2.17 (H2 ', 6H,), 1.88-1.72 (H7, 9H,), 1241 (isopropyl CH3, 12H, d, J = 3.3 Hz). p NMR (CDC13): (146.40 and 146.90 (1P, phosphoramidite P, 2s), -11.10 (2P, phosphate P, ls).

Example 3 In this example, the synthesis of 5'O- (9-fluorenylmethoxycarbonyl-6N-benzoyldeoxyadenosine-3'-yl- (0-chlorophenyl) -phosphate-5'-yl-6N-benzoyldeoxyadenosine-3'-yl- (O-chlorophenyl) is illustrated. -phosphate- 5'-yl-6N-benzoyldeoxyadenosine-3'-o-methyl-N, N-diisopropyl phosphoramidite (Fmoc-dAAA-phosphoramidite) For the synthesis of Fmoc-d (AAA) -phosphoramidite, the same steps as in Example 1 are carried out, but as a nucleoside base, 6N-benzoyldeoxydenosine is used in steps a), b) and c), with the respective monomer, dimer and trimer formed in those steps. Likewise, the reported performance corresponds to the phosphorylation stage of the trimer and the results obtained from the identity analyzes were the following: Yield: 81%; R £ 0.43; * H NMR (CDC13, 300 Mhz) of the diastereomer mixture: (9.507 and 9.344 (3H, NH of dA, 2sa), 8.706, 8.666 and 8.580 (3H, H8 of dA, 3d, J = 7.8, 6.9 and 4.2 Hz ) 8.292 - 8.119 (3H, H2 of dA, m), 7.957 (2H, Ha of Fmoc, dd, J = 7.2 and 8.1 Hz), 7.686 (2H, Hd of Fmoc, dd, J = 3.3 and 7.5 Hz) , 7.533-7.050 (27H, 2Hc and 2Hb of Fmoc + 8H of o-chlorophenyl + 15H of benzoyl, m), 6.503 -6.358 (3H, Hl 'of dA, m), 5,500-5,378 (3H, H3' of dA , m), 4.561-4.334 (11H, 6H5 'and 3H4' of dA + CH2 of Fmoc, m), 4.162 (1H, CH of Fmoc, t, J = 6.9 Hz), 3.403 (3H, CH ^ OP, m ), 3.024 - 2.673 (8H, 6H2 'of dA + 2 CH of isopropyl, m), 1217 - 1184 (12H, isopropyl CH3, m). 31P NMR (CDC13): (156.328 and 150.020 (1P, P of phosphoramidite , 2s), -7.01, -7.270, -7.320, -7.534 and -7.666 (2P, phosphate P, 5s).

Example 4 In this example, the synthesis of 5'O- (9-fluorenylmethoxycarbonyl-4N-benzoyldeoxycytidine-3'-yl- (0-chlorophenyl) -phosphate-5'-yl-6N-benzoyldeoxyadenosine-3'-yl- (O-chlorophenyl) is illustrated. -phosphate- 5 '-2N- isobutyryldeoxyguanosine-3' -o-methyl-N, N-diisopropyl phosphoramidite (Fmoc-d (CAG) -phosphoramidite) For the synthesis of Fmoc-d (CAG) -phosphoramidite, the steps of example 1 are carried out, but as a nucleoside base it is used in step a) 4N-benzoyldeoxycytidine; in step b) 6N-benzoyldeoxyadenosine and in step c) 2N-isobutyryldeoxyguanosine. With the respective monomer, dimer and trimer formed in those steps. Likewise, the reported performance corresponds to the phosphorylation stage of the trimer and the results obtained from the identity analyzes were the following: Yield: 43.63%; R. 0.27; H NMR (CDC13, 300 Mhz) of the diastereomer mixture: (22,200"(1H, NH from dG, sa), 12.096 (1H, NH from dA, sa), 10.2 (1H, NH from dC, s), 9.9 (1H, HL of dG, s), 8.776 - 8.711 (2H, Ha of Fmoc, m), 8.037-7,884 (6H, 2Hd, 2Hc and 2Hb of Fmoc, m), 7.719 -7.255 (23H, 10H of 5 benzoyl + 8H of o-chlorophenyl + H5 and H6 of dC + H2 and H8 of dA + H8 of dG, m), 6.661 - 6.555 (1H, Hl 'of dG, m), 6.455 - 6.387 (1H, Hl 'of dA, m), 6.267 - 6.156 (1H, Hl' of dC, m), 5.487 (1H, H3 'of dA, sa), 5.160 (1H, H3' of dC, sa), 5.084 (1H, H3 'from dG, sa), 4.646 - 4.353 (11H, 2H5' and 1H4 'of dC + 102H5 'and 1H4' of dA + 2H5 'and 1H4' of dG + CH, of Fmoc), 4.186 (1H, CH of Fmoc, t, J = 6.9 Hz), 3.416 - 3.99 (3H, CH3OP, m), 2.922 -2.758 (4H, 2H2 'of dC + 2H2' of dA, m), 2.654 - 2.094 (5H, 2H2 'of dG + CH of Fmoc + 2 CH of isopropyl, m), 1198 - 1163 (18H, 2 CH3 of isobutyryl + 4 CH3 isopropyl, m). 31P NMR (CDC13): (150.101 and 149.982) 1P, phosphoramidite P, 2s), -6.705, -6.811, -6.855, -6.692 and -7.377, -7.577, -7.810, -7.936 (2P, P of phosphate , 8s).

Example 5 The synthesis of 5'- (9-fluorenylmethoxycarbonyl-thymidine-3'-yl- (O-chlorophenyl) -phosphate-5'-yl-2N-isobutyryldeoxyguanosine-3 'yl- (0-chlorophenyl) is illustrated in this example. ) - 5'-N-benzoyldeoxycytidine-3'-phosphate-o-methyl-N, N-diisopropyl phosphate phosphoramidite. (Fmoc- (TGC) -phosphoramidite) For the synthesis of Fmoc-d (TGC) -phosphoramidite, the same steps as in Example 1 are carried out, but as a nucleoside base it is used in step a) thymidine; in step b) 302N-isobutyryldeoxyguanosine and in step c) 4N-benzoyldeoxycytidine, with the respective monomer, dimer and trimer formed in those steps. Likewise, the reported performance corresponds to the phosphorylation stage of the trimer and the results obtained from the identity analyzes were the following: Yield: 65%; Rf 0.30; * H NMR (CDCl 3, 300 Mhz) of the diastereomer mixture: (12,200 (1H, NH from dG, sa), 10,800 and 10,600 (1H, HL from dG, 2s), 10,200 and 10,100 (1H, NH from dC, sa ), 9.300 (1H, NH of dT, sa), 7.918 -7.715 (8H, Fmoc aromatics, m), 7.554 - 7.129 (17H, 8H of o-chlorophenyl + 5H of benzoyl + H6 of dT + H8 of dG + H5 and H6 of dC, m), 6.298 -6.195 (3H, Hl 'of dT + Hl' of dC + Hl 'of dG, m), 5.486 -5.063 (3H, H3' of dT, H3 'of' dC + H3 'of dG, m), 4.601 -4.217 (12H, 2H5' and 1H4 'of dT + 2H5' and 1H4 'of dC + 2H5' and 1H4 'of dG + CH2 and CH of Fmoc, m), 3.390 - 3.299 (3H, CHDP, m), 2744-2.241 (8H, 2H2 'of dT + 2H2' of dC + 2H2 'of dG + 2 CH of isopropyl, m), 1719 (3H, dH H6, d .J = 6.3 Hz), 1258 - 1013 (18H, 2 CH3 isobutyryl + 4 CH3 isopropyl, P NMR (CDC13): (150.681 and 148.850) 1P, phosphoramidite P, 2s), -6.070, -6.673, -6.767, -6.962 , -7,520, -7,650, -7,823, -7,911 (2P, phosphate P, 8s).

Example 6 In this example, the use of Fmoc-trinucleotide-phosphoramidites for the construction of a library enriched with low multiplicity substitutions is illustrated.

As a study model, the inventors decided to mutagenize codons 19, 20 and 21 (coding for valine, proline and serine) of the gene encoding the β-galactosidase a peptide located in the plasmid pUC19 and characterized by containing a large amount of restriction sites. The fragment of the sequence is the following: 'TAG GAG GAT CCC CGG GTA GCG AGC TCG AAT TCA CTC GGA C 3' An Applied Biosystems 381 A DNA synthesizer was used using a 0.2 μmol synthesis scale and a standard protocol recommended by the equipment manufacturer. The synthesis of the mutagenic oligonucleotides requires the use of 5 vials of the synthesizer: four of them are loaded with the conventional DMT-nucleoside phosphoramidites of DMT-dA, DMT-dG, DMT-dC and DMT-T at 100 mM concentration in acetonitrile anhydrous, which were used to assemble the wild-type sequence, while the fifth vial (which we shall call X) was loaded with Fmoc-d (GCT) -31.1 mM phosphoramidite in anhydrous acetonitrile, for this particular example, although mixtures of different Fmoc-trinucleotide-phosphoramidites at different concentrations. A window of 3 codon mutagenesis was selected ie the three underlined codons GTA CCG AGC. To carry out the synthesis, we proceeded as follows: a) Sequentially coupling on a solid support, the phosphoramidites DMT-dA, DMT-dC, DMT-dG and DMT-T, to form the 3 'wild sequence adjacent to the window of mutagenesis, ie: TCG AAT TCA CTC GGA C b) initiate the first cycle of mutagenesis, coupling the previously synthesized fragment (adjacent 3 'sequence), simultaneously the mixture of mutagenic units (in this case only Fmoc-d (GCT) -phosphoramidite) and the DMT-dC-phosphoramidite , of which the same volume (65 μl of each) is added; c) the next two couplings are performed according to the wild-type sequence using DMT-dG-phosphoramidite and DMT-dA-phosphoramidite to reach the length of the sequence that has been replaced by the previous addition of Fmoc-d (GCT) -phosphoramidite; d) remove the DMT and Fmoc groups by consecutive treatments with solutions of mild acids (2% trichloroacetic acid in dichloromethane, for one min.) and soft basic solutions (Diazabicycloundecene or DBU, 100 mM), to conclude the first cycle of mutagenesis; e) repeating steps (b) to (d) for the next two codons to be mutated, selecting the appropriate DMT-nucleoside phosphoramidites according to the wild-type sequence; f) once the mutagenesis window has been completed, the 5 'adjacent sequence is assembled, which is: TAG GAG GAT CCC CGG, using the corresponding DMT-nucleoside-phosphoramidites.

The oligonucleotide combinatorial library thus generated was removed from the support by basic hydrolysis with concentrated ammonium hydroxide and subsequently purified by polyacrylamide gel electrophoresis.

The level of mutagenesis achieved with the Fmoc-d (GCT) -phosphoramidite in this example was 19.87%. According to the equation of binomial distribution mentioned in the previous chapter, the expected theoretical proportion of single, double and triple variants should be: 38.27, 9.49 and 0.78% respectively, while experimentally, after sequencing 50 clones obtained from the library combinatorial oligonucleotides, 22, 6 and 0% respectively of the variants cited above were obtained. In this way, although the theoretical and experimental results do not agree exactly due to the low population analyzed, it is well known that the experimental population of variants follows a distribution very similar to the binomial one.

It should be noted that the codon used in this example (GCT) codes for the amino acid alanine, therefore, additionally, it has the advantage that this methodology can be applied directly to "alanine scavenging studies", that is, for substitution sequential of each of the wild-type amino acids of a protein by alanine in the mutagenesis window, which allows knowing the function of each of the wild-type amino acids in the function of the protein.

REFERENCES Hermes, J. D., Parekh, S.M., Blacklow, S.C., Kdster, H. and Knowles, J.R. (1989). A reliable method for random mutagenesis: the generation of mutant librarles using spiked oligodeoxyribonucleotide primers. Gene 84, 143-151.

Lehtovaara, P.M., Koivula, A.K., Bamford, J. and Knowles, J.K.C. (1988). A new method for random mutagenesis of complete genes: enzymatic generation of mutant libraries in vitro. Pro. Eng. 2, 63-68.

Sondek, J. and Shortle, D. (1992). A general strategy for random insertion and substitution mutagenesis: substoichiometric coupling of trinucleotides phosphoramidites. Proc. Nati Acad. Sci. 85, 1777-1781.

Botstein, D. and Shortle, D. (1985). Strategies and applications of in vitro mutagenesis, Science 229, 1193-1201.

Myers, R.M. , Lerman, L.S., and Maniatis, T. (1985). A general method for saturation mutagenesis of cloned DNA fragments, Science 229, 242-247.

Kadonaga, J.T. and Knowles, J.R. (1985). A simple and efficient method for chemical mutagenesis of DNA, Nucleic Acids Res. 13, 1733-1745.

Sirotkin, K. (1986). Advantages to mutagenesis technical generating populations containing the complete spectrum of single codon changes. J. Theor. Biol. 123, 261-279.

Arkin, A.P. & Yuovan D.C. (1992). Optimizing nucleotide mixtures to encode specific subsets of aminoacids for semi-random mutagenesis. Bio / Tech. 10, 297-300.

Del Rio, G. Osuna, J. and Soberón, X. (1994). Combinatorial libraries of proteins: analysis of efficiency of mutagenesis techniques. BioTechniques 17, 1132-1139.

Virnekas, B., et al., And Moroney, S.E. (1994). Trinucleotide phosphoramidites: ideal reagents for the synthesis of mixed oligonucleotides for random mutagenesis. Nucleic Acids Res. 22, 5600-5607.

Ono, A., Matsuda, A., Zhao, J. and Santi, D.V. (nineteen ninety five). The synthesis of blocked-triplet-phosphoramidites and their use in mutagenesis. Nucleic Acids Res. 23, 4677-4682.

Lyttle, M.H. Napolitano, E.W. , Calió, B.L., and Kauvar, L.M. (nineteen ninety five) . Mutagenesis using trinucleotide β-cyanoethyl phosphoramidites. Nucleic Acid Res. 19, 274-280.

Kayushin, A.L., Korostelava, M.D., Miroshnikov, A.I., Kosch, W., Zubov, D. and Skin, N. (1996). A convenient approach to the synthesis of trinucleotide phosphoramidites-synthons for the generation of oligonucleotide / peptide libraries. Nucleic Acids Res. 24, 3748-3755.

Claims (11)

R E I V I N D I C A C I O N S

1. - Trinucleotide phosphoramidites characterized by being protected with the protecting group 9-fluorenylmethoxycarbonyl (Fmoc) in OH 5 'and a phosphoramidite group in OH 3'.

2. - Trinucleotide-phosphoramidites of claim 1, characterized in that the possible combinations of its pyrimidic and pyrimidic bases can be any of the 64 possible combinations of A, G, C and T.

3. - Trinucleotide phosphoramidites of claim 2 characterized in that the possible combinations of their bases are preferably: GCT, TTT, AAA, CAG and TGC.

4. - A synthesis route of the trinucleotide phosphoramidites of claim 1, characterized in that it comprises the following steps: a) the synthesis of the Fmoc-mononucleoside, which includes the reaction of the 9-fluorenylmethoxycarbonyl chloride and the corresponding nucleoside base, appropriately protected in the exocyclic amino, in pyridine as solvent; once the reaction is finished, the washing, purification and recovery of the product is carried out; b) synthesis of the Fmoc-dinucleoside monophosphate, which involves the Fmoc-mononucleoside reaction of part a) previously dried by co-evaporation with anhydrous pyridine and activated with the phosphorylation reagent o-chlorophenyldithriazole phosphate, with the following corresponding nucleoside base, appropriately protected in the exocyclic amino and in the presence of a catalyst for coupling this reaction, N-methylimidazole; once the reaction is completed, the washing, purification and recovery of the product is carried out; c) the synthesis of the Fmoc-trinucleoside diphosphate, which involves the reaction of the Fmoc-dinucleoside monophosphate of subsection b) previously dried by co-evaporation with anhydrous pyridine and activated with the phosphorylation reagent chlorophenylditriazolfosphate, with the following 3'-terminal nucleoside base defining the trinucleotide. This base must be properly protected. N-Methylimidazole is added to catalyze the coupling reaction. Once the reaction is completed, the washing, purification and recovery of the product is carried out; d) phosphorylation of the Fmoc-trinucleoside diphosphate, which involves the Fmoc-trinucleoside diphosphate reaction of part c) previously dried by coevaporation with anhydrous pyridine and followed by a coevaporation with tetrahydrofuran (THF), with N, N-diisopropylethylamine (DIPEA) ) and the chlorine N, N-diisopropyl-methylphosphonamide, the reaction is stopped by the addition of pyridine and diluted with an organic solvent, then the washing, purification and recovery of the product is carried out.

5. - A synthesis route of the trinucleotide phosphoramidites according to claim 4, characterized in that in steps a), b) and c) the washing of the products is preferably carried out with a neutral ionic solution.

6. - The use of the trinucleotide phosphoramidites of claim 1 as mutagenic units for the construction of libraries enriched with low multiplicity substitutions, characterized in that it comprises the following steps: a) coupling, sequentially on a solid support, the nucleoside-phosphoramidites protected by a conventional group, to form a wild-type sequence corresponding to the adjacent zone 3 'of the window of mutagenesis that it is desired to scan in a determined gene; b) initiating the first cycle of mutagenesis, by coupling to the wild-type sequence, in suitable proportions, a mixture containing the protected mutagenic units, and the next protected nucleoside-phosphoramidite by a conventional group corresponding to the wild-type sequence; c) the following two couplings are performed using nucleoside-phosphoramidites protected by a conventional group corresponding to the wild-type sequence, to reach the length of the sequence that has been replaced by the addition of the mutagenic units; d) removal of both conventional and orthogonal protective groups, by consecutive treatments with mild acidic and soft basic solutions regardless of the order, to conclude the first cycle of mutagenesis; e) optionally, repeating steps (b) to (d) as many times as necessary to conclude the number of codons that one wishes to explore; f) concluded the mutagenesis cycles, assemble the wild sequence corresponding to the adjacent zone 5 'of the mutagenesis window, using nucleoside-phosphoramidites protected by a conventional group.

7. The use of the trinucleotide phosphoramidites according to claim 6, characterized in that the conventional protective group is preferably dimethoxytrityl (DMT).

8. - The use of the trinucleotide phosphoramidites according to claim 6, characterized in that the window of mutagenesis is preferably flexible to any number of codons per library.

9. The use of the trinucleotide phosphoramidites according to claim 6, characterized in that the appropriate proportions of the mutagenic units and the nucleoside phosphoramidite are defined by the level of mutagenesis.

10. The use of the trinucleotide phosphoramidites according to claim 6, characterized in that the level of mutagenesis is preferably low for the generation of low multiplicity oligonucleotide mutants.

11. - The use of the trinucleotide phosphoramidites according to claim 6, characterized in that the use of a single support from the beginning of the synthesis favors its automation. R E S U E N The invention relates to a group of compounds consisting of trinucleotide phosphoramidites which are protected at OH 5 'with the 9-fluorenylmethoxycarbonyl group (Fmoc) and at OH 3' with the phosphoramidite group, their synthesis and their use as units mutagenic in the construction of combinatorial libraries enriched with low multiplicity substitutions. This constitutes a valuable tool for the study of the structure-function relationship of proteins and their engineering. By means of this strategy, few amino acid replacements can be generated per protein and thus prevent their functional destruction.