WO2008155054A1

WO2008155054A1 - Farnesoid-x-receptor mutants, and crystallisation thereof

Info

Publication number: WO2008155054A1
Application number: PCT/EP2008/004672
Authority: WO
Inventors: Joerg Benz; Dominique Burger; Catherine Joseph; Markus Rudolph; Armin Ruf; Ralf Thoma
Original assignee: F. Hoffmann-La Roche Ag
Priority date: 2007-06-20
Filing date: 2008-06-11
Publication date: 2008-12-24

Abstract

The invention provides an isolated polypeptide comprising a Farnesoid - X - receptor ligand binding domain (FXR-LBD), wherein the sequence of the ligand binding domain differs at least in one amino acid located at the surface of the ligand binding domain from the wildtype sequence.

Description

FARNESOID-X-RECEPTOR MUTANTS, AND CRYSTALLISATION THEREOF

The present invention relates to surface mutants of the Farnesoid-X-receptor ligand binding domain (FXR LBD) and their uses.

The Farnesoid-X-receptor (FXR) is a member of the nuclear hormone receptor superfamily of transcription factors. FXR was originally identified as a receptor activated by farnesol, and subsequent studies revealed a major role of FXR as a bile acid receptor.

FXR is expressed in liver, intestine, kidney, and the adrenal gland. Four splice isoforms have been cloned from humans.

Among the major bile acids, chenodeoxycholic acid is the most potent FXR agonist. Binding of bile acids or synthetic ligands to FXR induces the transcriptional expression of small heterodimer partner (SHP), an atypical nuclear receptor family member that binds to several other nuclear hormone receptors, including LRH-I and LXRα and blocks their transcriptional functions. CYP7A1 and CYP8B are enzymes involved in hepatic bile acid synthesis. FXR represses their expression via activation of the SHP pathway. FXR directly induces the expression of bile acid-exporting transporters for the ABC family in hepatocytes, including the bile salt export pump (ABCBIl) and the multidrug resistance associated protein 2 (ABCC2). FXR knockout mice have impaired resistance to bile acid- induced hepatotoxicity and synthetic FXR agonists have been shown to be hepatoprotective in animal models of cholestasis. These data show that FXR protects hepatocytes from bile acid toxicity by suppressing both cellular synthesis and import of bile acids and stimulating their biliary excretion.

The process of enterohepatic circulation of bile acids is also a major regulator of serum cholesterol homeostasis. After biosynthesis from cholesterol in the liver, bile acids are secreted with bile into the lumen of the small intestine to aid in the digestion and absorption of fat and fat-soluble vitamins. The ratio of different bile acids determines the hydrophilicity of the bile acid pool and its ability to solubilize cholesterol. FXR activation increases the hydrophilicity of the pool, decreasing the intestinal solubilization of cholesterol, effectively blocking its absorption. Decreased absorption would be expected to result in lowering of plasma cholesterol levels. Indeed direct inhibitors of cholesterol absorption such as ezetimibe decrease plasma cholesterol, providing some evidence to support this hypothesis. However ezetimibe has limited efficacy, which appears due to feedback upregulation of cholesterol synthesis in cells attempting to compensate for depletion of cholesterol. Recent data have shown that FXR opposes this effect in part by directly repressing the expression of HMGCoA reductase via a pathway involving SHP and LRHl. FXR also decreases hepatic synthesis of triglycerides by repressing Spebpl-c expression by an alternate pathway involving SHP and LXRα. Thus compounds which modulate FXR activity may show superior therapeutic efficacy on plasma cholesterol and triglyceride lowering than current therapies.

Most patients with coronary artery disease have high plasma levels of atherogenic LDL. The HMGCoA reductase inhibitors (statins) are effective at normalizing LDL-C levels but reduce the risk for cardiovascular events such as stroke and myocardial infarction by only about 30%. Additional therapies targeting further lowering of atherogenic LDL as well as other lipid risk factors such as high plasma triglyceride levels and low HDL-C levels are needed.

A high proportion of type 2 diabetic patients in the United States have abnormal concentrations of plasma lipoproteins. The prevalence of total cholesterol > 240 mg/dl is 37% in diabetic men and 44% in diabetic women and the prevalence for LDL-C > 160 mg/dl are 31% and 44%, respectively in these populations. Diabetes is a disease in which a patient's ability to control glucose levels in blood is decreased because of partial impairment in the response to insulin. Type II diabetes (T2D), also called non-insulin dependent diabetes mellitus (NIDDM), accounts for 80-90% of all diabetes cases in developed countries. In T2D, the pancreatic Islets of Langerhans produce insulin but the primary target tissues (muscle, liver and adipose tissue) develop a profound resistance to its effects. The body compensates by producing more insulin ultimately resulting in failure of pancreatic insulin-producing capacity. Thus T2D is a cardiovascular-metabolic syndrome associated with multiple co-morbidities including dyslipidemia and insulin resistance, as well as hypertension, endothelial dysfunction and inflammatory atherosclerosis.

The first line treatment for dyslipidemia and diabetes is a low-fat and low-glucose diet, exercise and weight loss. Compliance can be moderate and treatment of the various metabolic deficiencies that develop becomes necessary with, for example, lipid- modulating agents such as statins and fibrates, hypoglycemic drugs such as sulfonylureas and metformin, or insulin sensitizers of the thiazolidinedione (TZD) class of PPARD- agonists. Recent studies provide evidence that modulators of FXR may have enhanced therapeutic potential by providing superior normalization of both LDL-C and triglyceride levels, currently achieved only with combinations of existing drugs and, in addition, may avoid feedback effects on cellular cholesterol homeostasis. In 2003 the crystal structures of wildtype human and rat FXR-LBD were published in complex with the agonist Fexaramine (MoLCeIl. 11, 1079-1092, 2003) and the chemically modified natural ligand CDCA (Mol.Cell 11, 1093-1100, 2003), respectively. The FXR LBD folds into an α-helical sandwich consisting of 12 helices with the agonist binding pocket formed by amino acid residues from helices 3, 5, 7, 10/11 and 12. Space groups observed were P2₁2₁2 with cell axes a=99.8 A, b=107.5 A, c=69.3A and P2₁2₁2₁ with cell axes a=36.7 A, b=56.8 A, c=117.6A.

Crystallization of the FXR LBD with agonists proved to be difficult in obtaining well diffracting crystals as well as establishing a high hit rate for successful crystallization.

Therefore, there is a need for FXR LBD peptides allowing successful crystallisation and subsequent X-ray crystallograpy analysis.

In a first aspect, the present invention relates to an isolated polypeptide comprising a Farnesoid-X- Receptor ligand binding domain (FXR-LBD), wherein the sequence of the ligand binding domain differs at least in one amino acid located at the surface of the ligand binding domain from the wildtype sequence.

In a preferred embodiment, the amino acid sequence located at the surface of the FXR-LBD is selected from amino acids with a hydrophilic side chain, preferably lysine and glutamic acid. Preferably, the amino acid with a hydrophilic side chain has been replaced by alanine or arginine.

In a further preferred embodiment, the amino acid sequence differs in two amino acids from the wildtype sequence.

In a further preferred embodiment, the wildtype FXR-LBD comprises amino acids 248 - 476 of Seq. Id. No.l (Gene bank accession No. AAK60271). Preferably, glutamic acid at position 281 of Seq. Id. No.l has been replaced by alanine and glutamic acid at position 354 of Seq. Id. No.l has been replaced by alanine.

In a second aspect, the invention relates to a nucleic acid molecule encoding a polypeptide of the present invention.

In a third aspect, the present invention relates to a co-crystal of a polypeptide of the present invention and a ligand bound to the FXR-LBD, wherein the crystal has unit cell dimensions of a = 35.0 ± 3 A, b = 113 ± 3 A, c = 157 ± 3 A, α = β = γ = 90° ± 3° and the crystal belongs to space group P2₁2₁2₁.

In a preferred embodiment, the crystal has unit cell dimensions of a = 35.0 ± 2 A, b = 113 ± 2 A, c = 157 ± 2 A, α = β = γ = 90° ± 2° . In a further preferred embodiment, the crystal has unit cell dimensions of a = 35.0 ± 1 A, b = 113 ± 1 A, c = 157 ± 1 A, α = β = γ = 90°± 1° . In another preferred embodiment, the crystal has unit cell dimensions of a = 35.0 A, b = 113 A, c = 157 A, α = β = γ = 90° .

In a preferred embodiment, the ligand is l-(5-Phenyl-2H-pyrazol-3-yl)-2- thiophen-2-ylmethyl- lH-benzoimidazole.

In a fourth aspect, the present invention relates to a co-crystal of a polypeptide of the present invention and a ligand bound to the FXR-LBD, wherein the crystal belongs to space group C222i.

In a preferred embodiment, the crystal has unit cell dimensions ofa = 71 ± 3 A, b = 82 ± 3 A, c = 188 ± 3 A , α = β = γ = 90°± 3° , preferably a = 71 ± 2 A, b = 82 ± 2 A, c =

188 ± 2 A , α = β = γ = 90° ± 2°, more preferably a = 71 ± 1 A, b = 82 ± 1 A, c = 188 ± 1

A , α = β = γ = 90° ± 1°, even more preferably a = 71 A, b = 82 A, c = 188 A , α = β = γ =

90°.

In a further preferred embodiment, the ligand is (S)-2,N-Dicyclohexyl-2-[2-(4- hydroxymethyl-phenyl)-benzoimidazol-l-yl]-acetamide.

In a seventh aspect, the present invention relates to a co-crystal of a polypeptide of the present invention and a ligand bound to the FXR-LBD, wherein the crystal belongs to space group P2₁.

In a preferred embodiment, the crystal has unit cell dimensions of a = 55 ± 3 A , b = 183 ± 3 A, c = 55 ± 3 A , β = 98.4 ± 3°, preferably a = 55 ± 2 A , b = 183 ± 2 A, c = 55 ± 2 A, β = 98.4 ± 2 °, more preferably a = 55 ± 1 A , b = 183 ± 1 A, c = 55 ± 1 A , β = 98.4 ± 1°, even more preferably a = 55 A , b = 183 A, c = 55 A , β = 98.4°.

In a preferred embodiment, the ligand is (S)-2,N-Dicyclohexyl-2-{2-[4-(lH- tetrazol-5-yl)-phenyl]-benzoimidazol-l-yl}-acetamide.

In a further preferred embodiment of the present invention, the above described co-crystals further comprise a co-activator peptide, preferably the co-activator peptide has the amino acid sequence KDHQLLRYLLDKD (Seq. Id. No. 6).

In an eighth aspect, the present invention relates to a method for co-crystallising a polypeptide of the present invention with a ligand that binds to the active site of FXR- LBD, the method comprising: a) providing an aqueous solution of the polypeptide of the present invention, b) adding a molar excess of the ligand to the aqueous solution of the polypeptide, and c) growing crystals. In a preferred embodiment, the method is performed in presence of a co-activator peptide, preferably the co-activator peptide has the amino acid sequence KDHQLLRYLL DKD (Seq. Id. No. 6). The co-activator peptide is preferably present in a molar excess, more preferably in a 10 - 15 molar excess.

In a further aspect, the present invention provides a co-crystal of a polypeptide of the present invention and a ligand bound to the FXR-LBD having the structure defined by the co-ordinates of Fig. 1, Fig. 2 or Fig. 3, optionally varied by a root mean square deviation (rmsd) of less than 2.0 A.

Preferred ligands are selected from the group consisting of l-(5-Phenyl-2H- pyrazol-3-yl)-2-thiophen-2-ylmethyl-lH-benzoimidazole, (S)-2,N-Dicyclo-hexyl-2- [2- (4-hydroxymethyl-phenyl)-benzoimidazol-l-yl]-acetamide, (S)-2,N-Dicyclo-hexyl-2-{2- [4-(lH-tetrazol-5-yl)-phenyl]-benzoimidazol-l-yl}-acetamide.

Crystals of the present invention can be grown by a number of techniques including batch crystallization, vapor diffusion (either by sitting drop or hanging drop) and by microdialysis. Seeding of the crystals in some instances is required to obtain X-ray quality crystals. Standard micro- and/or macroseeding of crystals may therefore be used.

In a preferred embodiment of the invention, co-crystals are grown by vapor diffusion. In this method, the polypeptide solution is allowed to equilibrate in a closed container with a larger aqueous reservoir having a precipitant concentration optimal for producing crystals. Generally, less than about 10 μL of substantially pure polypeptide solution is mixed with an equal or similar volume of reservoir solution, giving a precipitant concentration about half that required for crystallization. This solution is suspended as a droplet underneath a coverslip, which is sealed onto the top of a reservoir. The sealed container is allowed to stand, from one day to one year, usually for about 2-6 weeks, until crystals grow.

The co-crystals of the present invention can be obtained by a method which comprises: providing a buffered, aqueous solution of 3.75 to 50 mg/ml of a polypeptide of the present invention, adding a molar excess of the ligand to the aqueous polypeptide solution, and growing crystals by vapor diffusion or microbatch using a buffered reservoir solution of 0 % to 30 % (w/v) PEG, wherein the PEG has an average molecular weight of 200 Da to 20000 Da. The PEG may be added as monomethyl ether. PEG may be used of an average molecular weight of 500 Da to 5,000 Da. The buffered reservoir solution further comprises 0 M to 2 M tri-ammonium citrate pH 7, 0 M to 1 M L-proline, 0 M to IM trimethylamine-N-oxide, 0 M to 1 M ammonium sulfate, 0 M to 1 M lithium sulfate, 0 M to 1 M ammonium acetate, 0 M to 1 M sodium or magnesium formate and 0 M to 1 M D/L-malic acid pH 7. Said microbatch may be modified.

Methods for obtaining the three-dimensional structure of the crystals described herein, as well as the atomic structure coordinates, are well-known in the art (see, e.g., D. E. McRee, Practical Protein Crystallography, published by Academic Press, San Diego (1993), and references cited therein).

The crystals of the invention, and particularly the atomic structure coordinates obtained therefrom, have a wide variety of uses. For example, the crystals and structure coordinates described herein are particularly useful for identifying compounds that bind to farnesoid-X-receptors as an approach towards developing new therapeutic agents.

The structure coordinates described herein can be used as phasing models in determining the crystal structures of additional native or mutated, as well as the structures of co-crystals of farnesoid-X-receptor with bound ligand. The structure coordinates, as well as models of the three-dimensional structures obtained therefrom, can also be used to aid the elucidation of solution-based structures of native or mutated farnesoid-X-receptors, such as those obtained via NMR. Thus, the crystals and atomic structure coordinates of the invention provide a convenient means for elucidating the structures and functions of a farnesoid-X-receptor.

Thus, the present invention also provides a method of identifying compounds that can bind to a farnesoid-X-receptor comprising the steps of: applying a 3 -dimensional molecular modeling algorithm to the atomic coordinates of a protein shown in Fig. 1, 2 or 3, ± a root mean square deviation from the backbone atoms of said amino acids of not more than 2 A; determine the spatial coordinates of the binding site of FXR-LBD, and electronically screening the stored spatial coordinates of a set of candidate compounds against the spatial coordinates of the farnesoid-X-receptor binding site to identify a compound that can bind to farnesoid-X-receptor.

In a preferred embodiment, the method comprises the steps of: generating a three dimensional model of a binding pocket of FXR-LBD using the relative structural coordinates of Fig. 1, 2 or 3 of residues ILE273, THR274, ILE277, ASN287, PHE288, ILE290, LEU291, THR292, MSE(MET)294, ALA295, HIS298, MSE(MET)332, PHE333, ARG335, SER336, ALA337, ILE339, PHE340, LEU352, ILE356, SER359, ILE361, ILE366, MSE(MET)369, PHE370, TYR373, HIS451, MSE(MET)454, LEU455, TRP458, PHE465, LEU469, TRP473 , ± a root mean square deviation (rmsd) from the backbone atoms of said amino acids of not more than 2 A; and performing computer fitting analysis to identify a compound that can bind to a FXR-LBD binding site. MSE stands for selenomethionine.

In a preferred embodiment of the method of identifying compounds that can bind to a farnesoid-X-receptor, the rmsd is less than 1 A, preferably less than 0.5 A.

The term "root mean square deviation" means the square root of the arithmetic mean of the squares of the deviations. It is a way to express the deviation or variation from a trend or object. For purposes of this invention, the "root mean square deviation" defines the variation in the backbone of a protein from the backbone of FXR-LBD or an active binding site thereof, as defined by the structure coordinates of FXR-LBD described herein.

Molecular docking of large compound databases to target proteins of known or modeled 3-dimensional structure is now a common approach in the identification of new lead compounds. This "virtual screening" approach relies on fast and accurate estimation of the ligand binding mode and an estimate of ligand affinity. Typically a large database of compounds, either real or virtual is docked to a target structure and a list of the best potential ligands is produced. This ranking should be highly enriched for active compounds which may then be subject to further experimental validation.

The calculation of the ligand binding mode may be carried out by molecular docking programs which are able to dock the ligands in a flexible manner to a protein structure. The estimation of ligand affinity is typically carried out by the use of a separate scoring function. These scoring functions include energy-based approaches which calculate the molecular mechanics force field and rule-based approaches which use empirical rules derived from the analysis of a suitable database of structural information. Consensus scoring involves rescoring each ligand with multiple scoring functions and then using a combination of these rankings to generate a hit list.

Short description of the figures

Figure 1 shows the coordinates of a crystal of human FXR LBD (amino acids 248 - 476 of Seq. Id. No. 1; E281AE354A) with agonist (S)-2,N-Dicyclohexyl-2-[2-(4- hydroxymethyl-phenyl)-benzoimidazol-l-yl]-acetamide; the coordinates of amino acids 249 - 475 of Seq. Id. No. 1 and amino acids 3 - 13 of Seq. Id. No. 6 are shown;

Figure 2 shows the coordinates of a crystal of human FXR LBD (amino acids 248 - 476 of Seq. Id. No. 1; E281AE354A) with agonist (S)-2,N-Dicydohexyl-2-[2-(4- hydroxymethyl-phenyl)-benzoimidazol-l-yl]-acetamide; the coordinates of amino acids 247 - 476 of Seq. Id. No. 1 and amino acids 2 - 12 of Seq. Id. No. 6 are shown; and Figure 3 shows the coordinates of a crystal of human FXR LBD (amino acids 248 - 476 of Seq. Id. No. 1; E281AE354A) with agonist (S)-2,N-Dicyclohexyl-2-{2-[4-(lH- tetrazol-5-yl)-phenyl]-benzoimidazol-l-yl}-acetamide; the coordinates of amino acids 247 - 475 of Seq. Id. No. 1 and amino acids 2 - 12 of Seq. Id. No. 6 are shown.

Examples

Example 1: Crystal structure of human FXR LBD (E281AE354A) with agonist 1- (5-Phenyl-2H-pyrazol-3-yl)-2-thiophen-2-yhnethyl-lH-benzoimidazole

Methods:

Cloning and purification:

DNA manipulation and sequence analysis

Preparation of DNA probes, digestion with restriction endonucleases, DNA ligation and transformation oiExoli strains were performed as described (Sambrook, ]., Fritsch, E.F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY.). Mutagenesis was performed by using the QuikChange Multi Kit from Stratagene. For DNA sequencing, the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit and ABI PRISM 310 Genetic analyzer were used. PCR were performed in the T3 Thermocycler (Whatman Biometra), using the Expand polymerase (Roche).

Production and Purification of recombinant human FXR ligand binding domain (LBD) in E.coli

The ligand binding domain of human FXR, residues 248-476 (hFXR-LBD), was amplified by PCR using a cDNA and the oligonucleotides 5'-GAC GAGC CAT ATG GAA CTC ACC CCA GAT CAA CAG-3' (with an Ndel site in bold) (Seq. Id. No. 2) and 5'-CGC GGA TCC CTA CTG CAC GTC CCA GAT TTC-3' (with a BamHI site in bold) (Seq. Id. No. 3). Using the two new restriction sites, the amplified DNA fragment was cloned into the pET15b vector (Novagen) to create a fusion with a N-terminal His- tag. The use of the Ndel restriction site added additional amino acids to the N-teπninus which would leave the amino acids glycine, serine and histidine after a thrombin digest at the N-terminus of FXR-LBD. The DNA sequence was confirmed. pET15b-hFXR(248- 476) was transformed into B121(DE3) and heterologously expressed at 20⁰C by induction with IPTG at an optical density of 0.8 at 600 nm. About 30% of the protein was in the soluble fraction of the cell homogenate. The protein was purified using sequential chromatography on Ni-NTA, thrombin digest overnight at room temperature, a second Ni-NTA to remove impurities and finally a Superdex 200 size exclusion chromatography equilibrated with 50 mM Tris/ HCl pH 7.8, 0.1 M NaCl, 3 mM TCEP, 1 mM EDTA; 10% glycerol. The purified protein had a purity >90% and was monodisperse as shown by HPLC and analytical ultracentrifugation, respectively.

The wild-type hFXR-LBD could either not be crystallized with agonists or the obtained crystals were not of sufficient quality for diffraction experiments. Therefore surface mutants have been introduced to improve the crystal packing and crystal quality. Lys and GIu residues located at the surface were mutated to Ala or Arg. If the Lys and GIu amino acid residues form clusters on the surface they were combined to double mutants offering the best opportunity for crystal contact engineering. Mutagenesis was carried out as described by Strategene for the QuikChange Multi Kit. The oligonucleotides 5'-caa ata aaa ttt taa aag aag cat tea gtg cag aag aaa att ttc-3' (Seq. Id. No. 4) and 5'-ca ttc tga cct att gga age acg gat teg aaa tag tgg tat c -3' (Seq. Id. No. 5) were used for the mutation E281A and E354A, respectively (mutated codons are marked in bold). In both cases an EcøRI restriction site was destroyed and used for identification of correct clones. Expression and purification was carried out as described for the wild-type.

For the following mentioned crystals the double mutant E281AE354A showed superior properties with respect to hit rate and quality of crystals.

Crystallization:

Protein used for crystallization of FXR-LBD together with l-(5-Phenyl-2H- pyrazol-3-yl)-2-thiophen-2-ylmethyl-lH-benzoimidazole has been purified as described above. The protein was incubated with ligand in a 12 fold molar excess for 2 hours at room temperature. A short co-activator peptide (KDHQLLRYLLDKD) (Seq. Id. No. 6) from SRC-I was added in 12-fold molar excess and incubation continued overnight at 4 °C. The final DMSO concentration in the solution was adjusted to 2% to improve crystallization. Prior to crystallization experiments the protein was centrifuged at 20000 g for 10 min and concentrated to 12mg/ml. The crystallization droplet was set up at 22 °C by mixing 0.3 μl of protein solution with 0.6 μl reservoir in vapour diffusion hanging drop experiments. Crystals appeared out of 0.1 M HEPES/NaOH pH 7.0, 2.0 M ammonium sulfate after 1 day and grew to a final size of 0.1 mm x 0.1 mm x 0.05 mm within 2 days.

Crystals were harvested with paraffin oil as cryoprotectant and then flash frozen in a IOOK N₂ stream. Diffraction images were collected at a temperature of IOOK at the beamline XlOSA of the Swiss Light Source and processed with the programs MOSFLM and and SCALA (CCP4) yielding data to 2.3 A resolution. Standard crystallographic programs from the CCP4 software suite were used to determine the structure by molecular replacement using an in-house FXR- LBD structure as search model (CCP4 (Collaborative Computational Project, N. The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760-763 (1994)). Refinement and model building cycles were performed with REFMAC and MOLOC, repectively (Table 1).

Results:

Crystals belong to space group Pl^^i with cell axes a=35.0 A, b=l 13.12 A, c=l 57.72 A, ot=β=^y=90° and contained a dimer of FXR-LBD in the aysmmeric unit. The ligand was clearly defined in the initial Fo-Fc electron density map in both monomers. The ligand is bound to the protein by two hydrogen bonds and mainly hydrophobic interactions via residues from helices 3,5,7, 10/11 and 12. The loop between helix 10/11 and helix 12 is partially disordered in one molecule in the asymmetric unit but ordered in the other due to crystal packing effects. The structure is not isomorphous to the orthorhombic crystals of human FXR-LBD with Fexaramine in the PDB entry lOSH.pdb (A chemical, genetic, and structural analysis of the nuclear bile acid receptor FXR, Mol.Cell, 11, 1079-1092 (2003)).

Table 1: Data collection and structure refinement statistics for l-(5-Phenyl-2H-pyrazol-3- vl)-2-thiophen-2- vlmethvl- 1 H-benzoimidazole co-crvstal

Data Collection l-(5-Phenyl-2H-pyrazol-3-yl)-2-thiophen-2-ylmethyl- 1 H-benzoimidazole

Wavelength (A) 1.0714

Resolution¹ (A) 2.3 (2.359-2.3)

Unique reflections¹ ^' 27273

Completeness (%)¹ 99.3 (99.3)

D K-merge ((Oy/cy \ 1 >2 7.5 (26.4)

<I/σ>^! 5.2 (2.5)

Unit Cell (Space group a=35.0 A, b=l 13.12 A, c=157.72 A, α=β=-p90° P2,2₁2₁)

Refinement

Resolution (A) 2.3 (2.359-2.3)

Rcryst ' 23.2 (25.2)

R_fce^M 29.3 (32.0)

R.m.s. deviations from ideality 0.014 / 1.833 Bond lengths (A) / angles (°) Main chain dihedral angles (%) 92.4 / 6.9 / 0.5 / 0.2

Most favored/allowed/generous/ disallowed ⁵

¹ Values in parentheses refer to the highest resolution bins.

² Rmerge-∑ I I-<I> I /∑I where I is the reflection intensity.

³ R_cry_st-∑ I F_O-<F_C> I /ΣF₀ where F₀ is the observed and F₀ is the calculated structure factor amplitude.

⁴ R_fr_ee was calculated based on 5% of the total data omitted during refinement.

⁵ Calculated with PROCHECK [Laskowski, R.A., MacArthur, M.W., Moss, D.S. & Thornton, J.M. PROCHECK: a program to check the stereochemical quality of protein structure. J. Appl. Crystallogr. 26, 283-291 (1993)].

Example 2: Crystal structure of human FXR LBD (E281AE354A) with agonist (S)-2^V-Dicyclohexyl-2-[2-(4-hydroxymethyl-phenyl)-benzoimidazol-l-yl]- acetamide

Protein used for crystallization of FXR-LBD together with (S)-2,N-Dicyclohexyl- 2-[2-(4-hydroxymethyl-phenyl)-benzoimidazol-l-yl]-acetamide has been purified as described above. The protein was incubated with ligand in a 12 fold molar excess for 2 hours at room temperature. A short co-activator peptide (KDHQLLR YLLDKD) (Seq. Id.

No. 6) from SRC-I was added in 12 fold molar excess and incubation continued overnight at 4 °C. The final DMSO concentration in the solution was adjusted to 2% to improve crystallization. Prior to crystallization experiments the protein was centrifuged at 20000 g for 10 min and concentrated to 12mg/ml. The crystallization droplet was set up at room temperature by mixing 0.3μl of protein solution with 0.6 μl reservoir solution in vapour diffusion hanging drop experiments. Crystals appeared out of 0.1 Bis-Tris pH

6.5, 25% PEG 3350 immediately after setup and grew to a final size of 0.2 mm x 0.05 mm x 0.05 mm within 2 days.

Crystals were harvested with paraffin oil as cryoprotectant and then flash frozen in a IOOK N₂ stream. Diffraction images were collected at a temperature of IOOK at the beamline XlOSA of the Swiss Light Source and processed with the programs MOSFLM and and SCALA (CCP4) yielding data to 2.5 A resolution. Standard crystallographic programs from the CCP4 software suite were used to determine the structure by molecular replacement using an in-house FXR- LBD structure as search model (CCP4 (Collaborative Computational Project, N. The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760-763 (1994)). Refinement and model building cycles weie performed with REFMAC, autoBUSTER (Acta Crystallogr., D56, 1313-1323 (2000)) and MOLOC, respectively (Table 2).

Crystals belong to space group C222_Ϊ with cell axes a= 71.84 A, b=82.96 A, c=188.97 A, α=β=γ=90° and diffracted to 2.5 A resolution. The asymmetric unit is formed by a dimer of FXR LBD. In both monomers the ligand was clearly defined in the initial Fo-Fc electron density map. The ligand is bound to the protein via one hydrogen bond and mainly hydrophobic interactions. The structure is not isomorphous to the orthorhombic crystals of human FXR LBD with Fexaramine in the PDB entry lOSH.pdb (A chemical, genetic, and structural analysis of the nuclear bile acid receptor FXR, MoLCeIl, 11, 1079-1092 (2003)) as well as to the structure in example 1.

Table 2: Data collection and structure refinement statistics for (S)-2,N-Dicvclohexyl-2- r2-(4-hvdroxymethyl-phenyl^N)-benzoimidazol- 1 -yli -acetamide co-crystal

Data Collection (S)-2,N-Dicyclohexyl-2-[2-(4-hydroxymethyl- phenyl)-benzoimidazol- 1 -yl] -acetamide

Wavelength (A) 0.97890 Resolution¹ (A) 2.5 (2.65-2.5) Unique reflections¹ 19947 Completeness (%)¹ 100 (100)

R fO/ \ 1.2 merge v '⁰J 8.3 (61.5)

<I/σ>^! 5.5 (1.2)

Unit Cell (Space group C222,) _a=71.84 A, b=82.96 A, c=l 88.97 A, ot=β^=90^c Refinement

Resolution (A) 2.5 (2.65-2.5) n 1,3 JVryst 23.2 (24.55)

R_te ^M 30.06 (27.05)

R.m.s. deviations from ideality 0.012 / 1.633 Bond lengths (A) / angles (°)

Main chain dihedral angles (%) 90.6/ 8.6 / 0.7 / 0.0

Most favored/allowed/generous/ disallowed ⁵

Values in parentheses refer to the highest resolution bins.

where I is the reflection intensity. ³ R_cry_sF∑ I F_O-<F_C> I /ΣF₀ where F₀ is the observed and F_c is the calculated structure factor amplitude.

Example 3: Crystal structure of human FXR LBD (E281AE354A) with agonist (S)-2,N-DicyclohexyI-2-{2-[4-(lH-tetrazol-5-yI)-phenyl]-benzoimidazol-l-yl}- acetamide

Protein used for crystallization of FXR-LBD together with (S)-2,N-Dicyclohexyl- 2-{2-[4-(lH-tetrazol-5-yl)-phenyl]-benzoimidazol-l-yl}-acetamide has been purified as described above. The protein was incubated with ligand in a 12-fold molar excess for 2 hours at room temperature. A short co-activator peptide (KDHQLLR YLLDKD) (Seq. Id. No. 6) from SRC-I was added in 12-fold molar excess and incubation continued overnight at 4 degress. The final DMSO concentration in the solution was adjusted to 2% to improve crystallization. Prior to crystallization experiments the protein was centrifuged at 20000 g for 10 min and concentrated to 12mg/ml. The crystallization droplet was set up at room temperature by mixing 3 μl of protein solution with 1 μl reservoir solution in vapour diffusion hanging drop experiments. Crystals appeared out of 0.1 Bis-Tris pH 5.5, 25% PEG 3350.

Crystals were harvested with paraffin oil as cryoprotectant and then flash frozen in a IOOK N₂ stream. Diffraction images were collected at a temperature of IOOK at the beamline XlOSA of the Swiss Light Source and processed with the programs DENZO and SCALEPACK (Methods in Enzymology, 276: Macromolecular Crystallography, part A, 307-326 (1997)) yielding data to 2.2 A resolution. Standard crystallographic programs from the CCP4 software suite were used to determine the structure by molecular replacement using an in-house FXR-LBD structure as search model (CCP4 (Collaborative Computational Project, N. The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760-763 (1994)). Refinement and model building cycles were performed with REFMAC and COOT (Acta Crystallogr. D 60, 2126-2132 (2004)), respectively (Table 2).

Crystals belong to space group P2j with cell axes a= 55.4 A, b=183.9 A, c=55.8 A, β=98.4° and diffracted to 2.2 A resolution. The asymmetric unit is formed by a tetramer of FXR LBD. In all monomers the ligand was clerarly defined in the initial Fo-Fc electron density map. Interactions of the ligand with the protein were similar as in example 2. The structure is not isomorphous to the orthorhombic crystals of human FXR-LBD with Fexaramine in the PDB entry lOSH.pdb (A chemical, genetic, and structural analysis of the nuclear bile acid receptor FXR, Mol.Cell, 11, 1079-1092 (2003)) as well as to the structure in example 1 and example 2.

Table 2: Data collection and structure refinement statistics for (S)-2,N-Dicyclohexyl-2- {2-r4-(lH-tetrazol-5-ylVphenyll-benzoimidazol-l-vU-acetamide co-crystal

Data Collection (S)-2,N-Dicyclohexyl-2- {2-[4-(l H-tetrazol-5-yl)- phenyl]-benzoimidazol-l -yl} -acetamide

Wavelength (A) 1 Resolution¹ (A) 35.0-2.2 (2.28-2.20) Unique reflections¹ 55434 Completeness (%)' 99.9 (99.8) merge \ /•>/ 6.0 (66.7) ,

<Vσ>¹ 5.5 (1.2)

Unit Cell (Space group P2i) a=55.4 A, b=183.9 A, c=55.8 A, β=98.4°

Refinement

Resolution (A) 33.8-2.2 (2.26-2.20)

R 4firee 26.2 (35.2)

R.m.s. deviations from ideality 0.011 / 1.183 Bond lengths (A) / angles (°)

Main chain dihedral angles (%) 95.3/ 4.7 / 0.0 / 0.0

Most favored/allowed/generous/ disallowed ⁵

¹ Values in parentheses refer to the highest resolution bins.

²

the reflection intensity.

³ R_cry_sF∑ I F_O-<F_C> I /XF₀ where F₀ is the observed and F_c is the calculated structure factor amplitude.

⁵ Calculated with PROCHECK [Laskowski, R.A., MacArthur, M.W., Moss, D.S. & Thornton, J.M. PROCFfECK: a program to check the stereochemical quality of protein structure. J. Appl. Crystallogr. 26, 283-291 (1993)].

While there are shown and described presently preferred embodiments of the invention, it is to be distinctly understood that the invention is not limited thereto but may be otherwise variously embodied and practiced within the scope of the following claims.

Claims

1. An isolated polypeptide comprising a Farnesoid-X-Receptor ligand binding domain (FXR-LBD), wherein the sequence of the ligand binding domain differs at least in one amino acid located at the surface of the ligand binding domain from the wildtype sequence.

2. The polypeptide of claim 1, wherein the amino acid sequence located at the surface of the FXR-LBD is selected from amino acids with a hydrophilic side chain, preferably lysine and glutamic acid.

3. The polypeptide of claim 2, wherein the amino acid with a hydrophilic side chain has been replaced by alanine or arginine.

4. The polypeptide of claims 1 to 3, wherein the amino acid sequence differs in two amino acids from the wildtype sequence.

5. The polypeptide of claims 1 to 4, wherein the wildtype FXR-LBD comprises amino acids 248 - 476 of Seq. Id. No.1.

6. The polypeptide of claim 4, wherein in Seq. Id. No.l glutamic acid at position 281 has been replaced by alanine and glutamic acid at position 354 has been replaced by alanine.

7. A nucleic acid molecule encoding a polypeptide of claims 1 to 6.

8. A co-crystal of a polypeptide of claims 1 to 6 and a ligand bound to the FXR-

LBD, wherein the crystal has unit cell dimensions of a = 35.0 ± 3 A, b = 113 ± 3 A, c = 157 ± 3 A, α = β = γ = 90° ± 3° and the crystal belongs to space group P2₁2₁2₁.

9. A co-crystal of a polypeptide of claims 1 to 6 and a ligand bound to the FXR- LBD, wherein the crystal belongs to space group C222i.

10. The co-crystal of claim 9, wherein the crystal has unit cell dimensions of a =

71 ± 3 A, b = 82 ± 3 A, c = 188 ± 3 A , α = β = γ = 90° ± 3°.

11. A co-crystal of a polypeptide of claims 1 to 6 and a ligand bound to the FXR- LBD, wherein the crystal belongs to space group ?2_L

12. The co-crystal of claim 11, wherein the crystal has unit cell dimensions of a = 55 ± 3 A , b = 183 ± 3 A, c = 55 ± 3 A , β = 98.4° ± 3°.

13. A method for co-crystallizing a polypeptide of claims 1 to 6 with a compound that binds to the binding site of said polypeptide, the method comprising: a) providing an aqueous solution of the polypeptide, b) adding a molar excess of a ligand to the aqueous solution of the polypeptide, and c) growing crystals.

14. A crystal obtained by the method of claim 13.

15. A method for identifying a compound that can bind to the binding site of FXR- LBD comprising the steps: a) determining an active site of FXR-LBD from the three dimensional model of FXR- LBD using the atomic coordinates of Fig. 1, 2 or 3 ± a root mean square deviation from the backbone atoms of said amino acids of not more than 2 A; and b) performing computer fitting analysis to identify a compound that can bind to the FXR-LBD active site.

16. The method of claim 15 comprising the steps: a) generating a three dimensional model of an active site of FXR-LBD using the relative structural data coordinates of Fig. 1, 2 or 3 of residues ILE273, THR274, ILE277, ASN287, PHE288, ILE290, LEU291, THR292, MSE(MET)294, ALA295, HIS298, MSE(MET)332, PHE333, ARG335, SER336, ALA337, ILE339, PHE340, LEU352, ILE356, SER359, ILE361, ILE366, MSE(MET)369, PHE370, TYR373, HIS451, MSE(MET)454, LEU455, TRP458, PHE465, LEU469, TRP473, ± a root mean square deviation from the backbone atoms of said amino acids of not more than 2 A; and b) performing computer fitting analysis to identify a compound that can bind to the FXR-LBD active site.

17. A co-crystal of a polypeptide of claims 1 to 6 and a ligand bound to the FXR- LBD having the structure defined by the coordinates of Fig. 1, Fig. 2 or Fig.3, optionally varied by an rmsd of less than 2.0 A.

18. The polypeptides, crystals and methods substantially as hereinbefore described, especially with reference to the foregoing examples.