KR20040070286A - Nuclear receptor ligands and ligand binding domains - Google Patents

Abstract

The present invention provides compositions for the production of nuclear receptor synthetic ligands based on the three-dimensional structure of nuclear receptors, in particular thyroid receptors (referred to herein as "TR"), and new methods thereof, in particular methods using computers do. Also provided are crystals, nuclear receptor synthetic ligands, and related methods.

Description

NUCLEAR RECEPTOR LIGANDS AND LIGAND BINDING DOMAINS}

The present invention relates to ligands that bind to nuclear receptors, determination of nuclear receptors, methods of designing synthetic ligands of nuclear receptors using computers, and methods of using synthetic ligands.

Nuclear receptors represent a superfamilly of proteins that specifically bind to physiologically relatively small molecules such as hormones or vitamins. Although nuclear receptors may have transcription-independent actions, by binding to a nuclear receptor, the nuclear receptor changes the cell's ability to transcrib DNA. That is, nuclear receptors regulate the transcription of DNA. Unlike intrinsic membrane receptors and membrane related receptors, nuclear receptors are present in the cytoplasm or nucleus of eukaryotic cells. Thus, nuclear receptors comprise a class of ligand regulatory transcription factors that are soluble in cells.

Nuclear receptors may be involved in receptors for glucocorticoids (GR), receptors for androgen (AR), receptors for mineralocorticoids (MR), receptors for progestins (PR), receptors for estrogens (ER), thyroid hormones Receptors for vitamin D (TR), receptors for vitamin D (VDR) and receptors for retinoids (RAR and RXR). In addition, the so-called "orphan receptor" is part of the nuclear receptor phase and is structurally homologous to typical nuclear receptors such as steroids and thyroid receptors. To date, ligands have been identified that do not have all-pan receptors, but small molecular ligands for the class of ligands of transcription factors are likely to be found soon. In general, nuclear receptors specifically bind to physiologically relatively small molecules with high affinity and the apparent Kd is generally from 0.01 to 20 nM and depends on the nuclear receptor / ligand pair.

Despite the myriad of physiological processes and the importance of nuclear receptors in diseases such as hypertension, inflammation, hormone-dependent cancers (e.g. breast and prostate cancer), regulation of reproductive organs, hyperthyroidism, hypercholesterolemia and obesity, The development of synthetic ligands that specifically bind to nuclear receptors has gone through many drug design trials and errors. Previously, new ligands specific for nuclear receptors have been discovered without knowledge of the three-dimensional structure of nuclear receptors having ligands bound to them. Prior to the present invention, researchers were actually finding nuclear receptor ligands by simply probing, without knowing how amino acids of the nuclear receptor captured the ligand.

In conclusion, we devised a method and composition that can detect ligands for nuclear receptors, synthesize these compounds, and reduce the time required to administer these compounds to organisms to modulate physiological actions regulated by nuclear receptors. It would be advantageous to pay.

The present invention provides for the determination of nuclear receptor ligand binding domains having a ligand bound to a ligand binding domain (LBD). The crystals of the present invention provide good atomic resolution of amino acids that interact with nuclear receptor ligands, particularly thyroid receptor ligands. A three-dimensional model of nuclear receptor LBD with bound ligand revealed a previously unknown structure for nuclear receptors, indicating that the ligands are bound in binding cavities inaccessible to the water of the ligand binding domain of the nuclear receptor.

The present invention also provides a method of using a three-dimensional model of the nuclear receptor based on the determination of the nuclear receptor LBD using a computer. In general, methods for designing nuclear receptor ligands using a computer use a three-dimensional model of crystallized protein, including nuclear receptor LBD with bound ligand, to interact with the chemical residue (one or more) of the ligand. A second chemistry having a structure that determines the amino acid (s) of and reduces or increases the interaction between the interacting amino acid and the second chemical moiety as compared to the interaction between the interacting amino acid and the corresponding chemical moiety on the natural hormone. It is to select a chemical modification (one or more) of the chemical moiety that produces the moiety.

1 is a diagram illustrating how a computer can be used to design a ligand that interacts with a nuclear receptor on a nuclear receptor phase.

2 is a schematic representation of the nuclear receptor structure showing the function of homologous regions and various domains within family members.

Figure 3 shows the aligned amino acid sequence of the ligand binding domain of several members on the nuclear receptor.

4 is a ribbon diagram of rat TR-α LBD with labeled secondary structural elements. Ligand (magenta) was depicted as the space filling model. Alpha helix and coil form is yellow and beta strands are blue.

Figure 5 shows two cross-sections of the space-filled model of rat TR-a exposing ligands (magenta) packed into the receptor.

6 is a schematic of a ligand binding cavity. Residues that interact with the ligand appear almost at the site of interaction. Hydrogen bonds are indicated by dotted lines between bond partners, and the distance for each bond is recorded. Unbonded contact is represented as a radical spoke facing the interacting atom.

7 shows the distribution of crystallographic temperature factors in purified rat TR-α LBD. The distribution is expressed as a color grade in the range of less than 15 (dark blue) and greater than 35 (yellow-green).

8 is a ribbon diagram of rat TR-α LBD showing c-terminal activation domain for ligand. Residues containing the c-terminal activation domain (Pro393-Phe405) are indicated by bars. Hydrophobic residues, in particular Phe401 and Phe405 (blue), are directed inward of the ligand and Glu403 (red) is projected out of the solvent.

9 is an electrostatic potential surface of rat TR-α LBD calculated using GRAPH. The negative electrostatic potential is red; The positive electrostatic potential is blue. The c-terminal activation domain forms a large hydrophobic (white). Glu403 is shown as a single fragment of negative charge.

10 is a diagram comparing agonists and antagonists for several nuclear receptors.

11 is a synthesis scheme for manufacturing TS1, TS2, TS3, TS4, and TS5.

12 is a synthesis scheme for producing TS6 and TS7.

13 is a synthesis scheme for producing TS8.

14 is a synthesis overview for producing TS10.

Figure 15 depicts the chemical structure of several TR ligands.

16 is a graph showing the competitive assay compete with T ₃ and triac T ₃ the cover to (triac) is binding to human TR-α or human TR-β.

FIG. 17 shows a scattercat analysis of labeled T ₃ binding to TR-α and TR-β.

FIG. 18 is a chart demonstrating the effect of TS-10 on transcriptional regulation of DR4-ALP reporter gene in the presence or absence of T3 as assayed in TRAFα1 reporter cells.

19 is a chart demonstrating the effect of TS-10 on the transcriptional regulation of the DR4-ALP reporter gene in the presence or absence of T3 as assayed in TRAFβ1 reporter cells.

FIG. 20 is a chart showing the effect of TS-10 on transcriptional regulation of DR4-ALP reporter gene in the presence or absence of T3 as assayed in HepG2, ie, liver reporter cell line.

FIG. 21 shows a partial ribbon plot of TR-α LBD with T3 in ligand binding cavities. Selected interacting amino acids were labeled including Ile221, Ile222 and Ser260, Ala263, Ile299 and Leu276.

FIG. 22 shows a partial ribbon plot of TR-α LBD with T3 and Dimit nested in ligand binding cavities. The interaction of Ile221, Ile222, Ala260, Ile299 and Leu276 was labeled.

FIG. 23 is a partial ribbon plot of TR-α LBD with T3, showing three polar arginine residues (Arg228, Arg262 and Arg266 (indicated by black bars)) of three pockets of water: HOH502, HOH503 and HOH504 , Dotted line represents a hydrogen bond.

FIG. 24 is a partial ribbon plot of TR-α LBD with triac, showing three polar arginine residues (indicated by black bars) of water pockets (HOH503, HOH504 and HOH600), dotted lines showing hydrogen bonds Indicates.

FIG. 25 is a partial ribbon diagram of TR-α LBD with T3 and triac nested in ligand binding cavities. The figure shows several interacting amino acid residues in the polar pockets which remain unchanged whether T3 or triac occupy the ligand binding cavity: Arg262, Asn179, HOH503 and HOH504, and Ser277. Both Arg228 and Arg266 occupy two different positions depending on whether T3 or triac bind.

FIG. 26 is a stereochemical diagram of TRα LBD with bound dimit. FIG.

Appendix 1 is an appendix to the references.

Appendix 2 is a chart of amino acids interacting with TR ligands for TR complexed with Dimit, Triac, IpBr2 and T3.

Introduction

The present invention provides a novel method for preparing nuclear receptor synthetic ligands based on the three-dimensional structure of nuclear receptors, in particular thyroid receptors (herein referred to as "TR"), in particular methods and compositions using computers. Previously, the lack of information of three-dimensional structures for ligand binding domains of nuclear receptors, in particular the absence of information of three-dimensional structures for nuclear receptors with bound ligands, has hindered the field of nuclear receptor drug discovery.

First described herein are crystallographic and three-dimensional structure information from the ligand binding domain (LBD) of nuclear receptors with bound ligands. Such crystals provide excellent resolution at the atomic level and the ability to visualize the coordination of nuclear receptor ligands by amino acids, including LBD. The present invention also provides a method of designing a nuclear receptor synthetic ligand using a computer using such crystals and three-dimensional structure information to generate a synthetic ligand that modulates the morphological changes of LBD of the nuclear receptor. Such synthetic ligands can be designed using methods using the computers described herein and some of which are shown in FIG. 1. Such computerized methods are particularly useful for designing antagonists or partial agonists for nuclear receptors, wherein the antagonists or partial agonists are observed activation against naturally occurring ligands or other ligands that mimic naturally occurring ligands such as agonists. It has extended residues that interfere with any one of a number of ligand-induced molecular events that alter the effect of receptors on the regulation of gene expression, such as disrupting normal coordination of domains. As described herein, synthetic ligands of nuclear receptors will be useful for modulating nuclear receptor activity in a variety of medical conditions.

Applicability to Nuclear Receptors

The present invention, in particular the computerized method, includes a receptor for glucocorticoid (GR), a receptor for androgen (AR), a receptor for mineralocorticoid (MR), a receptor for progestin (PR), a receptor for estrogen (ER ), Drugs for various nuclear receptors such as receptors for thyroid hormones (TR), receptors for vitamin D (VDR), receptors for retinoids (RAR and RXR), and receptors for peroxysome proliferators (PPAP) It can be used to design. The present invention can also be applied to “all-pan receptors,” which are structurally homologous in terms of the modular structure and primary structure for typical nuclear receptors such as steroids and thyroid receptors. For example, when compared to rat RARα and human TR-β receptors, the amino acid homology of all-fan receptors with other nuclear receptors ranges from very low levels (<15%) to 35%. In addition, the overall folding of the ligand bound phase and members is likely to be similar, as revealed by the structural analysis described herein and the X-ray crystallographic structure of the TR. Although the ligands are not identical to all-pan receptors, once such ligands have been identified, those skilled in the art will be able to apply the invention for the design and use of such ligands, and their overall structural modular motifs are similar to other nuclear receptors described herein. something to do.

Modular action domain of nuclear receptor

The present invention generally refers to nuclear receptor activation, structure, and as a result of determining the three-dimensional structure of nuclear receptors having significant three-dimensional structures or crystallized protein structures of ligand binding domains for different binding ligands, TR-α and TR-β, and Pattern of regulation, and all nuclear receptors described herein. Proteins with nuclear receptor superstructures represent substantial regions of amino acid homology as described herein and known in the art shown in FIG. Members of this family represent the overall structural motifs of the three modular domains (similar to the TR three modular domain motifs):

1) variable amino terminal domains;

2) a highly conserved DNA binding domain (DBD); And

3) Less conserved carboxyl terminal ligand binding domain (LBD).

This modularity allows the different domains of each protein to individually achieve different actions, even though the domains may affect each other. Separation of domains is generally conserved when a particular domain is isolated from the residual protein. By using conventional protein chemistry techniques, existing domains can sometimes be separated from the parent protein. By using conventional molecular biology techniques, each domain can be expressed individually, generally having its original intact function, or chimeras of two different nuclear receptors can be constructed, which are produced It retains the properties of the individual functional domains of each nuclear receptor from chimeras.

2 provides a schematic of a family member structure showing family members of various domains and homologous regions within functional groups.

Amino terminal domain

The amino terminal domains are the least conserved of the three domains and significantly change the size of the nuclear receptor phase and members. For example, such domains contain 24 amino acids in VDR and 603 amino acids in MR. These domains are involved in transcriptional activation, and in some cases, their uniqueness may define the activation of target genes by selective receptor-DNA binding and specific receptor isoforms. Such domains may exhibit synergistic and opposing interactions with domains of LBD. For example, studies of mutated and / or deleted receptors show positive interactions of amino and carboxy terminal domains. In some cases, deletion of one of these domains will eliminate the transcriptional activation action of the receptor.

DNA-binding domain

DBD is the most conserved structure on nuclear receptors. These generally comprise about 70 amino acids folded with two zinc finger motifs, which zinc ions are coordinating with four cysteines. The DBD includes two vertically oriented α-helices extending from the bases of the first and second zinc fingers. The two zinc fingers work together along non-zinc finger residues that are straight at the receptor homodimer or heterodimeric interface and straight to the nuclear receptor for specific target sites on the DNA. The various amino acids of the DBD affect the separation between two half sites (typically six nucleotides) for receptor dimer binding. For example, the GR subfamily and the ER homodimer bind to half sites spaced by three nucleotides and oriented as palladium. Optimal spacing promotes cooperative interactions between DBDs and the D box residues are part of the dimerization interface. Another area of DBD promotes DNA-protein and protein-protein interactions required for RXR homodimerization and heterodimerization in direct repeat elements.

LBD can affect the DNA binding of DBDs, which can also be controlled by ligand binding. For example, TR ligand binding affects the extent to which TR binds to DNA as a monomer or dimer. This dimerization also depends on the spacing and orientation of half the DNA regions.

The nuclear receptor superfamily is divided into two subclasses: 1) GR (GR, AR, MR and PR) and 2) TR (TR, VDR, RAR, RXR and most all-fan receptors) based on the DBD structure, It interacts with heat shock protein (hsp) and has the ability to form heterodimers. GR subgroup members are tightly bound by hsp in the absence of ligand and, after ligand binding and dissociation of hsp, dimerize and show homology at the half-region of DNA to which they bind. Also, these half sites tend to be arranged as palindroms. TR subgroup members tend to bind to DNA or other chromatin molecules when separated from the ligand, and may bind to DNA as monomers and dimers, but form heterodimers and form DNA in various orientations and spacings in half regions. It binds to the element and also shows homology to the nucleotide sequence in half. ER is similar to the GR subfamily in hsp interactions and therefore does not belong to the TR subfamily in nuclear localization and DNA binding properties.

Ligand Binding Domain

LBD is the second most conserved domain in the receptor. The integrity of several different LBD sub-domains is important for ligand binding, whereas truncated molecules containing only LBD have normal ligand binding activity. In addition, these domains have other functions, including dimerization, nuclear translocation, and transcriptional activity, as described herein. Importantly, these domains bind ligands and undergo ligand induced morphological changes as described in detail herein.

Most superfamily members, including all-pan receptors, have two or more transcriptionally active subdomains, one of which is a constituent and is present in the amino terminal domain (AF-1), and these (AF-2 (also TAU 4) The other) is in the ligand binding domain whose activity is regulated by the binding of the agonist ligand. The action of AF-2 requires an activation domain (also called a transactivation domain) that is highly conserved in the receptor supernatant (about amino acids 1005-1022). Most LBDs contain an activation domain. Some mutations in these domains abolish AF-2 function but neglect ligand binding and do not affect other functions. Ligand binding allows the active domain to provide an interaction site for a substantial coactivator protein that acts to stimulate (or in some cases inhibit) transcription.

As described herein, the carboxy-terminal activating subdomain is located three-dimensionally close to the LBD for the ligand such that the ligand is coordinated (or interacted) with amino acids in the activating subdomain to bind to the LBD. As described herein, the LBDs of nuclear receptors are expressed and crystallized and their three-dimensional structure is bound to bound ligands (using crystal data from the same receptor or from different receptors or combinations thereof) and ligands for LBD. In particular, it is determined by methods using a computer used to design ligands containing extension residues that are coordinated with the activation domain of the nuclear receptor.

As described herein and known in the art, once a ligand (CDL) designed using a computer is synthesized, an assay is used to establish the activity and affinity of the agent, partial agent or antagonist as described herein. Can be tested. After this test, CDL can be further purified by inducing LBD crystals with CDL bound to LBD. The structure of the CDL is then further purified using the chemical modification methods described herein for three-dimensional models to improve the activity and affinity of the CDL and to improve the improved properties of the good agents or antagonists described herein. A second resulting CDL having was prepared.

Nuclear receptor isoform

The present invention can also be applied to prepare new synthetic ligands for distinguishing nuclear receptor isoforms. As described herein, CDLs that distinguish isoforms may be prepared to prepare tissue specific or action specific synthetic ligands. For example, except for two PR isoforms A and B, which are translated from the same mRNA by cross initiation from different AUG codons, the GR subfamily members generally have one receptor encoded by a single gene. This method is particularly applicable to TR subfamily with several receptors encoded by two (TR) or three (RAR, RXR and PPAR) genes or having cross RNA splicing, examples of which are described herein. Described in

Nuclear receptor crystals

The present invention provides crystals made from nuclear receptor ligand binding domains having a ligand bound to the receptor. As illustrated in the examples, TR crystallizes with the ligand bound thereto. Crystals are prepared from purified nuclear receptor LBD generally expressed by cell culture, preferably E. coli culture. Preferably, different crystals (co-crystals) for the same nuclear receptor are one or more bromo- or iodo-substituted synthetic ligands that act as analogs or antagonists of different ligands such as naturally occurring ligands and naturally occurring ligands. It is prepared separately using. These bromo- and iodo- substituents act as heavy atom substituents in the crystals of nuclear receptor ligands and nuclear receptor proteins. This method has the advantage in the stage of the crystal bypassing the need to obtain typical heavy metal derivatives. If the three-dimensional structure is measured for the ligand-bound nuclear receptor LBD, the three-dimensional structure can be used in computerized methods to design synthetic ligands for the nuclear receptor and further active structural relationships are described herein. And can be measured through periodic tests using assays known in the art.

Expression and Purification of Other Nuclear Receptor LBD Structures

High levels of expression of nuclear receptor LBD can be obtained by the techniques described herein as well as other documents. TR including members of steroid / thyroid receptor superfamily such as estrogen receptor (ER), androgen receptor (AR), mineralocorticoid receptor (MR), progesterone (PR), RAR, RXR and vitamin D receptor (VDR) And high levels of expression in E. coli of the ligand binding domains of other nuclear receptors can also be achieved. Yeast and other eukaryotic expression systems can be used for nuclear receptors that bind to heat shock proteins because these nuclear receptors are generally difficult to express in bacteria except for ER, which can also be expressed in bacteria. to be. Representative nuclear receptors or their ligand binding domains have been cloned and sequenced: human RAR-α, human RAR-γ, human RXR-α, human RXR-β, human PPAR-α, human PPAR-β, human PPAR-γ, Human VDR, human ER (described in Seielstad et al., Molecular Endocrinology, vol 9: 647-658 (1995)), human GR, human PR, human MR and human AR. The ligand binding domain of each of these nuclear receptors has been identified and is shown in FIG. 3. Using the information of FIG. 3 in conjunction with the methods described herein and known in the art, one of ordinary skill in the art can express and purify any LBD of nuclear receptors, including those shown in FIG. 3, bind them to suitable ligands, The ligand may crystallize the LBD of the bound nuclear receptor.

3 is a straight line of several dogs of the steroid / thyroid hormone receptor superfamily that represent amino acids included in suitable expression vectors.

Extracts of expressing cells are suitable sources of receptors for purification and preparation of crystals of selected receptors. To obtain this expression, the vector is structured in a manner similar to that used for the expression of rat TR alpha [April et al. Protein Expression and Purification, 6: 368-370 (1995). Nucleotides encoding amino acids surrounding the ligand binding domain of the expressed receptor, e.g., estrogen receptor ligand binding domain (hER-LBD) (R to position 1025 at position 725, normalized in a straight line as shown in FIG. (Corresponding to L in) is inserted into an expression vector such as that used by Aprilleti (1995) et al. In order to obtain substances which yield good crystals, it is preferred to include at least one amino acid corresponding to human TR-β positions 725 to 1025. If more structural information is desired, extended sequences of contiguous amino acid sequences may be included. Thus, expression vectors for human estrogen receptors can be prepared by inserting nucleotides encoding amino acids from positions 700 to c-terminal positions 1017. Such vectors provide high yields of receptors in E. coli, which can bind to hormones. Seielstad et al. Molecular Endocrinology Vol 9: 647-658 (1995). However, the c-terminal region via position 1025 is variable hydrolyzed and can be advantageously blocked from the structure, and this technique for avoiding variable hydrolysis can also be applied to other nuclear receptors.

TR-α and TR-β and TR expression, purification and crystallization functions as examples of nuclear receptor LBD structures

As an example of the nuclear receptor structure of the ligand binding domain, the α and β isoforms of TR are crystallized from proteins expressed from an expression structure, preferably a structure that can be expressed in E. coli. Other expression systems may be used, such as yeast and other eukaryotic expression systems. For TR, LBD can be expressed without DBD or amino terminal domain moieties. If additional structural information of an amino acid proximate to the LBD is desired, a portion of the DBD or amino terminus may be included. In general, for TR, the LBD used for the determination will be less than 300 amino acids long. Preferably, the TR LBD will be at least 150 amino acids in length, preferably at least 200 amino acids in length, most preferably at least 250 amino acids in length. For example, the LBD used for crystallization may comprise the amino acids of Met 122 to Val 410 of rat TR-α, Glu 202 to Asp 461 of human TR-β.

Typically, TR LBD is homogeneously purified for crystallization. Purity of TR LBD is determined by SDS-PAGE, mass spectrometry and hydrophilic HPLC. The purified TR for crystallization should have a purity of at least about 97.5%, preferably at least 99.0%, most preferably at least 99.5%.

Initial purification of the ligand unbound receptor can be obtained by conventional techniques such as hydrophobic interaction chromatography (HPLC), ion change chromatography and heparin affinity chromatography.

In order to obtain more advanced purification for nuclear receptors, in particular TR subclasses and for the modified determination of TR, ligand transfer purification of nuclear receptors is carried out using columns that separate the receptors according to charge, such as ion exchange or hydrophobic interaction columns. It is then preferred to bind the ligand, in particular the receptor eluted with the agonist. When the rechromatated receptor on the same column elutes at the position of the ligand bound receptor, the ligand induces a change in the surface charge of the receptor such that the receptor is removed by the original column performed with the non-ligand bound receptor. In general, saturation concentrations of ligands are used in the column, and the protein can be preincubated with the ligand prior to hanging them on the column. Structural studies detailed herein demonstrate the general applicability of this technique to obtain high purity nuclear receptor LBD for crystallization.

More recently developed methods involve treating a "tag", such as histidine, located at the end of a protein, such as at the amino terminus, and then using a nickel chelating column for purification. Janknecht R., Proc . Natl. Acad. Sci. USA Vol 88: 8972-8976 (1991).

In order to determine the three-dimensional structure of TR LBD or LBD from another member of the nuclear receptor phase, it is desirable to co-crystallize the LBD and the corresponding LBD ligand. In the case of TR LBDs, it is preferred that the heavy atoms they contain co-crystallize with different ligands such as T3, IpBr and dimit, respectively. Other TR ligands, such as the ligands known in the art and surrounded by formula (I) described herein, are also used to generate co-crystals of TR LBD and TR ligand. In the compounds included in formula (I), it is generally preferred to use one or more ligands having one or more bromo or iodine substituents in R ₃ , R ₅ , R ₃ ′ or R ₅ ′, preferably such compounds Will have two or more, more preferably three or more such substituents. As described herein, advantageously, such substituents are used as heavy atoms to help solve the state problem for the three-dimensional structure of TR LBD, halogen (eg, I or Br) substituted ligands, In particular, it can be used as generalizing the method of using a ligand for a nuclear receptor.

Conventionally purified LBDs, such as TR LBDs, are homogenized to saturation concentrations of ligands at temperatures capable of fully preserving the protein. Although the receptors tend to be more stable within the range of 2-20 ° C., the ligand equilibrium can be at 2-37 ° C.

Preferably the crystals are prepared by the hanging drop method described in detail herein. Adjusted temperature control is desirable to improve the stability and quality of the crystals. Temperatures of 4 to 25 ° C. are generally used and it is desirable to test crystallization over the temperature range. In the case of TR, a crystallization temperature of 18 to 25 ° C, preferably 20 to 23 ° C, and most preferably 22 ° C is preferred.

3,5,3'-triiodotyronine (T ₃ ), 3'-isopropyl-3,5-dibromotyronine (IpBr ₂ ), 3'-isopropyl-3,5-dimethyltyronine (di Mitt), and complexes of TR-a LBD with various agents, including 3,5,3'-triiodotyroacetic acid (triac), are prepared by the methods described herein. Cocrystals of rTR-α LBD pre-bound with ligands are prepared from 15% 2-methyl-2,4-pentanediol (MPD) by vapor diffusion at ambient temperature. The crystal is radiation sensitive and must be frozen to measure complete diffraction data. On a rotating anode X-ray light source, the crystals are diffracted about 3 μs; Synchrotron radiation extends the separation limit by 2.0 Hz for the T ₃ cocrystal. Compositions of thyroid hormones combined with the ability to prepare and co-crystallize receptors complexed with various analogs allow for abnormal phase matching methods. This phase matching method can be applied to ligands of nuclear receptors by inducing I and Br substitutions of the ligands. In this method, co-crystals of TR LBD containing different hormone analogs (T ₃ , IpBr ₂ , dimit, and triac) at the 3, 5 and 3 ′ positions provide derivatives of the same type. In these analogues, the halogen substituents (2Br and 3I atoms) act as heavy atoms and the dimit cocrystal (3 alkyl group) acts as the parent. Initial 2.5 μs multiple isomorphic substitution / modification scattering / density The altered electron density map allows LBD to be traced from the skeleton generated in the molecular graphics program O5 [Jones, TA et al., ACTA Cryst, 47: 110-119 (1991) ] (Quoted here by reference). The LBD model consists of four fragments, Arg157-Gly184, Trp186-Gly197, Ser199-Pro205, and Val210-Phe405, and is modified in XPLOR using a position transformed and stimulated annealing original. Missing residues are formed with the help of different densities. The final model is refined to R _crystal = 21.8% and R _glass = 24.4% for data from 5.0 to 2.2 ms (see Table 3).

This phase matching method can be applied to ligands of nuclear receptors described herein by causing I and Br substitutions of ligands.

3d structure of TR LBD

Structure of TR LBD

As an example of the three-dimensional structure of the nuclear receptor, the wrinkles of TR-α ₁ LBD are shown in FIG. 4. TR-α LBD consists of a monostructure domain filled in three layers consisting of twelve α-helices, H1-12, and four short β strains, S1-4, forming a mixed β-sheet. The buried hormone and three antiparallel α-helices, H5-6, H9, and H10 form the central layer of the domain as shown in FIG. 4. H1, H2, H3 and S1 form one side of the LBD and the opposite side is formed by H7, H8, H11 and H12. The first 35 amino acids of the N-terminus (Met122-Gln156) are not visible on the electron density map. Heterodimer RXR; The three-dimensional structure of amino acids Met 122-Gln151 of TR DBD, a TR DNA-binding wholesale bound to DNA, allows extensive contact with small grooves of DNA8. Five complex imino acids (Arg152-Gln156) present between the final residue of the TR DBD and the initial visualization residue of the DBD represent an effective “hinge” that connects the LBD and the DBD at the free receptor.

The distinct helix composition and the stratified arrangement of the secondary structure are identical to the arrangement of ligand unbound hRXRα, indicating that there is a common nuclear receptor fold between the two nuclear receptors.

The TR LBD starts at Arg157 and extends to the beginning of H1 in the form of an extended coil. The rotation of the α-helix, H2, covers the hormonal binding cavity and covers S4, the short β-strand, S1, which forms the edge of the β-sheet mixed parallel to the three outermost antiparallel strands. The chain is irregular up to the beginning of H3, which is mostly antiparallel to H1. H3 bends in Ile221 and Ile222 and remains in contact with the ligand.

The chain rotates about 90 ° at the end of H3 to form an incomplete α-helix, H4. The first node core helix, H5-6, extends along the axis altered by torsion near the ligand at Gly 253. Helix consists of most hydrophobic branched chains suspended in two surprising exceptions: Arg262 is an unusable solvent and interacts with ligand carboxylates (1-substituents), and Glu256 is Arg329 and H11 from H9. Encounter polarity with Arg375 from. H5-6 terminates at the short β-strand of the four strand mixed sheet, ie S2. S3 and S4 are connected in a left turn and are further connected by a salt bridge between Lys284 and Asp272. Following S4, H7 and H8 form L stabilized by the salt bridge between Lys268 and Asp272. The rotation between H7 and H8 takes an abnormal form as a result of the interaction of the ligand with its glycine rich sequence. H9 is a second core helix antiparallel to adjacent H5-6. In addition, two buried polar branched chains, Glu315 and Gln320, are found. Glu315 together with His358 and Arg356 form a buried salt bridge. The oxygen of Gln320 forms a hydrogen bond with the buried branched chain of His175. The chain is then reversed again to form H10, which is antiparallel to H9. H11 extends diagonally across the entire length of the molecule. Immediately after H11, the chain forms a Type II rotation about 90 ° with respect to H11. The chain then rotates again to form H12, which is loosely charged to H3 and H11 as part of a hormone or ligand binding cavity. The last five amino acids at the C-terminus, Glu406-Val410, are disordered.

Ligand binding cavity of TR LBD as an example of a buried ligand cavity of nuclear receptor

The three-dimensional structure of TR LBDs explains the surprising discovery that ligand binding cavities of LBDs are inaccessible solvents when T3 or its identical conformation binds to LBDs. These surprising results could explain a new model of the three-dimensional structure and function of nuclear receptors, and this new model is an application of methods for designing nuclear receptor synthetic ligands containing extended positions that inhibit the normal activation of activation domains. And identifying methods using computers for ligand design.

Dimit, ie, the ligand bound to the receptor, is the same conformation of T ₃ and is a thyroid hormone agonist. Thus, the binding of dimit should reflect the binding of T ₃ and the dimit binding receptor is expected to be the active form of TR. The ligand is embedded in the receptor, providing a hydrophobic core for the subdomain of the protein as shown in FIGS. 5A and 5B. H5-6 and H9 contain hydrophobic cores for the remaining receptors.

The extending coupling cavity consists of several structural elements. The cavity is surrounded by H5-6 (Met 256-Arg266) at the top, the bottom is surrounded by a loop (Leu287-Ile299) sandwiched between H7 and H8 and the sides are H2 (185-187), S3 and S4 Rotation between (Leu276-Ser277), H3 (Phe215-Arg228), H11 (His381-Met388) and H12 (Phe401-Phe405). The common volume formed by these elements is essentially the volume of the hormone (530 mm 3) as calculated by GRASP (Columbia University, United States of America (600 m 3)). The remaining volume is filled by water molecules surrounding the amino propionic acid substituents. FIG. 6 shows various contacts (or interactions) between TR LBD and ligand.

The plane of the inner and outer rings of the ligand (prime ring) is rotated about 60 ° from the plane relative to each other, forming a 3'-terminal form, where the 3 'substituents of the outer ring protrude downward and away from the inner ring. . The amino-propionic acid and outer phenol rings each take the trans form on both sides of the inner ring. The torsion angle (X ₁ ) for amino-propionic acid is 300 °.

The amino-propionic acid substituents are loosely packed in polar pockets formed by branched chains from H2, H4 and S3. The carboxylate group forms a hydrogen bond directly with the guanidinium group of Arg228 and amino N of Ser277. Arg262, Arg266 and Asn179 also interact with carboxylates through water mediated hydrogen bonding. Three arginine residues generate markedly positive local electrostatic potentials, which can stabilize the negative charge of the carboxylates. Hydrogen bonds are not formed by amino nitrogen. Amino-contributing interaction propionic acid substituent are consistent with the fact that triac, which lacks the amino nitrogen, has the same binding affinity and binding of the T ₃ affinity, long aliphatic chain of amino nitrogen and T ₃ is gender binding affinity It does not.

Diphenyl ether, on the other hand, is embedded in the hydrophobic core. The inner ring is filled in the hydrophobic pocket formed by H3, H5-6, and S3. The pockets for 3-methyl and 5-methyl substituents are not fully filled. The reason is expected to be that the van der Waals radius of the methyl substituent on dimit is smaller than the iodine substituent provided by the thyroid hormone T ₃ . Such pockets typically range from 25 to 100 square ohms strong (though smaller pockets for the substituent range from 40 to 80 square ohms) and may be more densely filled with chemical substituents that fit well as described herein.

The outer ring is densely packed in the pocket formed by the loops between H3, H5-6, H7, H8, H11 and H12, and H7 and H8. Ether oxygen is found in the hydrophobic environment formed by Phe218, Leu287, Leu276, and Leu292. The absence of hydrogen bonds to the ether oxygens is evidence of the correct stereochemistry of the phenyl ring, as shown by the structurally similar bonds and possible bonds with hormonal analogs with reduced or negligible hydrogen bond ability. Match the role. 3'-isopropyl substituents contact Gly290 and 291. The presence of glycine at this location in the pocket makes it possible to explain the observed relationship between the activity and the size of the 3'-substituent. Activity is maximal for 3'-isopropyl and is reduced by the amount added. Only hydrogen bonds in the hydrophobic cavity are formed between phenolic hydroxyl and His381 Ne2. His381 is stabilized by filling the contacts provided by Phe405 and Met256.

The presence of a 5 'substituent larger than hydrogen affects the binding affinity for hormones. The broader thyroid hormone, ie 3,5,3 ', 5'-tetraiodo-L-tyronine (T ₄ ) contains iodine at this position and binds the receptor with 2% affinity of T ₃ Let's do it. The structure suggests that the distinction to T ₄ is made through a combination of possible limitations imposed by the steric impact by Met256 and the structure of hydrogen bonding from His381 to phenolic hydroxyl. The 5 ′ position is a preferred region for introducing chemical changes of CH at the 5 ′ position of the T3 and / or TR agonist as described herein extending from the prime ring and producing an antagonist or partial agonist.

Deletion and antibody competition studies suggest the involvement of residues Pro162 to Val202 in ligand binding. The region is filled with residues that form polar pockets where H2 interacts with amino-propionic acid groups, but the region is not in direct contact with the hormone in the bound structure. One role for H2 is to stabilize these residues in a bound state. H2 with β-strands S3 and S4 may exhibit a predominant entry point to the ligand because the amino-propionic acid of the ligand is oriented towards this region. Studies of receptors that bind to the T ₃ affinity matrix demonstrate that only binding to amino-propionic acid is maintained, indicating that steric hindrance present in other bindings prevents binding. The crystallographic temperature factor also suggests that the coil and β-strand regions are the most flexible portions of the domain Figure 7. Partial involvement of this region in binding hormones, namely the hinge domain between DBD and LBD, Structural means for ligands that bind to functional DNA binding may be provided, because part of the hinge domain is in contact with DNA.

TR LBD Transcription Activation Helix As an Example of Nuclear Receptor Activation Domain

In addition to the surprising discovery that ligand binding cavities are not available solvents when loaded with ligands, the activating helix of TR LBD shows a surface for ligand cavities for activation between ligands bound to one or more amino acids. The C-terminal 17 amino acid of TR, called activating helix or AF-2 (example of LBD activation domain), is involved in mediating hormone dependent transcriptional activation. Although mutations of key residues in the domain reduce ligand dependent activation, it is unclear to the present invention whether such mutations directly affect ligand coordination. Although some mutations in these domains have been known to reduce or delete ligand binding, other mutations at more distant sites of LBD also have similar effects.

The activation domain in the nuclear receptor exhibits a similar three-dimensional relationship to the binding cavity, the region of LBD that binds to the molecular recognition domain of the ligand. That is, the activation domain represents part of itself for the binding cavity (essentially the molecular recognition domain of the ligand). Many nuclear receptors are expected to have domains including retinoid receptors, RAR, and RXR, glucocorticoid receptor GR, and estrogen receptor ER. Based on the sequence of the TR, the domain is proposed to accept an amphipathic helix structure. β-sheets or mixed secondary structures can be represented as activation domains at less relevant nuclear receptors.

Within the activation domain, the very well conserved motif ΦΦXEΦΦ, where Φ represents a hydrophobic residue, is proposed to mediate the interaction between the receptor and the transcriptional coactivator. Hormonal dependence caused several proteins to bind to TR. As one of these proteins, Trip1 is involved in putative auxiliary activator Sug1 and interacts with a subset of the C-terminal activation domain and basic transcriptional tissue, suggesting a role in trans activation by TR. RIP140, SRC1, Onate, S.A. et. al., Science 270: 1354-1357 (1995)] and TF-1 (see, eg, Ledouarim, B., et. al., EMBO J. 14: 2020-2033 (1995)] interact with other nuclear receptors via ligand-dependent methods via the C-terminal domain. Binding of such proteins can be altered using the TR ligands described herein, particularly TR ligands with extensions that stericly inhibit the interaction between very well conserved motifs and other proteins.

The C-terminal activation domain of TR forms the amphipathic helix, H12. This H12 is loosely positioned relative to the receptor to form part of the hormone binding cavity. Helix is filled with a hydrophobic moiety, and a charged moiety, directed inwards against the hormone binding cavity, including a very well conserved glutamate extending into the solvent as shown in FIG. 8. The activating helix of TR LBD shows Phe401 for the ligand binding cavity and is directly coordinated with the hormone. That is, such amino acids form van der Waals contacts with the plane of the outer phenyl ring. Phe 405 also interacts with His381, possibly stabilizing its hydrogen bond form, ie, the desired hydrogen bond interaction. The involvement of Phe401 and Phe405 in binding hormones explains how mutations of these residues reduce hormone binding affinity. In addition, the effect of these mutations on activation appears to be derived from the role of stabilizing domains within the bound structure via reduced hydrogen bond interactions of bipolar interactions. Glu 403 extends into the solvent to reinforce its important role in the interaction. In the observed form of Glu 403 present on the surface as an ordered residue against the background of a significantly hydrophobic surface, Glu 403 may interact with the activator protein shown in FIG. 9 and described herein. Other charged residues, Glu 405 and Asp 406, are irregular as helix collisions in Phe 405.

The other two sequences in TR, γ2 and γ3 activate transcription when expressed with the DNA binding domain as a fusion protein. This sequence found in TRB corresponds to TR-α residues Pro158-Ile168 at H1 (γ2) and Gly290-Leu319 at H8 and H9 (γ3). Unlike the C-terminal activation domain, γ2 and γ3 do not show molecular structural units in the rat TRα LBD and do not show a surface for protein-protein interactions: the important aspartate / glutamate residues of γ3 are two separate Located on the helix and not forming a single surface; Charged residues of γ2 interact with the residues of LBD in ion pairs. Thus, γ2 and γ3 do not act as activation domains in terms of total receptors.

How to design nuclear receptor LBD ligands using a computer

Identifying the three-dimensional structure of the nuclear receptor ligand binding domain provides an important and useful method for designing ligands for nuclear receptors using the computerized methods described herein. Referring to the figures, it can be determined that nuclear receptor ligands are bound within binding cavities in LBD that are inaccessible to water, and that chemical residues can be added at desired positions on the ligand. These chemical changes, usually extensions, can fill the binding cavities shown in the figures for denser packing (or less water), or contact with the ligand before chemical changes are introduced or appear in the three-dimensional model of the LBD. It may be used in contact with or cleaving amino acids, rather than in place thereof. Ligands that interact with the nuclear phase and members may act as agonists, antagonists and partial agonists, depending on which ligand-induced conformational changes are performed.

Agents induce changes in receptors that cause them to be in active form, affecting transcription either positively or negatively. There may be several different ligand induced changes in receptor form.

Antagonists bind to the receptor but do not induce morphological changes that alter the transcriptional regulatory properties or physiologically relevant forms of the receptor. Binding of the antagonist may block the binding of the agent and hence its action.

Partial agents bind only to the receptor and induce only partial changes in the receptor induced by the agent. The difference can be qualitative or quantitative. Thus, partial agents may induce some morphological changes induced by the agent, but do not induce morphological changes induced by others, or partial agents may induce specific changes to a limited extent.

Ligand-induced Morphological Change

As described herein, the receptor to which no ligand is bound is present in an inactive or slightly active form or in a form that exhibits inhibitory activity. Binding of the agonist ligand induces morphological changes in the receptor, thereby making the receptor more active, stimulating or inhibiting the expression of the gene. Receptors can exhibit non-genomic action, with some known forms of change and / or their subsequent forms described herein.

Heat shock protein binding

For many nuclear receptors, ligand binding induces degradation of the heat shock protein, allowing the receptor to form mostly dimers, after which the receptor binds to DNA and regulates transcription.

Nuclear receptors generally represent regions that bind to LBD and have a heat shock protein binding domain that can be changed by binding a ligand to LBD. As a result, extended chemical residues (or more) from ligands that stabilize the binding or contact of LBD with the heat shock protein binding domain can be designed using the computer-assisted methods described herein to generate partial agonists or antagonists. have. Typically extended chemical moieties may extend beyond the molecular recognition domains on the ligand and generally pass through the buried binding cavities of the ligand.

Dimerization and Heterodimerization

For receptors associated with hsp in the absence of ligand, dissociation of hsp dimerizes the receptor. Dimerization is due to receptor domains in DBD and LBD. The main stimulus of dimerization is the dissociation of hsp, but ligand-induced morphological changes at the receptor can have further facilitating effects. For receptors that are not associated with hsp in the absence of ligand, particularly TR, ligand binding can affect the pattern of dimerization / heterodimerization. This effect depends on the DNA binding site and may also depend on promoters associated with other proteins that can interact with the receptor. The common pattern should inhibit monomer formation while favoring heterodimer formation over dimer formation on DNA.

Nuclear receptor LBDs generally have a dimerization domain that provides a region for binding to other nuclear receptors and can be altered by binding a ligand to LBD. As a result, extended chemical residues (or more) from ligands that disrupt the binding or contact of dimerization domains can be designed using the computer-assisted methods described herein to generate partial agonists or antagonists. Typically extended chemical moieties may extend beyond the molecular recognition domains on the ligand and generally pass through the buried binding cavities of the ligand.

DNA binding

In nuclear receptors that bind to hsp, hsp is ligand-induced dissociation while successively forming dimers and promotes DNA binding. For unrelated receptors (as in the absence of ligands), ligand binding stimulates DNA binding of heterodimers and dimers and inhibits monomer binding to DNA. However, for DNA containing a single half site, ligands tend to stimulate the binding of the receptor to DNA. This effect is moderate and depends on the nature of the DNA site and on the presence of other proteins that can interact with the receptor. Nuclear receptors generally have a DBD (DNA binding domain) that provides a region for binding to DNA and can be regulated by binding the ligand to LBD. As a result, chemical residues (or more) extended from ligands that disrupt the binding or contact of DBDs can be designed by the methods using the computer described herein to generate partial agonists or antagonists. Typically extended chemical moieties may extend beyond the molecular recognition domains on the ligand and generally pass through the buried binding cavities of the ligand.

Inhibitor binding

Receptors not associated with hsp in the absence of ligand often act as transcription inhibitors in the absence of ligand. This action appears to be partly due to transcription inhibitor proteins that bind to the receptor's LBD. Agent binding induces dissociation of these proteins from the receptor. This action mitigates the inhibition of transcription and results in the transcriptional trans activation function of the receptor.

Trantrans Activation Function

Ligand binding induces transcriptional activation in two basic ways. The first method is by dissociation of hsp from the receptor. This degradation, which involves the binding of receptors to DNA or other proteins in nuclear chromatin and subsequent dimerization of these receptors, results in transcriptional regulatory properties of the receptors. This function may in particular be a function on the amino terminus of the receptor.

The second method is to change the receptor and contact it with other proteins involved in transcription. Such a protein may be a protein that directly or indirectly interacts with an element of a terminal promoter or a protein of a terminal promoter. In addition, the interaction may be via other transcription factors that interact directly or indirectly with proteins of the terminal promoter. Several different proteins have been described that bind to receptors in a ligand dependent manner. In addition, in some cases, ligand-induced morphological changes do not affect the binding of other proteins to receptors, but affect their ability to regulate transcription.

Nuclear receptors or nuclear receptor LBDs generally have an activation domain that provides a region for binding to DNA and can be altered by binding a ligand to LBD. As a result, a chemical moiety (or more) extending from a ligand that disrupts the binding or contact of an activation domain can be designed by the methods described herein using the computer to generate partial agonists or antagonists. Typically extended chemical moieties may extend beyond the molecular recognition domain on the ligand, generally beyond the buried binding cavity of the ligand and in the direction of the activation domain, which activation domain is often ground or more typically Is a helix as it appears in the three-dimensional model shown in the figure in two dimensions on a computer screen.

Ligand-induced Morphological Change

Plasma proteins bind hormones without morphological changes through static binding pockets formed between monomers or domains. For example, tetrameric thyroid binding plasma protein transtratin forms a solvent accessible hormone binding channel at the oligomeric interface. The structure of the protein does not change upon binding with the hormone in the generation of a buckled binding cavity with ligand binding.

However, it is demonstrated that the structural role for ligands bound to nuclear receptor LBD, such as rat TR-α LBD, may differ in the bound and unbound states of the receptor. In the absence of hormones, the receptor may have a cavity in its core, ie, a nonspecific globular protein. Ligands (eg, hormones) form a hydrophobic core of the active receptor after binding to the nuclear receptor. Ligand binding by receptors is a dynamic process and modulates receptor function by inducing altered forms.

An accurate description of hormone-induced morphological changes should compare the structure of ligand-bound and non-ligand-bound TRs. The structure of unligand bound human RXRα can be substituted as a model for unligand bound TR. Rat TR-α LBD and human RXRα LBD form similar wrinkles, and structural similarity extends to morphological changes after ligand binding.

There are three main differences between the two structures that are believed to be the result of ligand binding. First, the bound rat TR-α LBD structure is more compact and hormones are densely packed in the hydrophobic core of the receptor. In contrast, unligand bound human RXRα LBD contains several internal hydrophobic cavities. The presence of each cavity is not common in wrinkled proteins and reflects the ligand unbound state of the receptor. Two of these cavities have been proposed as possible binding sites for 9-cis retinoic acid, although only a few of these sites partially overlap with a single buried binding cavity observed in ligand-bound rat TR-α LBD.

The second difference is associated with H11 in rat TR-α LBD which contributes to some hormone binding cavities. Continuous H11 in rat TR-α LBD breaks down at Cys 432 in RXR to form a loop between H10 and H11 in hRXRα. This residue corresponds to His381 in TR which provides hydrogen bonding to the outer ring hydroxyl of the ligand. In addition, rhmone binding cavities filled by ligand in rat TR-α LBD are interrupted at hRXRα by the same loop to form separate hydrophobic pockets in RXR with H6 and H7. In bound rat TR-α LBD, the corresponding helix H7 and H8 are adjacent to the binding pocket and surround the hormone binding cavity from below.

The third difference between the two receptors is the location of the C-terminal activation domain. The C-terminal activation domain forms α-helices at both receptors, but the domains in the rat TR-α LBD have proline rich rotation and are located relative to the receptor and contribute to a portion of the binding cavity. On the other hand, the activation domain in hRXRα with no ligand bound is part of the helix longer than protruding into the solvent.

This difference leads to a model for another form of TR LBD assumed in the absence of ligand. In TR, where the ligand is not bound, the subdomain of the receptor surrounding the hormone binding cavity is loosely charged in the binding cavity occluded by H11, which is not partially constructed, providing a partial core to the receptor.

When bound to the hormone, the moiety that forms the coil in the unbound receptor binds to the ligand and leads to H11. The ordering of H11 does not block the hydrophobic cavities, allowing H7 and H8 to interact with the hormone. The elongated hydrophobic cavity then breaks around the hormone to create a tightly bound structure.

It will be possible to predict ligand induced morphological changes in the C-terminal activation domain that depend in part on the extended structure in ligand unbound TR that is recharged upon ligand binding. Ligand-induced morphological changes resulted in the amino acid sequence of rat TR-α in rotation (393-PTELFPP-399) reducing the propensity of the peptide chain of rat TR-α to form α-helix, resulting in more refill volume. It can be difficult to distinguish because it can be done in small chains.

After ligand-induced morphology has occurred, the shape of the C-terminal activation domain in the bound structure changes the charge compared to the unbound form of the receptor. Binding of the ligands improves the stability of the activation domain. Activation domains are loosely charged even in bound structures, as measured by the distribution of charge interactions over the entire LBD. The packing density of the activation domain, defined as the number of atoms within 4.5 kV, is 1.5 standard deviations below the mean. By comparison, another surface helix, H1, is less than 0.5 standard deviation of the mean, and the poorest-filled structural parts, ie, irregular coils from residues Ile196-Asp206, exhibit 2.0 standard deviations below the mean. In addition, the main charging contact for the C-terminal domain at the bound receptor is provided by the residue itself or the ligand itself, which interacts with the ligand, such as His381. The shape of these residues can be expected to differ between the bound and unbound receptors and is expected to be different by stretching the shape of the C-terminal activation domain that depends on their interaction. Without the stabilization provided by the bound ligand, the C-terminal activation domain is disordered before hormone binding.

Correlation of ligand induced morphological changes is evident as described herein. For example, His381 from H11 and Phe405 from H12 interact within the bound structure to provide specific hydrogen bonds to the phenolic hydroxyl. Ligand-induced changes affecting H11 and H12 are enhanced to form a tightly bound state.

Calculation method using ligand extension and three-dimensional model

The three-dimensional structure of ligand bound TR receptors has not been suggested and would be of great help in the development of new nuclear receptor synthetic ligands such as thyroid receptor antagonists. In addition, the superfamily of such receptors is well suited to conventional methods, including three-dimensional structural identification and combinatorial chemical analysis as disclosed in EP 335 628, US Pat. No. 5,463,564, which is incorporated herein by reference. The availability of the receptor allows for structural measurements using X-ray crystallography. Computer programs using crystallographic data in the practice of the present invention will enable the rotational design of ligands to these receptors. Programs such as Rasmol can be used for atomic coordination from crystals produced by carrying out the invention, or for carrying out the invention by forming three-dimensional models and / or measuring structures associated with ligand binding. Computer programs such as INSIGHT and GRASP can realize additional operations and possibilities for introducing new structures. In addition, large amounts of binding and bioactivity assays can be derived by using purified recombinant proteins and novel reporter gene transcription assays described herein and known in the art to measure the activity of CDL.

In general, computational methods for designing nuclear receptor synthetic ligands include the following two steps:

1) determining that the amino acid (s) of nuclear receptor LBD interacts with the first chemical portion (one or more) of the ligand using a three-dimensional model of the crystallized protein comprising nuclear receptor LBD having a binding ligand; And

2) reducing the interaction between the interacting amino acid and the second chemical moiety as compared to the interaction between the interacting amino acid and the first chemical moiety by selecting a chemical modification (one or more) of the first chemical moiety Creating a second chemical moiety having an enhancing structure.

As presented herein, the interacting amino acid forms a contact with the ligand and the center of the atom of the interacting amino acid is generally 2-4 kW (angstroms) away from the center of the atom of the ligand. Typically this distance is determined by a computer, as described herein and in McRee, 1993, but the distance can be determined manually once the three-dimensional model is manufactured. Examples of interacting amino acids are described in Appendix 2. See also the following literature on stereochemistry of three-dimensional models (Wagner et al., Nature 378 (6558): 670-697 (1995)). More typically, the atoms of the ligand and the atoms of the interacting amino acids are 3 to 4 mm apart. The present invention can be practiced by repeating steps 1 and 2 to purify the alignment of ligands in LBD and determine better ligands such as agents. As shown in the figure, the three-dimensional model of TR is expressed in two dimensions to determine the amino acid contacting the ligand and to determine the position on the ligand for the change in interaction with a specific amino acid as compared to the chemical and pre-chemical interactions. You can choose. Chemical modifications can be made using computers, manually using two-dimensional representations of three-dimensional models, or by chemically synthesizing ligands. Three-dimensional models can be manufactured using Appendix 2 and the drawings. As in further steps, three-dimensional models can be prepared using atomic coordinates of nuclear receptor LBD from crystallized proteins known in the art. See McCree (1993), herein incorporated by reference.

The ligand can also interact with distant amino acids after chemical modification of the ligand to form new ligands. Remote amino acids are generally not in contact with the ligand prior to chemical modification. Chemical modifications can alter the ligand's structure to prepare as a new ligand that interacts with distant amino acids, typically at least 4.5 kHz away from the ligand. Often the distant amino acids will not align the surface of the binding cavity to the ligand because they are too far from the ligand to be part of the pocket or surface of the binding cavity.

Interactions between atoms of LBD amino acids and atoms of LBD ligands can be made by any force or attraction described naturally. In general, interactions between atoms of amino acids and ligands will result in hydrogen bond interactions, charge interactions, hydrophobic interactions, van der Waals interactions or dipole interactions. In the case of hydrophobic interactions it is recognized that this is not part of the interaction between the amino acid and the ligand itself, but rather as part of the repulsive force of other hydrophilic groups or water from the hydrophobic surface. Reduction and increase in the interaction of LBD and ligand can be measured by calculating or testing binding energy computationally or by using thermodynamic or kinematic methods known in the art.

Chemical modifications will often increase or decrease the interaction between atoms of LBD amino acids and atoms of LBD ligands. Steric hindrance will be a common means of changing the interaction of LBD binding cavities with active regions. Chemical modifications are preferably induced at the C-H, C- and C-OH positions in the ligand, where carbon is the portion of the ligand structure that remains the same after modification is complete. In the case of C-H, C may have 1, 2 or 3 hydrogens, but generally only one hydrogen will be replaced. H or OH is removed after the modification is complete and replaced with the desired chemical moiety.

Since thyroid receptors are components of the larger superfamily of hormone-binding nuclear receptors, rules in the development of agonists and antagonists will be recognized by those skilled in the art as useful in designing ligands in the entire superfamily. Examination of the structures of known and antagonists of estrogen and androgen receptors supports most of the antagonist mechanisms of action as shown in FIG. 10.

Overall folding of the receptor based on the comparison of the recorded structure of unliganded RXR with amino acid sequences of other phases and components indicates that the total folding of receptors with the phases is similar. As such, it is expected from structures where there is a general pattern of folding of nuclear receptors around agonist or antagonist ligands.

The three-dimensional structure of a nuclear receptor with ligand binding allows the binding cavity to align the agent, in particular the agonist molecule recognition domain, and the antagonist has a chemical structure that extends beyond the ligand, in particular the agent, and inhibits the folding of the receptor around the ligand. As a result they form buried binding cavities or other groups that have the same effect, resulting in a clear observation that the nuclear receptor is polled around the agonist ligand. The expansion position can affect the folding in various ways as indicated by the structure. Such expansion on antagonists is shown in FIG. 10 for various receptors compared to the corresponding agents.

For example, expansion to the carboxy terminal activated helix affects the filling / folding of the helix into the receptor body. This includes the amino terminus of the receptor, which can interact directly or indirectly with the carboxy terminus trans activation domain (including Helix 12), or the transcriptional transactivation action on the opposite end of the linear receptor, interacting with other parts of the receptor or other proteins. May affect the ability of said portion of the nuclear receptor to do so. Expansion in this direction can also affect the filling of TR helix 11 (or similar helix in nuclear receptors) into the body of the receptor and can optionally affect the dimerization and heterodimerization of the receptor. Extension towards helix 1 can affect the relationship between the DNA binding domain and the hinge region of the receptor having a ligand binding domain and optionally or additionally interacts with the receptor binding to DNA and / or said region of the receptor May affect the interaction of proteins with receptors. Other expansions towards Helix 11 may affect dimers by affecting the charging of Helix 11 and Helix 1 and 10. Such chemical modifications can be assessed through the computerized methods described herein. Also in some cases, the extension may protrude through a receptor that is otherwise or completely incompletely folded around the ligand. The overhang extension may be a three-dimensional blockade of interaction with a coactivator or other protein.

A three-dimensional structure with a ligand embedded in a binding cavity simply describes a nuclear receptor having a binding cavity containing a hinge and a lead consisting of one or more structural elements that move to accumulate and surround the ligand. Ligands for TR can be modified on specific sites with specific classes of chemical groups that act to leave the lead and hinge regions in an open, partially open or closed state to achieve partial agonist or antagonist action. In this state, the biological response of the TR is different so that the structure can be used to design specific compounds with the desired effect.

Knowledge of the three-dimensional structure of the TR-T ₃ complex leads to a general model for agonist and antagonist design. An important new feature of the structural data is that the T ₃ ligand is completely embedded in the central hydrophobic core of the protein. Other ligand-receptor complexes belonging to the nuclear receptor superfamily have similarly embedded ligand binding sites so this model will be useful for agonist / antagonist design for the entire superfamily.

If the design of an antagonist is desired, the necessary binding affinity is achieved either through preservation of the major binding contact of the natural hormonal agent or through the interaction of extensions with receptor regions while incorporating "extension groups" that interfere with the normal workings of the ligand-receptor complex. Needs to be generated.

The model applied to the antagonist design and described herein is referred to as an “extension model”. Antagonist compounds for nuclear receptors must contain identical or similar groups that promote high affinity binding to the receptors, and in addition, the compounds must contain branched chains that can be large and / or polar. This branched chain may be an actual extension, given in a large state, or a side group with a different charged action than the agent ligand. For example substitution of CH ₂ OH with CH ₃ at position 21, and denaturation of the cortisol to the keto group at position 11 results in glucocorticoid antagonist activity. Robsseau, GG, et al., J. Mol. Biol. 67: 99-115 (1972). However, in most cases an effective antagonist has a larger extension. As such, antiglucocorticoid (and antiprogestin) RU486 contains a bulk side group at the 11 position. See, Horwitz, KB Endocrine Rev. 13: 146-163 (1992). The antagonist compound will then bind within the embedded ligand binding site of the receptor with a reasonable high affinity (100 nM), but the expansion action will prevent the receptor-ligand complex from adopting the necessary form required for transcription factor action. Antagonism (which can occur in agonists or antagonists) prevents homo / heterodimer formation in HRE, prevents coactivator binding to receptor monomers, homodimers or homo / heterodimers, or ligands to HRE. It can manifest itself at the molecular level in a number of ways, including by a combination of these effects that prevents the transcription of hormone reactive genes mediated by an inducing effect. There are several antagonistic compounds for nuclear receptors in the prior art in which antagonism can be explained by the expansion hypothesis [Horwitz, KB, Endocrine Rev., 13: 146-163 (1992), Raunnaud JP et al. , J. Steroid Biochem. 25: 811-833 (1986), Keiel S., et al., Mol. Cell. Biol. 14: 287-298 (1994). These compounds are shown in FIG. 10 along with their agent controls. Each of these antagonists contains large dilators attached to the agent or agent like core structure. Importantly, these antagonist compounds were discovered by chance and were not designed using structural action hypotheses such as the expansion principle.

One method of designing a thyroid antagonist using the extended hypothesis is provided below as a teaching example. The three-dimensional structure of the TR-αdimit complex combined with the prior art, in particular the structural activity data disclosed in the references herein, can be used to organize the ligand-receptor interactions that are most importantly involved in high affinity ligand binding. Physical pictures of these interactions are shown in FIG. 6. The figure illustrates an isolated essential contact for ligand binding. Since the ligand is embedded in the center of the receptor, a structure with a space between these isolated interactions is also important. As such, the present knowledge of the inventors of such systems dictates that, for example, a newly designed ligand for a receptor must contain two substituted aryls bound by a tyronine structural backbone, or one atomic spacer.

The general structure for an antagonist designed by the expansion hypothesis is illustrated by the general description of substituents of the following TR antagonist (associated with formula (1)): R1 is a carboxylate, phosphonate, phosphate, sulfate or sulfite It may have an anionic group and is linked to a ring having 0 to 3 atom linkers, preferably 2 carbon linkers, comprising at least one C, O, N, S atom. R 1 may be optionally substituted with an amine (eg —NH ₂ ). R3 and R5 are small hydrophobic groups such as -Br, -I, or -CH ₃ . R3 and R5 may be the same substituent or different substituents. R ₃ ′ may be a hydrophobic group which may be larger than for R 3 and R 5, such as —I, —CH ₃ , —isopropyl, —phenyl, —benzyl, 5 and 6 ring heterocyclics. R ₄ ′ is a group that can contribute to hydrogen bonding as a donor or acceptor. Such groups include -OH, -NH ₂ , and -SH. R ₅ ′ is an important expander that can make this compound an antagonist. R ₅ ′ is a long chain alkyl (eg straight or branched chain having 1 to 9 carbon atoms), aryl (benzyl, phenyl and substituted benzyl and phenyl rings (eg halogen, alkyl (1 to 5 carbon atoms)) , And may or may not be linked to the ring by -CH _2- , a heterocycle (eg, 5 or 6 atoms, preferably 5 carbons and 1 nitrogen, or 5 carbons) , They are polar (eg -OH, NH ₂ and -SH), cations (eg -NH ₃ , -N (CH ₃ ) ₃ ), and anions (carboxylates, phosphonates, phosphates or _May or may not include sulfate groups: R ₅ ′ is also polar (eg, —OH, NH ₂ and —SH), cation (eg, —NH ₃ , —N (CH ₃ ) ₃ ) And an anionic (carboxylate, phosphonate, phosphate or sulfate) group X is a spacer group suitably located on two aromatic rings. In contrast, 1 such as O, S, SO, SO ₂ , NH, NZ, where Z is alkyl, CH ₂ , CHOH, CO, C (CH ₃ ) OH and C (CH ₃ ) (CH ₃ ). Is a spacer of four atoms R 2, R 6, R 2 ′ and R 6 ′ may be —F, and —Cl and are preferably H.

TR ligands can also be described as substituted phenylated 3,5 diyodo tyrosine with substituted R5 'and R3' groups. R 5 ′ represents long chain alkyl (eg, straight or branched chain having 4 to 9 carbon atoms), aryl (benzyl, phenyl and substituted benzyl and phenyl rings (eg halogen, alkyl (1 to 5 carbon atoms)); And, connected or unlinked to the ring by -CH _2- , heterocycle (eg, 5 or 6 atoms, preferably 5 carbons and 1 nitrogen, or 5 carbons), They are polar (eg -OH, NH ₂ and -SH), cations (eg -NH ₃ , -N (CH ₃ ) ₃ ), and anions (carboxylates, phosphonates, phosphates or sulfates) May or may not include groups: R 5 ′ may also be polar (eg, —OH, NH ₂ and —SH), cation (eg, —NH ₃ , —N (CH ₃ ) ₃ ), and And an anionic (carboxylate, phosphonate, phosphate or sulfate) group, R 3 ′ is —IsoPr, halogen, —CH ₃ , alkyl (1-6 carbon) or aryl (benzyl, phenyl and Ringed benzyl and phenyl rings (e.g., halogen, alkyl having 1 to 5 carbon atoms, connected or not connected to the ring by -CH _2- ), heterocycles (e.g., 5 or 6 atoms, Preferably 5 carbons and 1 nitrogen, or 5 carbons, these are polar (eg -OH, NH ₂ and -SH), cations (eg -NH ₃ , -N ( CH ₃ ) ₃ ), and anionic (carboxylate, phosphonate, phosphate or sulfate) groups, with or without groups.

TR antagonists may also be modified T ₃ agonists (with diphenyl structures), wherein R ₅ ′ is alkyl, aryl, 5- or 6-membered heterocyclic aromatic, heteroalkyl, heteroaryl, arylalkyl, hetero Aryl alkyl, polyaromatic, polyheteroaromatic, polar or charged groups, wherein R ₅ ′ may be substituted with polar or charged groups. R ₅ ′ is defined as described herein.

Using this method, the ligand of this example preferably has the following properties:

1. Compounds should be bound to TRs having high affinity (eg 100 nM).

2. The compound must bind to the receptor in the same basic orientation as the natural hormone.

3. The expanding group R5 'must protrude towards the activating helix of the receptor (C terminal helix).

4. A suitable substituent at R ₅ ′ must disrupt the activating helix from its optimal local structure required for mediated transcription.

Antagonists can also be designed with multiple extensions to block more than one side of the folding at any time.

TR ligands (eg, hyperagents) can be designed (and synthesized) to enhance the interaction of one or more amino acids with one or more chemical moieties on the ligand molecule recognition domain. One method is to enhance the charge and polar interactions by replacing the carboxylate of T ₃ (R1 region) with phosphonates, phosphates, sulfates or sulfites. This enhances the interaction with Arg 262, Arg 266 and Arg 228. The interaction of one or more chemical moieties with one or more chemical moieties on the ligand molecule recognition domain can also be enhanced by increasing the size of the R1 group to fill the space filled by water when the dimit is bound (associated with R1). . Preferably the groups have hydrophobicity to complementary charged and bonded steels.

Another way to improve the interaction of one or more chemical residues with one or more amino acids on the ligand molecule recognition domain is to limit the form of the dihedral angle between the two phenyl rings of the tyronine ligand in solution. The plane of the two phenyl rings in the solution is at right angles with two angles of 90 °. In the TR demit structure, the dihedral angle is close to 60 °. TR ligands that fix the angle between two phenyl rings will make the bond stronger. The ligand can be prepared by linking the R6 'and R5 positions of tyronine or substituted tyronine diphenyl. The size of the cyclic linkage can fix the angle between two phenyl rings. Specifically with respect to formula (1), the following cyclic modifications are preferred: 1) R ₅ is linked to R ₆ ′, 2) R ₃ is linked to R ₂ ′ or 3) R ₅ is linked to R ₆ ′ And R ₃ is connected to R ₂ ′. The linkage can be made by alkyl or heteroalkyl chains having 1 to 6 atoms and preferably 2 to 4 carbon atoms or other atoms. All positions of the heteroalkyl chain may be N, O, P or S. S and P heteroatoms along the heteroalkyl chain are any of their possible oxidation states. N hetero atoms or carbons along an alkyl or heteroalkyl chain may have one or more Z substituents, where Z is alkyl, heteroalkyl, aryl, heteroaryl, 5- or 6-membered heterocyclic aromatic. Such compounds are subject to claim, provided that Formula (1) does not include any compounds in the prior art for the filing date of this application.

The interaction of one or more chemical residues with one or more amino acid residues on the ligand molecule recognition domain also fills the unfilled space between the TR ligand and LBD in the region of bridging oxygen (eg, oxygen at T3, triac or dimit). By selecting chemical modifications. As such, somewhat larger residues replacing ether oxygen may enhance binding. The linker may be a mono- or geminal-disubstituted carbon group. In addition to small, hydrophobic groups on disubstituted carbon, groups that are about the same size as oxygen but have greater hydrophobicity are preferred.

TR-α and TR-β selectivity for thyroid hormone receptors

Ligands that selectively bind to alpha TR rather than beta TR can be designed using the methods described herein. The X-ray crystallographic structure of the TR-α LBD of the rat provides identification into the design of the ligand.

The three-dimensional structure shows that the main difference between TR-α and TR-β in the ligand binding cavity is at amino acid Ser 277 (having the side group -CH ₂ OH) in the TR-α of the rat and its corresponding residue is human TR asparagine 331 at -β with the side group -CH ₂ CONH ₂ . The branched chain in human TR-β has different hydrogen bond potentials that synthesize larger, charged, and differentiating compounds.

For example, in the complex of TR-α and triac, Ser 277 does not participate in ligand binding. Triac's equivalent affinity for alpha and beta isoforms lacking kinetics for Ser 277 (asn 331 in beta), and indirectly alpha / beta selectivity with amino acid substitution of Asn 331 for Ser 277 and Arg 228 Support the claim that it exists in the interaction.

In terms of ligand design, these differences mean that for β-selective ligands, some or all of the following differences should be used:

1. The presence of larger branched asparagine.

2. The ability of branched carbonyl groups to provide strong hydrogen bond donors.

3. The ability of branched amino acids to provide two hydrogen bond receptors.

4. Adjust the polarity to reorganize the water trapped in the T3 pocket.

In terms of pharmaceutical design, these differences mean that for the α-selective ligand, some or all of the following differences should be used:

1. The presence of smaller branched chains.

2. Ability of hydroxyl groups on the —CH ₂ OH side groups, branched carbonyl groups to provide weak hydrogen donors.

3. Adjust the polarity to reorganize the water trapped in the T3 pocket.

In both cases these differences can be used in several ways. For example, they can also be used using software built for the construction of new organic molecules such as Biosym-MSI.

Processing method

Compounds of formula (1) exhibit biological activities that can be useful for medical treatment and can be demonstrated in the following tests:

(i) Induction of mitochondrial α-glycerophosphate dehydrogenase (GPDH: EC 1.1,99.5). This assay is directed in some species, especially rats, to specifically induce mitochondrial α-glycerophosphate dehydrogenase by thyroid hormones and tyro mimetics in a closely related manner in reactive tissues such as liver, kidney and heart. Is particularly useful because it is [References: Westerfield, WW, Richert, DA and Ruegamer, W. R., Endocrinology, 1965, 77, 802. This assay allows direct determination of the thyroid hormone-like effect of a compound in rats, and in particular the direct measurement of the thyroid hormone-like effect in the heart;

(ii) an increase in basal metabolic rate as measured by an increase in total body oxygen consumption [Barker et al., Ann, N.Y. Acad. Sci., 86: 545-562 (1960);

(iii) stimulation of atrial rhythm rate isolated from animals previously administered with a tyro mimetic (Stephan et al., Biochem, Phannacol, (1992) 13: 1969-1974; Yokoyama et al., J. Med. Chem. 38: 695-707 (1995);

(iv) Changes in total plasma cholesterol levels determined using cholesterol oxidant kits (eg, Merck CHOD iodine colorimetric kits). Stephan et al. (1992);

(v) determination of LDL (low density lipoprotein) and HDL (high density lipoprotein) cholesterol in lipoprotein fractions separated by ultracentrifugation; And

(vi) Changes in total plasma triglyceride levels measured using enzyme color tests, eg, the Merck system GPO · PAP method.

The compound of formula (1) is

(a) at a dose that does not significantly modify cardiac GPDH levels, which increases the rate of metabolism in the test animal, increases hepatic GPDH levels,

(b) at doses that do not significantly modify cardiac GPDH levels, may be found to exhibit mimic activity with selective tyrosity by lowering plasma cholesterol and triglyceride levels and the ratio of LDL to HDL cholesterol.

Therefore, the compounds of formula (1) can be used for the treatment of diseases that can be alleviated by compounds that selectively mimic the effects of thyroid hormones in some tissues, while having little or no direct tyrosine effect in the heart. For example, compounds of formula (1) that increase hepatic GPDH and metabolic rate at doses that do not significantly modify cardiac GPDH levels are suggested for the treatment of obesity.

Agents of formula (1) that lower the ratio of total plasma cholesterol, LDL-cholesterol to HDL-cholesterol, and triglyceride levels at doses that do not significantly modify cardiac GPDH levels are intended to be used as general antihyperlipidemic agents (hyperlipoproteinases). That is, to treat patients with elevated levels of plasma lipids (cholesterol and triglycerides). In addition, in view of these effects on plasma cholesterol and triglycerides, they are also proposed for use as specific anti-cholesterol and anti-triglyceride inhibitors.

Patients with elevated plasma lipid levels are considered at risk of developing signs of coronary artery disease or other atherosclerosis as a result of their high plasma cholesterol and / or triglyceride concentrations. In addition, LDL-cholesterol versus HDL-cholesterol is believed to be a lipoprotein that induces atherosclerosis, and since HDL-cholesterol is believed to transport cholesterol from the vessel wall and interfere with the production of atherosclerosis plaques. Anti-hyperlipidemic agents that lower the ratio of sterols are incorporated herein by reference. Patents 4,826,876 and 5,466,861 are pointed out as anti atherosclerosis agents.

The present invention also provides a method for generating mimetic activity selective to some tissues other than the heart, wherein the method comprises the steps of a compound of formula (1) or a pharmaceutically acceptable salt thereof to the animal to produce said activity. Administering the required effective amount.

The present invention also relates to a method for lowering plasma lipid levels and a method for lowering the ratio of LDL-cholesterol to HDL-cholesterol levels by suitably administering a compound of the present invention or a pharmaceutically acceptable salt thereof.

In addition, the compounds of formula (1) may be presented for thyroid hormone replacement therapy in patients with compromised cardiac action.

Standard pharmaceutical compositions are generally administered in therapeutic uses of the compounds of the invention.

Therefore, in a further aspect the present invention provides a pharmaceutical composition comprising a compound of formula (1) or a pharmaceutically acceptable salt thereof and a pharmaceutically acceptable carrier. The composition comprises a compound suitable for oral, parenteral or rectal administration.

Pharmaceutical composition

The compound of formula (1) and its pharmaceutically acceptable salts, which are active when provided orally, can be formulated as liquids such as syrups, suspensions or emulsions, tablets, capsules and lozenges.

Liquid compositions are generally compounds or pharmaceuticals with suitable liquid carriers, for example non-aqueous solvents such as ethanol, glycerin, sorbitol, polyethylene glycol, suspensions, preservatives, surfactants, wetting agents, sweeteners or coloring agents in oil or water. It will consist of a suspension or solution of an acceptable salt. Alternatively, liquid formulations can be prepared from reconstitutable powders.

Powders containing the active compounds, suspensions, sucrose and sweeteners can be reconstituted with water to form suspensions and syrups can be prepared from powders containing the active ingredients, sucrose and sweeteners.

Compositions in tablet form may be prepared using suitable pharmaceutical carriers that are routinely used to prepare solid compositions. Examples of such carriers include magnesium stearate, starch, lactose, sucrose, microcrystalline cellulose and binders such as polyvinylpyrrolidone. Tablet forms may also be provided in a color film coating, or in a color included as part of a carrier. In addition, the active compounds may be formulated in controlled release dosage forms as tablets comprising a hydrophilic or hydrophobic matrix.

Compositions in capsule form can be prepared using routine encapsulation methods, for example by incorporation of the active compound and excipients into hard gelatin capsules. Alternatively, the active compound and high molecular weight polyethylene glycol of the semi-solid matrix can be prepared and filled into hard gelatin capsules, or a solution of the active compound in polyethylene glycol or a suspension in edible oil, for example liquid paraffin or fractionated coconut oil. It can be prepared and filled into soft gelatin capsules. The compound of formula (1) and its pharmaceutically acceptable salts, which are active when provided parenterally, can be formulated for intramuscular or intravenous administration.

Typical compositions for intramuscular administration will consist of suspensions or solutions of the active ingredient in oils such as arachis oil or sesame oil. Universal compositions for intravenous administration include, for example, active ingredients, dextrose, sodium chloride, auxiliary solvents such as polyethylene glycol, and any chelating agents such as ethylenediamine tetraacetic acid and antioxidants such as sodium It will consist of a sterile isotonic aqueous solution containing metabisulfite. Alternatively, the solution may be freeze dried and then reconstituted with a suitable solvent prior to administration.

Compounds of formula (I) and their pharmaceutically acceptable salts which are active at rectal administration can be formulated as suppositories. Universal suppository formulations will generally consist of the active ingredient in combination with binders and / or lubricants such as gelatin or cocoa butter or other low melting vegetable or synthetic waxes or fats.

The compound of formula (I) and its pharmaceutically acceptable salts which are active at topical administration may be formulated as a transdermal composition. Such compositions include, for example, baking, active compound reservoirs, control membranes, linear and contact adhesives.

The universal daily dosage of the compound of formula (1) will vary depending on the individual's requirements, the disease to be treated and the route of administration. Suitable dosages are generally from 0.001 to 10 mg / kg of recipient body weight per day.

Within this general dosage range, the dosage can be chosen such that the compound of formula (1) lowers plasma cholesterol levels and speeds up metabolism with little or no direct effect on the heart. Generally, non-exclusively, the dosage will range from 0.05 to 10 mg / kg.

In addition, within the usual dosage ranges, the dosage may be selected to lower plasma cholesterol levels, at which the compounds of formula (1) have no or little effect on the heart without increasing metabolic rate. Generally, non-exclusively, such dosages will be between 0.001 and 0.5 mg / kg.

It is to be understood that the two subranges mentioned above are not excluded from each other and that the specific activity, which consists of a particular dosage, varies depending on the nature of the compound of formula (1) used.

Preferably, the compound of formula (1) is present in unit dosage form, such as a tablet or capsule, so that the patient can be self-administered in a single dose. Generally, unit dosages contain 0.05 to 100 mg of compound of formula (1). Preferred unit dosages contain 005 to 10 mg of the compound of formula (1).

The active ingredient may be administered 1 to 6 times a day. As such, the daily dosage is generally between 0.05 and 600 mg per day. Preferably, the daily dose is in the range of 0.05-100 mg per day. Most preferably in the range of 0.05 to 5 mg per day.

Example 1 Synthesis of TR Ligands

Many TR ligands are known in the art, including T4 (thyroxine), T3, T2, and TS-9 [Ref. Jorgensen Hormones and Analogs, in 6 Hormonal Proteins and Peptides, Thyroid Hormones 107-204 (Choh) Hao Li ed., 1978) (incorporated herein by reference).

The method of synthesizing several TR ligands is described below.

Synthesis of TS1, TS2, TS3, TS4 and TS5

TS1, TS2, TS3, TS4 and TS5 and all of their analogs also include T3 (3,5,3 'triiode-L-tyronine), T4 (thyroxine) and 3,5-dioodothyronine) It can be prepared by simple acylation of the nitrogen atom of a tyronine analog. T1 was reacted with Ph ₂ CHCO ₂ NHS (N-hydroxy succinimide-2,2-diphenylacetate) and C ₁₆ H ₃₃ CO ₂ NHS to synthesize TS1 and TS2, respectively. TS3 was synthesized by reacting T3 with FMOC-Cl (fluorenylmethyloxycarbonylchloride). TS4 was synthesized by reacting T3 with tBOC ₂ O (tBOC anhydride or di-t-butyldicarbonate). TS5 was synthesized by reacting 3,5-dioodothyronine with tBOC ₂ O, which is different from TS1-4 because TS5 has -H instead of -I at the R ¹ ₃ position. General reaction design for TS1, TS2, TS3, TS4 and TS5 is shown in FIG. It should be noted that in the reaction design both TS5 and the precursor have hydrogen instead of iodine at the R ¹ ₃ position.

Synthesis of TS6 and TS7

TS6 was synthesized by reacting TS5 with paranitrophenyl isocyanate. TS7 was synthesized by reacting TS6 with TFA (trifluoroacetic acid), which cleaves the tBOC group. These reactions are simple organic synthesis reactions that can be performed by anyone skilled in the art. The composite design for TS6 and TS7 is shown in FIG. 12.

Synthesis of TS8

In the presence of triethylamine, and TBTU (hydroxybenztriazolouronium tetrafluoroborate) or HBTU (hydroxybenztriazolouronium hexafluorophosphate) and an amide producing condensation reagent, TS5 was converted to Ph ₂ CHNH ₂ (di Phenylmethylamine) to synthesize TS8. The synthetic design for TS8 is shown in FIG. 13.

Synthesis of 3,5-Diiodo-3'isopropyltyronine Derivatives

To design antagonists, it is important to have hydrophobic groups in the 3 'and 5' positions. Preferred hydrophobic groups in the 3 'position are methyl, benzyl, phenyl, iodine and heterocyclic structures. The synthesis of 3,5-diiodo-3'-isopropyl-5'-substituted tyronine is described below. The examples provided describe specific steps for synthesizing TS10 compounds, but such general reaction schemes are useful to those skilled in the art in any number of 3,5-diiodo-3'-isopropyl-5'-substituted tyronine derivatives. It can be used to synthesize and is characterized by having an extension in the 5 'position. Known organic synthesis techniques can be used to synthesize further compounds of this class.

The synthesis of TS10 is described below and shown in FIG. 14. The members used in the reaction design for TS10 representing the reaction product for each step are indicated in parentheses.

2-formyl-6-isopropylanisole (1) : 2-formyl-6-isopropyl ( prepared according to Casiraghi, et al. JCS Perkin I, 1862 (1980), incorporated herein by reference) Anisole (10.0 g, 61 mmol) was added dropwise to a suspension of sodium anhydride (3.7 g, 153 mmol) in 50 mL THF and 50 mL DMF in a round bottom flask. This addition causes an exothermic reaction and forms a gray solid. Then methyl iodine (26.0 g, 183 mmol) was added dropwise and the reaction mixture was stirred for 5 hours at room temperature. The reaction mixture was quenched with 20 mL of water, poured into 500 mL of water and extracted with ether (2x300 mL). The ether layers were combined, washed with water (5 × 1000 mL), dried over magnesium sulfate and concentrated in vacuo to yield 10.2 g (94%) of the title compound, having ¹ H NMR (CDCl ₃ ) properties as follows: d 10.30 (s, 1H), 7.63 (d, 1H, J = 3 Hz), 7.50 (d, 1H, J = 3 Hz), 7.13 (t, 1H, J = 3 Hz), 3.81 (s, 3H), 3.31 (heptet, 1H, J = 7.5 Hz), 1.19 (d, 6H, J = 7.5 Hz).

2- (2-hydroxynonyl) -6-isopropylanisole (not shown in the design):

Octyl magnesium chloride (8.4 mL, 16.9 mmol, 2.0 M) was added dropwise to a solution of Compound (1) (1.5 g, 8.4 mmol) in 10 mL THF at -78 ° C. The reaction mixture was stirred for 2 hours while warming to room temperature. The reaction mixture was diluted with 50 mL ether and poured into 50 mL water. The ether layer was washed with brine (1x50 mL), dried over sodium sulfate and concentrated in vacuo. Flash chromatography (silica gel, 10% ether / hexanes → 15% ether / hexanes) afforded 734 mg (30%) of the title compound, having ¹ H NMR (CDCl ₃ ) properties as follows: d 7.33-7.10 (m, 3H), 5.00 (br, s, 1H), 3.81 (s, 3H), 3.33 (heptet, 1H, J = 7 Hz), 1.90-1.19 (m, 14H), 0.86 (t, 3H, J = 6.5 Hz); HRMS (EI), found: 292.2404; Theorem: 292.2402.

2-nonyl-6-isopropylanisole (2) : Compound (2) (663 mg, 2.3 mmol) was dissolved in a solution of 5 mL ethanol and 5 mL acetic acid, and a carbonaceous palladium catalyst was added as a spatula tip. The reaction mixture was then charged with hydrogen gas (using a simple balloon and needle) and the mixture was stirred overnight at room temperature. The next day, the reaction mixture was poured into ether (100 mL) and the ether layer was extracted with saturated sodium bicarbonate (3 × 100 mL). The ether layer was dried over sodium sulfate and concentrated in vacuo to give 581 mg (91%) of compound (2), with ¹ H NMR (CDCl ₃ ) properties as follows: d 7.14-7.00 (m, 3H), 3.75 ( s, 3H), 3.36 (heptet, 1H, J = 6.8 Hz), 2.63 (t, 2H, J = 7.5 Hz), 1.68-1.15 (m, 14H), 0.86 (t, 3H, J = 5.5 Hz); HRMS (EI), found: 276.2459; Theory: 276.2453.

Tyronine addition product (4) : fuming nitric acid (0.071 mL) was added to 0.184 mL acetic anhydride cooled to -5 ° C. Iodine (66 mg) was added to the mixture, followed by trifluoroacetic acid (0.124 mL). The solution was allowed to warm to room temperature with stirring for 1 hour, at which point all iodine dissolved. The reaction mixture was then concentrated in vacuo to give an oily semisolid. The residue was dissolved in 0.7 mL acetic anhydride and cooled to -20 ° C. A drop of anisole (2) (581 mg, 2.1 mmol) in 1.2 mL acetic anhydride and 0.58 mL TFA was added dropwise. The reaction mixture was stirred for 1 hour at 20 ° C., then warmed to room temperature with stirring overnight. The reaction mixture was partitioned between water and methylene chloride. The methylene chloride layer was dried over sodium sulfate and concentrated in vacuo to yield the iodonium salt (3) as an oil. This material was introduced directly into the coupling reaction without purification or characterization.

(Cited by reference herein) [N. Lewis and P. Wallbank, Synthesis 1103 (1987)] N-trifluoroacetyl-3,5-diiodotyrosine methyl ester (552 mg, 1.0 mmol), and all crude iodine from above The nium salt (3) was dissolved in 5 mL anhydrous methanol. Diazabicyclo [5.4.0] undecane (DBU) (183 mg, 1.2 mmol) and bronze copper were added with a spatula tip and the resulting mixture was stirred at rt overnight. The next day, the reaction mixture was filtered and the filtrate was concentrated in vacuo. The crude residue was purified by flash chromatography (silica gel, 10% ethyl acetate / hexanes) to yield 30 mg (4%) of protected tyronine addition product (4).

Deprotected tyronine (TS10) : Protected tyronine (4) (30 mg, 0.04 mmol) was dissolved in a mixture of 2.25 mL acetic acid and 2.25 mL 49% hydrobromic acid. The reaction mixture was heated to reflux for 5 hours. The reaction mixture was cooled to room temperature and the solvent was removed in vacuo. The oily residue was triturated to a gray solid by the addition of water. This solid was filtered off, washed with water and dried over P ₂ O _{5 in} vacuo to afford 24 mg (81%) of the desired compound TS10, having ¹ H NMR (CDCl ₃ ) properties as follows: d 7.57 (1 s) , 1H), 6.86 (s, 1H), 6.45 (s, 1H), 6.34 (s, 1H), 4.81 (m, 1H), 3.86 (s, 3H), 3.71 (s, 3H), 3.33-3.05 ( m, 3H), 2.58-2.47 (m, 2H), 1.62-0.76 (m, 23H); MS (LSIMS): M ⁺ = 817.0.

As mentioned above, one of ordinary skill in the art can alter the reaction design to synthesize a class of compounds characterized by 3,5-diiodo-3 'isopropyltyronine derivatives, wherein (1) 3 'isopropyl groups can be replaced with hydrophobic groups including methyl, benzyl, phenyl, iodine and heterocyclic structures, and (2) alkyl groups, two-dimensional aryls, heterocyclic groups, or polar groups and / or charged groups A wide range of chemical structures, including groups, can be incorporated at the 5 'position.

In this reaction design, aldehyde (1) is a versatile synthetic intermediate that provides for the attachment of various chemical moieties at the 5 'position of the final tyronine derivative. In addition, various chemical reactions may be used to attach chemical moieties. These reactions are well known in the art and include organometallic addition of aldehydes (including Grignard reagents, organolithiums, etc.), reductive amination reactions of primary or secondary amines with aldehydes, phosphorus ones. There is a Wittig olefination reaction with lead or stabilized phosphonate anions. Another possible reaction is the reduction of aldehyde to benzyl alcohol, which allows etherification in the 5 'position. As noted above, these methods allow a wide range of chemical structures to be incorporated at the 5 ′ position of the final tyronine derivative, including alkyl groups, two-dimensional aryls, heterocyclic groups or polar groups and / or charged groups.

Synthesis of 3,5-dibromo-4- (3 ', 5'-diisopropyl-4'-hydroxyphenoxy) benzoic acid (Compound 11).

(a) A mixture of 2,6-diisopropyl phenol (20 g, 0.11 mol), potassium carbonate (62 g, 0.45 mol), acetone (160 mL) and methyl iodide (28 mL, 0.45 mol) was refluxed for 3 days. The reaction mixture was filtered through celite, evaporated, dissolved in ether, washed twice with 1M sodium hydroxide, dried over magnesium sulfate and concentrated to give 15.1 g of 2,6-diisopropyl anisole as light yellow oil ( 0.08 mol, 70%).

(b) Fuming nitric acid (12.4 mL, 265 mmol) was added dropwise to 31.4 mL of acetic anhydride cooled in a dry ice / carbon tetrachloride bath. A small amount of iodine (11.3 g, 44.4 mmol) was added, followed by dropwise addition of trifluoroacetic acid (20.5 mL, 266 mmol). The reaction mixture was stirred at room temperature until iodine dissolved. Nitrogen oxides were removed by flushing nitrogen into the vessel. The reaction mixture was concentrated and the residue was dissolved in 126 mL of acetic anhydride and cooled in a dry ice / carbon tetrachloride bath. To the stirred solution was added 2,6-diisopropylanisole (51 g, 266 mmol) in 150 mL acetic anhydride and 22.6 mL trifluoroacetic acid, one by one. The reaction mixture was left overnight at room temperature and then concentrated. The residue was dissolved in 150 mL of methanol and treated with 150 mL of 10% aqueous sodium sulfite solution and 1 L of 2M sodium borotetrafluoride solution. After the aggregates precipitated, petroleum ether was added and the supernatant was decanted. The precipitate was triturated with petroleum ether, filtered, washed with petroleum ether and dried at room temperature in vacuo. This gave 34 g (57 mmol, 65%) of bis (3,5-diisopropyl-4-methoxyphenyl) iodium tetrafluoroborate as a white solid.

(c) Thionyl chloride (3 mL) was added dropwise to a stirred solution of 3,5-dibromo-4-hydroxybenzoic acid (12 g, 40.5 mmol) in 250 mL methanol. The reaction mixture was refluxed for 5 days and water was added to filter out the precipitated product. The residue was dissolved in ether ether acetate. From the aqueous phase, methanol was removed by concentration. The aqueous phase was then saturated with sodium chloride and extracted with ethyl acetate. The combined organic phases were dried over magnesium sulphate, filtered and concentrated. This gave 12.5 g (40.5 mmol, 100%) of 3,5-dibromo-4-hydroxymethyl benzoate as a white crystalline solid.

(d) The products obtained in steps (b) and (c) were reacted with each other according to the following protocol. Dichloromethane in bis (3,5-diisopropyl-4-methoxyphenyl) iodium tetrafluorobenzoate (2.86 g, 4.8 mmol) and bronze copper (0.42 g, 6.4 mmol) in 7 mL dichloromethane at 0 ° C. 3,5-dibromo-4-hydroxymethylbenzoate (1.0 g, 3.2 mmol) and triethylamine (0.36 g, 3.5 mmol) in 5 mL were added dropwise. The reaction mixture was stirred for 8 days in the absence of light and then filtered through celite. The filtrate was concentrated and the residue was purified by column chromatography (silica gel, 97: 3 petroleum ether / ethyl acetate) to give 3,5-dibromo-4- (3 ', 5'-diisopropyl- as a solid. 0.62 g (1.2 mmol, 39%) of 4'-methoxyphenoxy) methyl benzoate was obtained.

(e) The product (0.2 g, 0.4 mmol) from step (d) was dissolved in 2 mL of dichloromethane, left under nitrogen and cooled to -40 ° C. 1 M BBr ₃ (1.2 mL, 1.2 mmol) was added dropwise to the stirred solution. The reaction mixture was allowed to reach room temperature and left overnight. Cool to 0 ° C. and hydrolyze with water. Dichloromethane was removed by concentration and the aqueous phase was extracted with ethyl acetate. The organic phase was washed with 1M hydrochloric acid and brine. Then dried over magnesium sulfate, filtered and concentrated. Chromatography of the residue (silica, 96: 3.6: 0.4 dichloromethane / methanol / acetic acid) gave 3,5-dibromo-4- (3 ', 5'-diisopropyl-4'- as a white solid. 93 mg (0.2 mmol, 51%) of hydroxyphenoxy) benzoic acid were obtained. ¹ H NMR (CDCl ₃ ) δ1.23 (d, 12H, methyl), 3.11 (m, 2H, CH), 6.50 (s, 2H, 2.6-H), 8.33 (s, 2H, 2 ', 6'- H).

Table 1 and Figure 15 illustrate the structure of several TR ligands.

Example 2 Receptor Binding Assay of TR Ligands

To test the ability of synthesized TR ligands to bind thyroid receptors (TR), see JD Apriletti, JB Baxter, and TN Lavin, J. Biol. Binding affinity of TR ligands to TR was assayed using TR's prepared from 125 ₁ T ₃ and rat liver nuclei as described in Chem., 263: 9409-9417 (1988). The apparent Kd was calculated using the method described by Aprilletti (1995) and Aprilleti (1988). The apparent Kd is listed in Table 2. 1 nM [ ¹²⁵ I] T ₃ , and 50 μg / mL core histone in Buffer E (400 mM KCl, 200 mM potassium phosphate, pH 8.0, 0.5 mM EDTA, 1 mM MgCl ₂ , 10% glycerol, 1 mM DDT) in a volume of 0.21 mL. The apparent Kd (App.Kd) was measured in the presence of the sample to be assayed. After overnight incubation at 4 ° C., 0.2 mL of the incubation mixture was equilibrated with Buffer E in a Quick-Sep Sephadex G-25 column (2.7 × 0.9 cm, 1.7 mL bed volume). )). Excluded peaks of protein bound [ ¹²⁵ I] T ₃ were eluted with 1 mL of Buffer E and collected in vitro and counted. Specific T ₃ binding was calculated by subtracting nonspecific binding from total binding.

Example 3 Increased Nucleoprotein Co-Activation by TR Ligands

To test the ability of TR ligands to activate the binding of TR to the nuclear activating protein RIP-140 (a nuclear protein capable of binding to nuclear receptors such as the estrogen receptor), see herein after the TR ligand has ligand bound to TR. Cited in [V. Cavailles, et al., EMBO J., 14 (15): 3741-3751 (1995), incubated with RIP-140. In this assay, the 35 _S -RIP-140 protein binds to ligand bound TR but not to ligand unbound TR. Many TR 35 _S ligands activated RIP-140 binding as described in Table 2.

Example 4 TR Ligand Binding and TR Activation in Cultured Cells

To test the TR activation of transcription in the cellular environment, TR ligands were assayed for their ability to activate reporter genes, ie, chloramphenicol transferases ("CAT"), with viablely linked TR DNA binding sequences. GC or L937 cells (commercially available from ATCC) can be used, respectively. In this assay, the TR ligand crosses the cell membrane, binds to TR and activates TR, which in turn activates gene transcription of CAT by binding the TR DNA binding region upstream of the CAT gene. CAT gene activation was assayed at various concentrations as described herein and in literature to determine the effective concentration (EC ₅₀ ) for half maximum gene activation. The results of the CAT gene activation experiments are listed in Table 2.

CAT Gene Activation Assay

The thyroid hormones (3, 3) in rat pituitary cell lines (ie GC cells) containing endogenous thyroid hormone receptors (TRs), or U937 cells containing exogenous TRs expressed as known in the art. Functional responses to 5,3'-triiodo-L-tyrosine, T ₃ ) and TR ligands were assayed. GC cells were cultured in 10 cm dishes in RPMI 1640 with 10% neonatal bovine serum, 2 mM glutamine, 50 units / mL penicillin and 50 μg / mL streptomycin. For transfection, cells are treated with trypsin, resuspended in buffer (PBS, 0.1% glucose), mixed with TREtkCAT plasmid (10 mg) or phage in 0.5 mL buffer (15 ± 5 million cells), and 0.33 Electroporation was performed using a Bio-Rad gene pulser at kv and 960 mF. The TREtkCAT plasmid was derived from the T ₃ response element (AGGTCAcaggAGGTCA) cloned at the Hind III site of the pUC19 polylinker upstream of the minimal (-32 / + 45) thymidine kinase promoter bound to the CAT (tkCAT) coding sequence. It contains two copies. After performing electroporation, cells were pooled in growth medium (PRMI with 10% charcoal treated, hormone depleted neonatal serum), plated in 6 well dishes and treated with ethanol or hormone. Leitman, RCJ Ribeiro, ER Mackow, JDBaxter, BL West, J. Biol, Chem. 266, 9343 (1991), CAT activity was measured after 24 hours, which is incorporated herein by reference.

Effect of TS-10 on transcriptional regulation of DR4-ALP reporter gene in the presence or absence of T3.

Characteristics of TRAF Cells: TRAFa1 is a CHO K1 cell stably transformed with an expression vector encoding human thyroid hormone receptor α1 and DR4-ALP reporter vectors; TRAFb1 is a CHO K1 cell stably transformed with an expression vector encoding human thyroid hormone receptor β1 and DR4-ALP reporter vectors.

Interpretation of the effect of compound TS-10 on the regulation of transcription of the DR4-ALP reporter gene in the presence or absence of T3.

TRAFa1 reporter cells: TS-10 alone (open source) partially activates the expression of the ALP reporter protein, which accounts for about 27% of the maximum effect by the natural thyroid hormone T3. If T3 (filled circle) is present, TS-10 has a weak antagonistic effect. EC50 concentrations for the antagonistic effects of TS-10 and EC50 concentrations for the T3 antagonistic effects are shown in FIG. 18, respectively.

In FIG. 18, open circles and circles filled with dotted lines show the dose dependent effect of TS-10 / T3 on toxic markers (MTS / PMS), and the reduction of tetrazolium salt in mitochondria is displayed on the right Y-axis as optical density. It is. At the highest concentration of TS-10 (32 mM) in both the absence and presence of T3, as can be seen under the microscope, there is no obvious toxic effect of TS-10 with respect to MTS-PMS markers, but the cell morphology There is a definite effect as to (not shown in the figure).

TRAFb1 Reporter Cells : TS-10 alone (open circle) partially expresses ALP reporter protein up to about 35% of maximum effect by T3. EC50 concentrations for the antagonistic effects of TS-10 are shown in FIG. 19. In the presence of T3 (circled circles), TS-10 shows a rather slight increase in the T3 effect on the expression of the ALP reporter protein. At the highest concentrations used (32 mM), the T3 inhibitory effect of TS-10 is toxic rather than T3 antagonist.

In FIG. 19 , open circles and circles filled with dotted lines show a dose dependent effect of TS-10 / T3 on toxic markers (MTS / PMS), a decrease in tetrazolium salts in the mitochondria, on the right Y axis as optical density. Is displayed. At the highest concentration of TS-10 (32 mM) in both the absence and presence of T3, as can be seen under the microscope, there is no obvious toxic effect of TS-10 with respect to MTS-PMS markers, but the cell morphology There is a definite effect as to (not shown in the figure).

HepG2 (HAF18) Reporter Cells : TS-10 (open circle) alone activates ALP reporter protein to a level slightly higher than about 50% of maximal effect by T3. EC50 concentration for the antagonistic effect of TS-10 is shown in FIG. 20. In the presence of T3 (circled circles), TS-10 does not exhibit any effect, ie, any T3 antagonism or augmentation / increase effect on T3. Open circles and circles filled with dotted lines indicate the dose dependent effect of TS-10 / T3 on toxic markers (MTS / PMS), a decrease in tetrazolium in the mitochondria, and is displayed on the right Y axis as optical density. At any concentration of TS-10 / T3 used, no obvious toxic effects are seen with respect to MTS-PMS markers, as can be observed using a microscope.

Example 5-T ₃ And bound to TR LBD by Triac [ ¹²⁵ I] T ₃ Comparison of human TR-α and human TR-β competition for

The drug, ie triac, is a thyroid hormone antagonist. Triacs are 3,5,3'-triiodotroacetic acid and are described in Jorgensen, Thyroid Hormones and analogs in 6 Hormonal Protein and Peptides, Thyroid Hormones at 150-151 (1978). Another compound that may be used instead of triac is 3,5-diiodo-3'-isopropyltyroacetic acid. Competition assays were performed to compare the replacement of [ ¹²⁵ I] T ₃ from binding to human TR-α LBD or human TR-β LBD by unlabeled T ₃ or triac. The results of this assay are shown in FIG. 16.

Prepare standard binding reactants containing 1 nM [ ¹²⁵ I] T ₃ , 30 fmol of human TR-α (empty symbol) or β (solid symbol), and competing unlabeled T ₃ (circle) or triac at varying concentrations It was. This assay was performed twice.

Coupled to TR [ ¹²⁵ I] T ₃ Standard black

Human TR-α (left panel) or human TR-β (right panel) for T ₃ binding was assayed with increasing concentration of [ ¹²⁵ I] T ₃ . The equilibrium dissociation constant (20 pM for α and 67 pM for β) was calculated using a linear regression test and is shown in FIG. 17.

3,5-Dibromo-4- (3 ', 5'-diisopropyl-4'-hydroxyphenoxy) benzoic acid is a TR-α selective synthetic ligand.

3,5-dibromo-4- (3 ', 5'-diisopropyl-4'-hydroxyphenoxy) benzoic acid (compound that binds to two different isoforms of TR, TR-α and TR-β 11) (having the structure shown above) was assayed. Compound (11) showed an IC50 of 1.6 μM to bind TR-α and an IC50 of 0.91 μM to bind TR-β. Assays for determining selective binding to TR-α or TR-β LBD may include reporter assays as described herein. [References: Hollenberg, et al., J. Biol. Chem., 270 (24) 14274-14280 (1995).

Example 6 Preparation and Purification of TR-α LBD

The rat TR-α LBD, residues Met122-Val410 were purified from E. coli (“LBD-122 / 410”). Expression vectors encoding rat TR-α LBDs were freshly transfected with E. coli strain BL21 (DE3) and incubated at 22 ° C. in a 50 liter (1) fermenter using 2 × LB medium. At A ₆₀₀ of 2.5-3, IPTG was added to 0.5 mM and incubation continued for 3 hours before harvesting. Bacteria pellets were quickly frozen in liquid nitrogen and stored at -70 ° C until processing. Extraction and purification steps were performed at 4 ° C. Bacteria were thawed in extraction buffer (20 mM Hepes, pH 8, 1 mM EDTA, 0.1% MTG, 0.1 mM PMSF, and 10% glycerol) at 10 mL buffer / g bacteria ratio. Bacteria were lysed by incubating with 0.2 mg / mL lysozyme for 15 minutes, sonicated to maximum power, and at the same time Brinkmann homogenizer (generator PTA 35/2 until the solution lost its viscosity). Homogenized with model PT10 / 35). After centrifugation at 10,000 g for 10 minutes, the supernatant was adjusted to 0.4 MKCl and treated with 0.6% PEI to precipitate fragmented DNA and centrifuged at 10,000 g for 10 minutes. The TR-α LBD of rats in supernatant was then precipitated with 50% ammonium sulfate and centrifuged at 10,000 g for 10 minutes. The precipitate was resuspended in Buffer B (20 mM Hepes, pH 8.0, 1 mM EDTA, 1 mM DTT, 0.1 mM PMSF, 0.01% Lubrol and 10% Glycerol) to give a final conductivity of 9 mS / cm (about 0.7 M ammonium sulfate). ) And centrifuged at 100,000 g for 1 hour. The supernatant was frozen in liquid nitrogen and stored at -70 ° C.

The crude extract was thawed, bound to labeled amounts of [ ¹²⁵ I] T _3, and loaded directly onto a phenyl-Toyopearl hydrophobic interaction column (2.6 × 18 cm, 95 mL bed volume) at a rate of 1.5 mL / min. I was. The column was eluted with a salt free 20% glycerol from glycerol free 0.7 ammonium sulfate in Buffer C (20 mM Hepes, pH 8.0, 0.5 mM EDTA, 1 mM DDT, 0.2 mM PMSF) in a 2 hour gradient. Detection of TR-α LBD in rats pre-bound to label [ ¹²⁵ I] T ₃ (less than 0.005% of total rat TR-α LBD) using a flow-through gamma emission detector On the other hand TR-α LBD of non-ligand bound rats was assayed using a postcolumn [ ¹²⁵ I] T ₃ binding assay (as described herein).

TR-α LBD peak fraction of phenyl-Toyopearl ligand bound rats was pooled and diluted with Buffer B to have a conductivity of 0.5 mS / cm (corresponding to about 20 mM ammonium sulfate) and a TSK-DEAE anion exchange column (2x15 cm) , 47 ml bed volume), at a rate of 4 mL / min and eluted with a 60 minute gradient with 50 to 200 mM NaCl in Buffer B.

TR-α LBD peak fraction of ligand unbound rats from TSK-DEAE was pooled, diluted 2-fold with Buffer B, and 0.75 mL / min on TSK-heparin HPLC column (0.8x7.5 cm, 3 ml bed volume). And was eluted with a gradient of 50-400 mM NaCl in Buffer B.

The pool of TR-α LBD peak fraction of ligand unbound rats from TSK-heparin column was adjusted with 0.7 M ammonium sulfate and 0.75 mL / min on a TSK-phenyl HPLC column (0.8 × 7.5 cm, 3 ml bed volume). It was loaded at speed and eluted for 60 minutes with a gradient of glycerol-free 0.7 M ammonium sulfate to salt free 20% glycerol in Buffer C. Fractions containing TR-α LBD of ligand unbound rats were pooled and incubated for at least 5 times the hormone for 1 hour, the salt concentration was adjusted to 0.7 M ammonium sulfate, the sample was reloaded and Chromatography was performed on the same column.

Example 7 Crystallization of Ligand-Bound TR-α LBDs

The material was split into batches from a single LBD-122 / 410 formulation and allowed to quantitatively bind with one of the following ligands: Dimit, T ₃ , or Triac IpBr ₂ for the final purification step. (3,5-dibromo-3 'isopropyltyronine).

To maintain complete saturation of the TR-α LBD of the rat with ligand and to prepare a complex for crystallization, the TR-α LBD of the ligand bound rat was concentrated, 10 mM Hepes (pH 7.0), 3.0 mM DDT, and Amicon Centricon-10 microconcentrator using 1.0 nM to 10 nM ligands (McGrath et al, Biotechniques, 7: 246-247 (1989)), which is incorporated herein by reference. Desalination).

17 (with 0.5 mL reservoir using 1 μl protein solution, 1 μl precipitant solution and coverslips treated with silane) as in McPherson, Preparation and Analysis of Protein Crystals (1982), incorporated herein by reference. Coefficient Crystallization Screening Study for TR-α LBD of Rats Bound to Selected Ligands Using Hanging Drop Vapor Diffusion at ° C [Jancarik & Kim, J. Appl. Crystallogr . 24: 409-411 (1991), which is incorporated herein by reference. The rats' TR-α LBDs were stored at −80 ° C. because they were not stable at 4 ° C., where they maintained their total affinity for the hormone and its crystallinity for 2-3 months. McGrath, ME et al., J. Mol Biol . 237: 236-239 (1994). After crystals were obtained under condition (21) of the screening study (see Jancarik & Kim 1991), the conditions were optimized. The wedge-shaped crystals were 15% 2-methyl-2,4-pentanediol (MPD), 0.2M ammonium acetate and 0.1M sodium cacodylate (pH 6.7), 3 mM DDT for 3 days, 2 μl protein solution, 1 μl precipitant With a 0.6 mL reservoir using coverslips treated with solution and silane, and 8.7 mg / mL (dimit), 5.5 mg / mL (IpBr ₂ ), 5 mg / mL (triac), or 2.3 mg / mL ( T ₃ ) was obtained reproducibly by quenching vapor diffusion at 22 ° C. Under these conditions, diffraction characteristic determination (dimension 0.5 × 0.2 × 0.0075 mm ³ ) can be grown at ambient temperature (22 ° C.). The best crystals have a limited dimension of about 100 μm and are obtained at a protein concentration of 2.3 to 8.7 mg / mL in the presence of 3 mM DTT. The crystal is monoclinic space group C2 with one monomer of asymmetric units.

Example 8-T ₃ Or crystallization of human TR-β LBD complexed with triac

TR-β LBD of human complexed with T ₃ and TR-β LBD of human complexed with triac were purified according to the same method as described above for TR-α LBD of rats with the following changes.

According to the amino acid numbering scheme for the various nuclear receptor LBDs shown in FIG . 3 , expression of human TR-β LBD is characterized in that the human TR-β LBD residue extends from the amino acid at position 716 through the amino acid at position 1022. Different from TR-α LBD in rats. 3 shows a numbering scheme that can be applied to all listed nuclear receptors and any additional cognate nuclear receptors. In FIG. 3, the vertical lines at positions 725 and 1025 describe the preferred minimum amino acid sequence needed to obtain proper binding of the ligand. The amino acid sequence at positions 716 to 1022 according to the numbering design of FIG. 3 corresponds to amino acid positions 202 to 461 according to the conventional numbering of the amino acid sequences of TR-β in publicly available. In addition, human TR-β LBD was expressed with a histidine tag as described in Crown et al., Methods in Molecular Biology 31: 371-387 (1994), incorporated herein by reference.

Purification of human TR-β LBD was performed in the same manner as described above for rat TR-α LBD with the following exceptions. First, prior to the purification step using a hydrophobic interaction column, human TR-β LBD expressed using a nickel NTA column (commercially available from Qiagen, Catsworth, CA) according to the manufacturer's instructions Was purified and eluted with 200 mM imidazole. The second difference is that in the purification of human TR-β LBD, the purification step using a heparin column was omitted.

The crystallization of TR-β LBD in humans bound to T ₃ or triac is as follows. At ambient temperature (22 ° C.), using a coefficient crystallization screening study by Jancarik & Kim (1991) and using the product commercially available from Hampton Research, Riverside, Crystals were obtained under condition 7 of the counting screen. Optimum conditions were as follows: 1 mM of 3 mM DTT, 3 μM from a suspension containing 1.0-1.2 M sodium acetate (pH not adjusted) and 0.1 M sodium cacodylate (pH 7.4) at a protein concentration of 7-10 mg / mL. Hexagonal double pyramid crystals were grown at 4 ° C. for 2-3 days with a 0.6 mL reservoir using protein solution, 1 μl precipitant solution or 2 μl protein solution, 1 μl precipitant solution and coverslip treated with silane. The best crystals have a limited dimension of 200 μm.

The determining system for TR-β LBD in humans bound to T ₃ or triac is a trigonal system with spatial group p3 ₁ 21. The dimensions of the unit cell are cell length a = cell length b = 68.448 mm 3 and cell length c = 130.559 mm 3. The angles are α = 90 °, β = 90 ° and γ = 120 °, respectively.

Example 9-Determination of Ligand Decombination TR-α LBD Crystal Structure

Three copper crystals on the Mar area detector at the Stanford synchrotron radiation laboratory beamline 7-1 (λ = 1.08 Hz) using 1.2 "vibrations (TR-α LBD, T3 and Data from each of IpBr ₂ ) was measured.

Data from the T ₃ copper crystals were measured with the b ^* axis nearly parallel to the axis. Crystals were flash frozen at −178 ° C. in a nitrogen gas stream with MPD mother liquor, which acts as a low temperature solvent. REFIX was used to determine the orientation matrix for each crystal [Kabsch, W., J. Appl. Crystallogr. 26: 795-800 (1993), incorporated herein by reference. Reflected light was collected with DENZO (commercially available from Molecular Structure Corp., Woodland, Texas), (Otwinowski, Z, Proceedings of the CCP4 Study Weekend; Reflectance was assessed by SCALEPACK as described in "Data Collection and Processing." 56-62 (SERC Darebury Laboratory, Warrington, UK 1993).

For the T ₃ data set, the Bijvoet pairs are stored separately and MADSYS (Dw. Hendrickson (Colombia University) and W. Weis (Standford University)) Evaluation was locally using.

Copper crystals prepared from three isomeric ligands are isoforms. MIR analyzes were performed using the program from the CCP4 suit [Ref. Collaborative Computational Project, NR Acta Crystallogr. D50: 760-763 (1994)], incorporated herein by reference. Dimit copper crystals were taken as the parent, and the difference Patterson was calculated for both T ₃ and IpBr ₃ . The positions of the three iodine atoms in the T ₃ difference Patterson were clearly determined from the Harker section of the density map as peak 11σ on the background. The positions for the two bromine atoms in the IpBr ₂ copper crystals are independently located as peak 8σ above the noise level. The phase for LBD-122 / 410 from solution to IpBr _{2 difference} Patterson was calculated and used to identify the location of the only third iodine of the T ₃ copper crystal. Halogen sites were purified with MLPHARE, including unusual contributions from iodine atoms (Otwinowski, Z. Proceedings of the CCPR study Weekend 80-86 (SERC Darebury Laboratory, Warrington, UK 1991)). The MIRAS phase was improved through solvent flattening / bar graph matching using DM (Cowtan, K., Joint CCP4 and ESF-EACBM Newsletter on Protein Crystallography 31: 34-38 (1994)) ( Which is incorporated herein by reference).

The LBD-122 / 410 model with bound dimit was built using Program O from a solvent-plated MIRAS 2.5 kV electron density map [Reference: Jones et al., Acta Crystallogr. A 47: 110-119 (1991), which is incorporated herein by reference. The initial model without ligand (Rcryst = 40.1%) was purified using XPLOR using least squares protocol. Dimit ligands were made with an indeterminate F _O -F _C difference density for the following number of times. Subsequent purifications used both least squared and simulated annealing protocols with XPLOR (Brunger et al., Science 235: 458-460 (1987), which is incorporated herein by reference). Individual atom B-factors were isotropically purified. As defined in PROCHECK, all residues are described in Lasowski et al., J. Appl. Crystallogr. 26: 283-291 (1993), within the permitted backbone torsion region. The current model is missing 34 residues (Met ₁₂₂ -Gln ₁₅₆ ) at the N-terminus and 5 residues (Glu406-Val410) at the C-terminus.

In addition, the following residues are not modeled above Cβ because of poor density: 184, 186, 190, 198, 206, 209, 240, 301, 330, 337, 340, 343, 359 and 395. Average for protein atoms B value is 34.5 ³² . The final model was LBD-122 / 410, residues Arg _157- Ser ₁₈₃ , Trp ₁₈₅ -Gly ₁₉₇ , Ser ₁₉₉ -Asp ₂₀₆ and Asp ₂₀₈ -Phe ₄₀₅ ; Three cacodylate changed cysteines: Cys ₃₃₄ , Cys ₃₈₀ and Cys ₃₉₂ ; And 73 solvent molecules (2003 atoms) modeled as water.

* R _sym = 100xΣ _hkl Σ _i | I _i -I | / Σ _hkl Σ _i I _i

† R _der = 100xΣ _hkl | F _PH -F _H | / Σ _hkl | F _P |

The occupancy for the two bromine sites is fixed at 35 electrons. The occupancy of the iodine site is proportional to this value.

Phasing power = <FH> / <ε>, where <FH> is the average calculated heavy atom structural factor magnitude, and <ε> is the average estimated deficit of the closure.

R _cullis = <ε> / <iso>, where <ε> is the average estimated deficit of the closure, and <iso> is the difference of the isoform.

¶ R _cryst = 100 _× Σ _hkl | F _O −F _C | / Σ _hkl | F _O |, where F _O and F _C are observed (for data F / σ> 2) and are the calculated structural factor magnitudes. R _free was calculated using 3% of the data, randomly selected and omitted by purification.

§ Correlation coefficient = Σ _hkl (| F _O |-| F _O |) x (| F _C |-| F _C |) / Σ _hkl (| F _O |-| F _O |) ² x Σ _nkl (| F _C |-| F _C |) ²

Example 10 Phase Formation of Triac and rTRα LBD Complexes

Due to the possible non-iso-form of the tree evil rTRα LBD complex, AMORE [Note from the starting model consisting of rTRα LBD complex with T ₃ Document: Navaza, J., Acta Crystallographica Section A-Fundamentals of Crystallography 50: 157-63 (1994 ) Was used to determine the molecular replacement solution, but in complex with the ligand, all water molecules, and the following residues were omitted: Asn179, Arg228, Arg262, Arg266 and Ser 277. Strong peaks in both rotational and translational motion studies Was obtained and no significant (> 0.5 times normal peak) false solution was observed (see Table 3). Strong positive density percentages in both anomalous and conventional differential Fourier maps convince the solution. After 15 cycles of the maximum possible purification, REFMAC [Murshudov, et al. The map was calculated using the Sigma-A weighting coefficient output by "Application of Maximum Likelihood Refinement," in the Refinement of Protein Structures, Proceedings of Daresbury Study Weekend (1996). Triacs, omitted residues, and water molecules 503, 504, and 534 (according to the numbering agreement for T3 and TR complexes) were made with the difference density generated using O, Jones et al .; The morphology of these residues was further confirmed in the simulated annealing omission map (Brunger et al.). The complete model was then refined using the least squares of positions, annealing of the simulations, and grouped B factor purification in the XPLOR, resulting in Rcryst of 23.6% and Rfree of 24.1%.

Example 11 Relationship of Structure and QSAR to Troid Hormone Receptors

The conclusion of the thyroid hormone receptor quantitative structure-activity relationship can be summarized as follows:

1) R ₄ '-hydroxyl groups act as hydrogen bond donors;

2) amino propionic acid electrostatically interacts with positively charged residues from the receptor via carboxylate anions;

3) the affinity of the R ₃ / R ₅ substituent is I>Br> Me >>H;

4) The preference of R ₃ '-substituent is Ipr>I>Br> Me >> H.

Antagonist 3,5,3'-triiodotyronine (T ₃ ), 3,5-dibromo-3'-isopropyltyronine (IpBr ₂ ), 3,5-dimethyl-3'-isopropyltyronine (Dimit), and the structure of the thyroid hormone receptor ligand binding domain complexed with 3,5,3'-triiodoacetic acid (triac) as provided herein

1) identification of receptor determinants of binding at the level of hydrogen bonding;

2) anticipation of the classical thyroid hormone receptor QSAR and the combination of these determinants; And

3) It enables the prediction that the determinants of binding are tight and flexible for both ligand and receptor.

The classification for antagonists with the form (R ₁ = amino-propionic acid, acetic acid; R ₃ , R ₅ = I, Br, Me; R ₃ '= Ipr, I) is shown below (for a representative ligand T ₃ ) same:

F = criterion (always satisfied)

A = adjustable

Based on the methods and data described herein, embodiments of the computer-assisted method of the present invention are as follows, based on the interaction between the structure of the amino acid residues of the receptor LBD and the four different ligands described herein. It allows the design of nuclear receptor ligands. Small molecular structures for ligands are available from the Cambridge Structure Database (CSD), and three-dimensional models can be constructed using the methods described throughout the specification. Factors to consider in designing synthetic ligands include:

1) Histidine 381 acts as a hydrogen bond acceptor for R ₄ 'hydroxyls with an optimal tautomer maintained by water molecules. See FIGS. 23 and 24. Histidine is the only hydrophilic residue in this hydrophobic pocket that encloses a R ₄ 'substituent. Histidine may be a hydrogen bond acceptor or donor depending on the tautomer state. Hydrogen bond donors are preferred, but for example, if methoxy is present at the R ₄ 'position of the ligand, it may also be acceptable to be a hydrogen bond acceptor.

2) Arginine 228, 262 and 266 interact with the electrostatic interactions provided by arginine 266 (as in the triac complex) directly and through water mediated hydrogen bonding with the R ₁ -substituent. Polar pockets are shown in FIGS. 23-25. 23 shows T ₃ in the TRα ligand binding cavity, wherein the amino-propion R 1 -substituent of T ₃ interacts with Arg228, HOH502, H9H503 and HOH504 by hydrogen bonding. FIG. 24 shows triacs in ligand binding cavities where the -COOH R ₁ substituent is present in the polar pocket. In FIG. 24, Arg228 no longer shares a hydrogen bond with the ligand, but the -COOH R ₁ substituent forms a hydrogen bond with Arg266. FIG. 25 superimposes T ₃ and triac in ligand binding cavities and shows several positionally invariant amino acid and water molecules, and optionally altered interacting amino acid and water molecules. The three figures show parts of the polar pocket that can change and those parts that do not move upon binding of different ligands. For example, Arg 262 at the top of the polar pocket does not move even when the R ₁ substituent is changed from -COOH to an aminopropionic acid group. However, the other two arginines, Arg228 and Arg266, proved to be flexible in polar pockets that respond to changes in size or chemical properties of the R ₁ substituent.

3) Internal and external pockets for R ₃ / R ₅ substituents include Ser260, Ala263, Ile299; And Phe218, Ile221, Ile222, respectively. See FIGS. 21 and 22. The inner pocket is filled with R ₃ or R ₅ substituents regardless of the size of the substituents and can act as binding determinants by placing the ligand in the receptor. Optimally, the internal pocket amino acids interact with R ₃ or R ₅ substituents that are not larger than the iodine group. If the inner pocket is filled with R ₃ substituents, the outer pocket interacts with the R ₅ substituents and vice versa. The outer pocket can be scaled to the size of the substituent via a backbone motion centered at the break point of Helix 3 (Lys220-Ile221), which results in the bending of H3 and the motion of the N-terminal portion of H3 to ligand binding. Imply that change can be represented. The outer pocket has greater flexibility in the inner pocket in the sense of accepting a larger group of substituents.

4) R ₃ '- pocket of a substituent is formed by Phe215, Gly290, Met388. Pocket R ₃ '- becomes incompletely filled by the iodine substituent, the flexible Met388 accommodates a branched chain and slightly larger 3'-isopropyl substituent by movement of the H7 / H8 loop. Such pockets can accommodate R ₃ '-substituents, for example phenyl groups, which are much larger than isopropyl.

This information will facilitate the design of high affinity antagonists and antagonists by improving automated QSAR methodologies and educational manual models of pharmaceutical lead compounds. For example, the inclusion of discrete water molecules provides a complete description of hydrogen bonds in polar pockets to address drug class development; In addition, the identification of mobile and non-mobile residues in the receptors presents physically reasonable limits for use in molecular dynamics / kinetic calculations.

Example 12 Design of Ligands with Increased Affinity

The direct interaction between the receptor and the ligand is localized in the polar pocket and interacts with the R ₁ substituent. Lack of complementarity may contain a relationship for biological regulation and also provides an opportunity to increase affinity by maximizing the interaction between the amino acid in the polar pocket and the R ₁ substituent of the synthetic ligand. The receptor-ligand interaction structures described herein allow for synthetic ligands with two complementary changes to be designed to have increased affinity:

1) Remove positively charged amines. The strong positive electrostatic potential expected for the polar pocket suggests that positively charged amines in the aminopropionic acid R ₁ substituents may be detrimental to binding. Suitable groups for substitution include adjacent hydrogen bond partners; For example, given by the properties of Thr 275 O or Ser 277 N (see table in Appendix 2). For example, any negatively charged substituents, including carboxylates, carbonyls, phosphonates and sulfates having from 0 to 4 carbon atoms, will be suitable for interacting with amino acids in polar pockets. Another example of a R ₁ substituent is oxamic acid, which replaces the amine of the natural ligand with one or more carbonyl groups.

2) Incorporation of hydrogen bond acceptors and donor groups into the R ₁ substituents provides extensive interaction with the polar pocket backbone. The hydrogen bond acceptor and donor groups incorporated into the R ₁ substituent allow interaction with water molecules in polar pockets that would otherwise occur. Specific water includes HOH 504 (hydrogen bond with Ala 225 O and Arg 262 NH); And HOH503 (Asn 179 OD1, Ala 180 N and hydrogen bonds), both of which include all four complexes (TR LBD complexed with T3, TR LBD complexed with IpBr ₂ , LBD complexed with Dimit and triac Combined TR LBD). Hydrogen bond network analysis in polar pockets suggests replacement of HOH 504 with a hydrogen bond acceptor and HOH 503 with a hydrogen bond donor (although the asparagine's chemical properties will allow flexibility at each site). As such, incorporating a hydrogen bond acceptor into an R ₁ substituent that can replace HOH 504, or incorporating a hydrogen bond acceptor into an R ₁ substituent that can positionally replace HOH 503, or a combination thereof is novel. It is a method for synthesizing TR ligand.

These two design methods can be used separately or in combination to design synthetic ligands, including the methods of Table 4 below.

The conclusion for this approach is to design interactions specific for residues Arg262 and Asn179. The goal is to complete the interaction for these residues by designing ligands with R ₁ substituents that form hydrogen bonds with water molecules or charged residues in polar pockets.

Table 4. Synthetic TR Ligands

The binding of the substituents in the biphenyl ether backbone structures described above can form pharmaceutically useful ligands for thyroid hormone receptors. These novel ligands may be antagonists of thyroid receptors.

The methods for constructing synthetic ligands using the computer described herein are summarized below:

In the above formula

A is a hydrogen bond acceptor,

D is a hydrogen bond donor,

O is -OH, -CO,

R ₁₀ may be -OH, -CO,

R ₂₀ may be -CO,

R ₃₀ may be —COOH, —CONH ₂ .

See also the table for synthetic TR ligands.

Table 3.LBD-122 / 410

All publications and patent applications mentioned in this specification are incorporated by reference in the same manner as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference. In these documents nuclear receptor ligands, in particular TR ligands, are incorporated herein by reference, but may be optionally excluded from the claimed compounds.

The upper part and the upper part are provided for the convenience of the reader and should not be used to explain the meaning of the terms used below the upper part and the upper part.

The present invention is now fully described, and it will be apparent to those skilled in the art that many changes and modifications can be made without departing from the spirit or scope of the appended claims.

The present invention provides for the use of computers to provide better nuclear receptors, in particular nuclear receptor synthetic ligands based on the three-dimensional structure of thyroid receptors.

Appendix 2

Interaction of Dimit with the thyroid hormone receptor amino acids

Description of Table XX. The table above describes the interaction with Dimit (DMT). The top of the column is as follows:

# 1: Atom of Dimit that interacts with amino acids of the receptor. They are also numbered in FIG.

# 2: amino acid at full length γTRα interacting with ligand.

# 3: Naming of atoms in amino acids where interaction occurs (standard nomenclature)

# 4: distance at A between dimit and protein atoms

Description of Table XX. This table describes the interactions with the triacs. The top of the column is as follows:

# 1: Triac atoms interacting with amino acids in their receptors. They are also numbered in FIG.

# 2: amino acid at full length γTRα interacting with ligand.

# 3: Naming of atoms in amino acids where interaction occurs (standard nomenclature)

# 4: distance in A between triac and protein atoms

Description of Table XX. The table above describes the interaction with IpBr ₂ . The top of the column is as follows:

# 1: Atom of IpBr ₂ that interacts with amino acids of the receptor. They are also numbered in FIG.

# 2: amino acid at full length γTRα interacting with ligand.

# 3: Naming of atoms in amino acids where interaction occurs (standard nomenclature)

# 4: Distance in A between IpBr ₂ and protein atoms

Description of Table XX. The table above describes the interaction with T3. The top of the column is as follows:

# 1: Atom of T3 that interacts with the amino acid of the receptor. They are also numbered in FIG.

# 2: amino acid at full length γTRα interacting with ligand.

# 3: Naming of atoms in amino acids where interaction occurs (standard nomenclature)

# 4: distance in A between T3 and protein atoms

Coordination structure of thyroid hormone receptor and dimit

Coordination structure of thyroid hormone receptor and triac

Coordination Structure of Tyloid Hormone Receptor and IpBr ₂

Coordination Structure of Troid Hormone Receptor and T ₃

Claims (22)

(1) TR LBD (Thyroid hormone receptor ligand binding domain) and TR LBD ligand; or (2) Determination of TR LBD comprising TR LBD without TR LBD ligand, diffracted at 2.0 to 3.0 microsecond resolution, and having crystal stability within 5% of unit cell dimensions. The crystal of claim 1, wherein the TR LBD has at least 200 amino acids. The method of claim 2, wherein the TR LBD is selected from the group consisting of TR amino acid sequences 122-410 of rat TR-α, TR amino acid sequences 157-410 of rat TR-α, and TR amino acid sequences 202-461 of human TR-β. A crystal characterized by being derived from the TR protein. The crystal of claim 2, wherein the TR LBD ligand is a compound of the formula: wherein the TR LBD ligand has an apparent Kd of 1 μM or less in binding to TR LBD: In the above formula, R ₁ is —O—CH ₂ CO ₂ H, —NHCH ₂ CO ₂ H, —CO ₂ H, —CH ₂ CO ₂ H, —CH ₂ CH ₂ CO ₂ H, —CH ₂ CH ₂ CH ₂ CO ₂ H , -CH ₂ CH (NH ₂ ) CO ₂ H, -CH ₂ CH [NHCOCHφ ₂ ] CO ₂ H, -CH ₂ CH [NHCO (CH ₂ ) ₁₅ CH ₃ ] CO ₂ H, -CH ₂ CH [NH- FMOC] CO ₂ H, -CH ₂ CH [NH-tBOC] CO ₂ H, or a carboxylate bound to the ring by 0 to 3 carbon linkers, -PO ₃ H ₂ , -CH ₂ PO ₃ H ₂ , -CH ₂ CH ₂ PO ₃ H ₂ , -CH ₂ CHNH ₂ PO ₃ H ₂ , -CH ₂ CH [NHCOCHφ ₂ ] PO ₃ H ₂ , -CH ₂ CH [NHCO (CH ₂ ) ₁₅ CH ₃ ] PO ₃ H ₂ , -CH ₂ CH [NH-FMOC] PO ₃ H ₂ , -CH ₂ CH [NH-tBOC] PO ₃ H ₂ , or 0 to 3 carbons Phosphonates or phosphates bound to the ring by a linker, -SO ₃ H, -CH ₂ SO ₃ H, -CH ₂ CH ₂ SO ₃ H, -CH ₂ CHNH ₂ SO ₃ H, -CH ₂ CH [NHCOCHφ ₂ ] SO ₃ H, -CH ₂ CH [NHCO (CH ₂ ) a ring bound to the ring by ₁₅ CH ₃ ] SO ₃ H, -CH ₂ CH [NH-FMOC] SO ₃ H, -CH ₂ CH [NH-tBOC] SO ₃ H, or 0 to 3 carbon linkers. Fight or sulfate, When bound to TR, it acts as a functional equivalent of CH ₂ CH (NH ₂ ) CO ₂ H in T3 in the molecular recognition domain, wherein R ₁ may be substituted or unsubstituted by an amine; R ₂ is H, halogen, CF ₃ , OH, NH ₂ , SH, CH ₃ or -Et, or When bound to TR, it acts as a functional equivalent of H in the molecular recognition domain; R ₃ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —N ₃ , —SH, —CH ₃ or —Et, or When bound to TR, it acts as a functional equivalent of I in the molecular recognition domain; R ₅ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —N ₃ , —SH, —CH ₃ or —Et, or is functional equivalent of I in the molecular recognition domain when bound to TR Acts as R ₃ may be the same as R ₅ ; R ₆ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —SH, or —CH _3, or When bound to TR, it acts as a functional equivalent of H in the molecular recognition domain, and R ₂ may be identical to R ₆ ; R ₂ ′ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —N ₃ , —SH, —CH ₃ or —Et, or is functional equivalent of H in the molecular recognition domain when bound to TR Acts as water; R ₃ ′ is a hydrophobic group comprising halogen, —CF ₃ , —SH, alkyl, aryl, 5-membered heterocycle, 6-membered heterocycle, or cyano, or the functionality of I in the molecular recognition domain when bound to TR Acts as an equivalent; R ₄ ′ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —NH ₃ , —N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate or sulfate, —SH, —CH ₃ Is an alkyl, aryl, 5-membered heterocyclic aromatic group or 6-membered heterocyclic aromatic group bonded to O, N or S via a urea or carbamate bond in the -Et, or R ₄ 'position, or bonded to TR Act as functional equivalents of OH in the molecular recognition domains, if any; R ₅ ′ is —H, —OH, —NH ₂ , —N (CH ₃ ) ₂ , —SH, —NH ₃ , —N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate, sulfate, carbon number 1 to 9 branched or straight chain alkyl, substituted or unsubstituted aryl, wherein substituted aryl is substituted with halogen or alkyl with 1 to 5 carbon atoms, and aryl is bonded or unbound to the ring by -CH _2- ), Aromatic heterocycle of 5 or 6 members, wherein the heterocycle is -OH, -NH ₂ , -SH, -NH ₃ , -N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate or May be substituted with one or more groups selected from the group consisting of sulfate), heteroalkyl, heteroaryl, arylalkyl, heteroaryl alkyl, polyaromatic groups or polyheteroaromatic groups, wherein R ₅ ′ is substituted with a polar or charged group Can be; R ₆ ′ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —SH, —CH ₃ or —Et, or when bound to TR and acts as a functional equivalent of H in the molecular recognition domain , X is O, S, SO ₂ , NH, NR ₇ , CH ₂ , CHR ₇ or CR ₇ R ₇ , wherein R ₇ is an alkyl, aryl, 5-membered heterocyclic aromatic group or 6-membered heterocyclic aromatic group . The unit cell dimension of claim 2, wherein the crystal of TR LBD ligand has a unit cell dimension of a = 117.00 ± 2%, b = 80.00 ± 2% and c = 63.00 ± 2% at a β angle of 120.00 ^o ± 2%. And a monoclinic space group C2. The crystal of claim 5, wherein the crystal is coordinating the TR LBD ligand and the crystal is selected from the group consisting of a crystal having one of the following properties: (1) Unit cell dimensions (a): a = 117.16, b = 80.52, c = 63.21, monoclinic space group C2, 2.2 Hz resolution and 87.0% completeness at β angle of 120.58 ^o , (2) unit cell dimensions: a = 117.19, b = 80.20, c = 63.23, monoclinic space group C2, 2.0 ms resolution and 82.4% completeness at β angle of 120.60 ^o , and (3) Unit cell dimensions: a = 117.18, b = 80.12, c = 63.13, monoclinic space group C2, 2.2 Hz resolution and 93.7% completeness at β angle of 120.69 ^o . The crystal of claim 2, wherein the crystal further comprises a human protein. The method of claim 7, wherein the crystal is 90.00 ^o of the α angle in the γ angle of 90.00 ^o of the β angle and ^{120.00 o a = b = 68.448 ±} 2% , and c = 130.559 ± 2% having a unit cell dimensions (Å) of , A trigonal space group p3 (1) 21. The crystal of claim 8, wherein the crystal further comprises a human protein. Compounds having the following formula and having an apparent Kd of 1 μM or less for the TR LBD ligand to bind TR LBD: In the above formula, R ₁ is —O—CH ₂ CO ₂ H, —NHCH ₂ CO ₂ H, —CO ₂ H, —CH ₂ CO ₂ H, —CH ₂ CH ₂ CO ₂ H, —CH ₂ CH ₂ CH ₂ CO ₂ H , -CH ₂ CH (NH ₂ ) CO ₂ H, -CH ₂ CH [NHCOCHφ ₂ ] CO ₂ H, -CH ₂ CH [NHCO (CH ₂ ) ₁₅ CH ₃ ] CO ₂ H, -CH ₂ CH [NH- FMOC] CO ₂ H, -CH ₂ CH [NH-tBOC] CO ₂ H, or a carboxylate bound to the ring by 0 to 3 carbon linkers, -PO ₃ H ₂ , -CH ₂ PO ₃ H ₂ , -CH ₂ CH ₂ PO ₃ H ₂ , -CH ₂ CHNH ₂ PO ₃ H ₂ , -CH ₂ CH [NHCOCHφ ₂ ] PO ₃ H ₂ , -CH ₂ CH [NHCO (CH ₂ ) ₁₅ CH ₃ ] PO ₃ H ₂ , -CH ₂ CH [NH-FMOC] PO ₃ H ₂ , -CH ₂ CH [NH-tBOC] PO ₃ H ₂ , or 0 to 3 carbons Phosphates or phosphonates bound to the ring by a linker, or -SO ₃ H, -CH ₂ SO ₃ H, -CH ₂ CH ₂ SO ₃ H, -CH ₂ CHNH ₂ SO ₃ H, -CH ₂ CH [NHCOCHφ ₂ ] SO ₃ H, -CH ₂ CH [NHCO (CH ₂ ) Sulfate bound to the ring by ₁₅ CH ₃ ] SO ₃ H, -CH ₂ CH [NH-FMOC] SO ₃ H, -CH ₂ CH [NH-tBOC] SO ₃ H, or 0 to 3 carbon linkers. Or sulfite, When bound to TR, it acts as a functional equivalent of CH ₂ CH (NH ₂ ) CO ₂ H in T3 in the molecular recognition domain, and R ₁ may be substituted or unsubstituted by an amine; R ₂ is H, halogen, CF ₃ , OH, NH ₂ , SH, CH ₃ or -Et, or when bound to TR, acts as a functional equivalent of H in the molecular recognition domain; R ₃ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —N ₃ , —SH, —CH ₃ or —Et, or is functional equivalent of I in the molecular recognition domain when bound to TR Acts as; R ₅ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —N ₃ , —SH, —CH ₃ or —Et, or is functional equivalent of I in the molecular recognition domain when bound to TR Acts as R ₃ may be the same as R ₅ ; R ₆ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —SH or —CH _3, or when bound to TR acts as a functional equivalent of H in the molecular recognition domain, and R ₂ is May be the same as R ₆ ; R ₂ ′ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —N ₃ , —SH, CH ₃ or —Et, or functional equivalent of H in the molecular recognition domain when bound to TR Acts as; R ₃ ′ is a hydrophobic group comprising halogen, —CF ₃ , —SH, alkyl, aryl, 5-membered heterocycle, 6-membered heterocycle, or cyano, or the functionality of I in the molecular recognition domain when bound to TR Acts as an equivalent; R ₄ ′ is —H, halogen, —CF ₃ , —OH, —NH ₂ , NH ₃ , —N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate or sulfate, —SH, —CH ₃ , An alkyl, aryl, 5-membered heterocyclic aromatic group or 6-membered heterocyclic aromatic group bonded to O, N or S via a urea or carbamate linkage at the -Et, or R ₄ 'position, or bonded to TR; In the case of a functional equivalent of OH in the molecular recognition domain; R ₅ ′ is —H, —OH, —NH ₂ , —N (CH ₃ ) ₂ , —SH, —NH ₃ , —N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate, sulfate, carbon number 1 to 9 branched or straight chain alkyl, substituted or unsubstituted aryl, wherein substituted aryl is substituted with halogen or alkyl with 1 to 5 carbon atoms, and aryl is bonded or unbound to the ring by -CH _2- ), Aromatic heterocycle of 5 to 6 atoms, wherein the heterocycle is -OH, -NH ₂ , -SH, -NH ₃ , -N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate or sulfate It may be substituted with one or more groups selected from the group consisting of), heteroalkyl, arylalkyl, heteroaryl alkyl, polyaromatic group or polyheteroaromatic group, R ₅ 'may be substituted with a polar or charged group; R ₆ ′ is —H, halogen, —CF ₃ , —OH, —NH ₂ , —SH, —CH ₃ or —Et, or when bound to TR and acts as a functional equivalent of H in the molecular recognition domain , X is O, S, SO ₂ , NH, NR ₇ , CH ₂ , CHR ₇ or CR ₇ R ₇ , wherein R ₇ is an alkyl, aryl, 5-membered heterocyclic aromatic group or 6-membered heterocyclic aromatic group . The compound of claim 10, wherein R ₁ is a carboxylate, phosphonate, phosphate or sulfite, bonded to the ring by 0 to 3 carbon linkers, R ₂ is H, R ₃ is -I, -Br or -CH ₃ , R ₅ is -I, -Br or -CH ₃ , R ₆ is -H, R ₂ 'is -H, R ₃ ′ is —I, —Br, —CH ₃ , —iPr, —phenyl, benzyl, 5 membered heterocycle or 6 membered heterocycle, R ₄ ′ is —OH, NH ₂ or —SH, R ₅ ′ is —H, —OH, —NH ₂ , —N (CH ₃ ) ₂ , —SH, —NH ₃ , —N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate, sulfate, carbon number 1 to 9 branched or straight chain alkyl, substituted or unsubstituted aryl, wherein substituted aryl is substituted with halogen or alkyl with 1 to 5 carbon atoms, and aryl is bonded or unbound to the ring by -CH _2- ), Aromatic heterocycle of 5 to 6 atoms, wherein the heterocycle is -OH, -NH ₂ , -SH, -NH ₃ , -N (CH ₃ ) ₃ , carboxylate, phosphonate, phosphate or sulfate It may be substituted with one or more groups selected from the group consisting of), heteroalkyl, arylalkyl, heteroaryl alkyl, polyaromatic group or polyheteroaromatic group, R ₅ 'may be substituted with a polar or charged group; R ₆ ′ is —H. The compound of claim 11, wherein the compound is (1) determining one or more interacting amino acids of nuclear receptor LBD that interact with one or more first chemical residues of the bound ligand, using a three-dimensional model of a crystallized protein comprising nuclear receptor LBD with bound ligand Doing; And (2) a first chemistry to produce a second chemical moiety having a structure that increases or decreases the interaction between the interacting amino acid and the second chemical moiety as compared to the interaction between the interacting amino acid and the first chemical moiety; A computer-aided method of designing a nuclear receptor synthetic ligand, comprising selecting one or more chemical modifications of residues, wherein a three-dimensional model is created by comparing isoform ligand derivatives to produce improved phasing Prepared by the process of The compound of claim 11, wherein the compound is (1) determining the structure of the molecular recognition domain of the agent using a three-dimensional model of a crystallized protein comprising nuclear receptor LBD; And (2) selecting one or more chemical modifications of the agent beyond the binding site to the agent and providing a ligand structure extending in the direction of one or more protein domains important for nuclear receptor function, the nuclear receptor from the nuclear receptor agent A computer-aided method of designing an antagonist, wherein the protein domain is (a) the transcriptional activation domain of LBD, (b) the repressor binding domain of LBD, (c) the DNA binding domain of nuclear receptor, (d) thermal shock of nuclear receptor. A protein binding domain, (e) a dimerization domain of LBD, or (f) a hinge region for a DNA binding domain. 12. The compound of claim 11, wherein the compound is a TR antagonist. 12. The compound of claim 11, wherein the compound is a TR agonist. 12. The compound of claim 11, wherein the compound is a TR α selective ligand. 12. The compound of claim 11, wherein the compound is a TR β selective ligand. A pharmaceutical composition having selective thyromimetic activity comprising the compound of claim 11 and a pharmaceutically effective carrier. 19. The composition of claim 18, wherein R ₅ 'is alkyl having 1 to 9 carbon atoms and is straight or branched. A pharmaceutical composition for reducing the ratio of LDL-cholesterol levels to HDL-cholesterol levels, comprising the compound of claim 11 and a pharmaceutically effective carrier. A pharmaceutical composition for lowering plasma lipid levels, comprising the compound of claim 11 and a pharmaceutically effective carrier. A pharmaceutical composition for treating thyroid hormone deficiency in a patient with reduced cardiac function, comprising the compound of claim 11 and a pharmaceutically effective carrier.