Rational Drug Design - American Chemical Society

Chapter 9

Rational Approaches to Inhibition of Human Osteoclast Cathepsin K and Treatment of Osteoporosis 1

1

1

2

Sherin S. Abdel-Meguid , Baoguang Zhao , Ward W . Smith , Cheryl A . Hanson , Judith LaLonde , Thomas Carr , Karla D'Alessio , Michael S. McQueney , H . - J . O h , Scott K. Thompson , Daniel F . Veber , and Dennis S. Yamashita 1

Downloaded by UNIV OF MINNESOTA on May 12, 2018 | https://pubs.acs.org Publication Date: July 7, 1999 | doi: 10.1021/bk-1999-0719.ch009

3

3

2

3

1

3

2

3

2

Department of Macromolecular Sciences, Department of Protein Biochemistry, and Department of Medicinal Chemistry, SmithKline Beecham Pharmaceuticals, 709 Swedeland Road, King of Prussia, PA 19406 3

N o v e l , potent and selective human osteoclast cathepsin K inhibitors have been designed based on knowledge derived from the crystal structure o f papain bound to a tripeptide aldehyde. Cathepsin K is a thiol protease belonging to the papain superfamily. U n l i k e previously known crystal structures o f that family o f enzymes i n which ligands bind to the nonprime side o f the active site, our papain structure shows the ligand i n the prime direction. This observation and the identification o f key interactions between the protein and the ligand inspired the design o f a novel class o f inhibitors spanning both sides o f the active site. The crystal structure o f the first member o f this class bound to cathepsin K confirmed our design hypothesis. Inhibitors o f cathepsin K are potential drugs for the treatment o f osteoporosis.

Recent success i n the rational design o f novel, potent HIV-1 protease inhibitors and the subsequent verification that they are highly effective drugs, has corifirmed the important role o f rational design i n the drug discovery process. Most o f these drugs were designed based on knowledge derived from the crystal structures o f H I V protease, renin and other aspartyl proteases (1,2). M a n y other examples o f rational drug design are now available (3). Here, we w i l l describe the structure-based design o f one chemical class o f cathepsin K inhibitors that are potential drugs for the treatment o f osteoporosis and we w i l l show how the crystal structure o f an inhibitor o f cathepsin K bound to papain inspired the rational design process.

© 1999 American Chemical Society

Parrill and Reddy; Rational Drug Design ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

141

142

Bone Remodeling and Osteoporosis


Bone remodeling is a normal and dynamic process involving deposition and resorption o f bone matrix. Bone is formed by mature osteoblast cells, while osteoclasts are responsible for bone resorption. Osteoclasts are multinuclear giant cells that solubilize mineralized bone matrix through secretion o f proteolytic enzymes into an extracellular, sealed, l o w p H compartment on the bone surface. It is believed that osteoporosis, a disease characterized by low density, high porosity and brittleness o f bone, results from imbalance between bone formation (osteoblasts) and resorption (osteoclasts).

Cathepsin K and its Role in Osteoporosis Cathepsin K is a recently discovered member o f the papain superfamily o f cysteine proteinases that is selectively and highly expressed i n osteoclasts (4,5). It is secreted as a 314 amino acid proenzyme containing a 99 amino acid leader sequence (6). The proenzyme self-processes at low p H to generate the mature form (7). The crystal structures o f cathepsin K i n the presence and absence o f bound ligands have been determined (8,9). The enzyme folds into two subdomains separated by the active site cleft, a characteristic o f the papain family o f cysteine proteases. Cathepsin K is believed to play an important role i n bone resorption and is a potential therapeutic target for treatment o f diseases involving excessive bone loss such as osteoporosis. This is supported by two pieces o f evidence. One, it has been known for over a decade that classical thiol protease inhibitors such as E-64 and leupeptin inhibit bone resorption (10,11). T w o , defects i n the gene encoding cathepsin K have been linked recently to pycnodysostosis, a disease characterized by skeletal defects such as dense, brittle bones, short stature and poor bone remodeling (12).

Papain as a Surrogate for Cathepsin K The absence o f sufficient cathepsin K for crystallographic structure determination early-on i n this study compelled us to search for a suitable model. Papain, having 46% identical amino acid sequence to cathepsin K , was chosen because o f the availability o f its structure i n the presence and absence o f ligands. A number o f crystal structures o f papain with bound inhibitors had been reported (13, 14, 15). The inhibitors i n all o f these structures were found to bind on the nonprime side o f the active site (Figure l a ) . Using these structures, we modeled a number o f our d i - and tri-peptide aldehyde inhibitors into the nonprime side o f the active site o f papain and into a homology model o f cathepsin K derived from papain. These modeling studies did not explain our S A R data which showed strong preference for the presence o f a C b z or other aromatic moiety at the amino terminus o f these peptides. Thus, to rationally design inhibitors o f cathepsin K it was necessary to obtain crystal structures using our own inhibitors. Again, papain was selected because it is commercially available i n large quantities ( I C N Biomedicals #1009-24) and because its crystallization and crystallographic studies are well documented (16).



143

Figure 1. a) Ribbon drawing o f the crystal structure o f the complex o f papain with leupeptin. b) Ribbon drawing o f the papain complex with the peptide aldehyde inhibitor C b z - L e u - L e u - L e u - O H . The figure was prepared with M O L S C R I P T (22).


144


Crystal Structure of Papain/Cbz-Leu-Leu-Leu Aldehyde W e have determined the crystal structure o f papain bound to the Cbz-Leu-Leu-Leu aldehyde whose chemical structure is shown i n Figure 2. The structure o f the complex was obtained using crystals grown by vapor diffusion from a solution o f 0.1 M TrisHC1 at p H 8.5 containing 0.5 M sodium citrate and 20% P E G 600. The crystals belong to the monoclinic space group C 2 , with a=100.5 A, b=50.7 A, c=62.3 A, P=99.9° and one molecule/asymmetric unit. They grow i n a space group different from those previously reported for papain/ligand complexes. The crystal structure o f the papain/Cbz-Leu-Leu-Leu aldehyde was solved by the molecular replacement method (/ 7), using the structure o f papain ( P D B code 1PIP; 18) as a starting model. Surprisingly, the inhibitor i n our structure was found to bind on the prime side o f the active site (Figure l b ) . A major point o f interaction between the inhibitor and the protein was an edge-to-face interaction between the phenyl ring o f the inhibitor and the indole ring o f T r p l 8 1 (Figure 3). This tryptophan is conserved between papain and cathepsin K . In order to ensure that the novel binding mode observed i n the crystal structure o f the papain/Cbz-Leu-Leu-Leu aldehyde was not an artifact o f crystallization, we produced crystals o f papain bound to leupeptin under exactly the same conditions as those used for the papain/Cbz-Leu-Leu-Leu aldehyde. The crystals were isomorphous, and our structure o f the papain/leupeptin complex was nearly identical to that previously reported (Figure la), with the inhibitor bound to the nonprime subsites (SI to S3; 19) o f the enzyme.

Design of a Novel Cathepsin K Inhibitor Based on the Crystal Structure of the Papain/Cbz-Leu-Leu-Leu aldehyde The observations that inhibitors containing C b z or other aromatic groups at the amino terminus bind to the prime side o f the active site and that such binding may be facilitated by the interaction with T r p l 8 1 led to the design o f novel inhibitors spanning both sides o f the active site (Figure 4). The prototype o f this class o f inhibitors was a symmetric inhibitor that resulted from an overlay o f the Cbz-Leu-Leu-Leu aldehyde and leupeptin papain crystal structures. The two inhibitors were merged computationally by replacing their aldehyde functions with a single ketone (Figure 4). The resulting model o f a ketone-containing inhibitor was further simplified by removal o f the side chains on both sides o f the ketone moiety. This was necessary since the arginyl and leucyl sidechains occupied the same region o f space. Furthermore, a homology model o f cathepsin K derived from the structure o f papain suggested that T r p l 8 4 o f cathepsin K (Trpl77 i n papain), a highly conserved residue within the papain superfamily, would form a better aromatic-aromatic interaction with the C b z moiety. Thus, the hypothetical inhibitor was shortened by one L e u residue from the right side (Figure 4), resulting i n a yet smaller molecule. A second C b z moiety was introduced on the left side (Figure 4), as a final step to make the inhibitor truly symmetric. This was done not to mimic any symmetry i n the active site (there is none), but rather to simplify the chemical synthesis o f this initial member o f a new class o f inhibitors. This C b z group was also hypothesized to reach to Tyr67 on the


145

2a.


Leupeptin

Leu-Leu-Arg-aldehyde

2b.

Cbz-tripeptide aldehyde

Cbz-Leu-Leu-Leu-aldehyde Figure 2. Chemical structure o f a) leupeptin and b) Cbz-Leu-Leu-Leu-aldehyde.

Figure 3. Stereo view o f the active site of papain bound to Cbz-Leu-Leu-Leu-OH. Inhibitor atoms are drawn as ball-and-stick. Parrill and Reddy; Rational Drug Design ACS Symposium Series; American Chemical Society: Washington, DC, 1999.


146

Figure 4. Schematic drawing o f the design o f the symmetric diacylaminomethyl ketone inhibitor based on the crystal structures o f papain bound to leupeptin and to Cbz-Leu-Leu-Leu-aldehyde.


147

nonprime side o f the cathepsin K active site for additional aromatic-aromatic interaction. The resulting diacylaminomethyl ketone (l,3-bis[[N[(phenylmethoxy)carbonyl]-L-leucyl]amino]-2-propanone) is shown i n Figure 4.


Binding of the Novel, Symmetric Diacylaminomethyl Ketone to Cathepsin K The novel diacylaminomethyl ketone is a selective, competitive, reversible inhibitor o f cathepsin K with a K j o f 23 n M (20). Spanning both sides o f the active site has allowed for enhanced potency and selectivity by taking simultaneous advantage o f interactions on the nonprime and prime sides o f the active site, and by allowing the use o f a less reactive electrophilic carbon for attack at the cysteine. Yamashita et al. (20) have shown that this diacylaminomethyl ketone is a relatively poor inhibitor o f papain, cathepsin L , cathepsin B and cathepsin S, with Kj app o f 10,000 n M , 340 n M , 1,300 n M and 890 n M , respectively. To confirm our design hypothesis, we have determined the crystal structure o f diacylaminomethyl ketone bound to cathepsin K . Data for the structure determination were obtained from crystals grown using vapor diffusion from a solution o f 10% isopropanol, 0.1 M NaPC>4-citrate at p H 4.2. These crystals belong to the tetragonal space group P 4 3 2 i 2 , with a=57.7 A, c=131.1 A, and the unit cell contains one molecule/asymmetric unit. The crystal structure was determined using the molecular replacement method (17) and a model consisting o f all atoms from the crystal structure reported by Zhao et al. (8). The diacylaminomethyl ketone inhibitor binds i n the cathepsin K active site as predicted. It spans both sides o f the active site (Figure 5) and makes a number o f key interactions with the enzyme (Figure 6). The phenyl groups on both ends o f the inhibitor engage T r p l 8 4 and Tyr67 i n a face-face and edge-face interaction, respectively (Figure 7). The crystal structure clearly shows the inhibitor covalently attached to the enzyme at the sulfur atom o f Cys25 (the active site cysteine) as expected. The P2 leucyl sidechain o f the inhibitor fits snugly i n the hydrophobic S2 pocket defined by residues Met68, Leu209, A l a l 3 4 , A l a l 6 3 and Tyr67 (Figure 6). Hydrogen bonding interactions are seen between N D 1 o f H i s l 6 2 , N E 2 o f G l n l 9 and the backbone amide nitrogens o f Cys25 and Gly66, all o f which donate a hydrogen to oxygen atoms o f the inhibitor. The remainder o f the inhibitor interacts poorly or not at all with the enzyme indicating potential for further optimization o f this class o f inhibitors. }

Conclusions The design hypothesis generated from the crystal structures o f papain bound to C b z Leu-Leu-Leu aldehyde and to leupeptin resulted i n the design and synthesis o f a novel, potent, reversible and selective symmetric inhibitor. Confirmation was achieved through crystallographic structure determination o f the resulting diacylaminomethyl ketone inhibitor bound to cathepsin K . This i n turn led to the generation o f numerous novel inhibitors o f cathepsin K that span both sides o f the active site as described by Yamashita et al (20) and Thompson et al. (21). These new inhibitors were then optimized through iterative cycles o f structure-based design. Although the papain



148

Human Cathepsin K - Diacylaminomethyl Ketone Figure 5. Ribbon drawing o f the crystal structure o f the complex o f human cathepsin K with the symmetric diacylaminomethyl ketone.



Figure 6. Schematic view o f the interactions i n the active site o f cathepsin K with the symmetric diacylaminomethyl ketone inhibitor.



150

Figure 7. Stereo view o f the active site o f the complex o f human cathepsin K with the symmetric diacylaminomethyl ketone. Inhibitor atoms are drawn as ball-andstick.


151


structures provided pivotal insights for the design o f the first inhibitors, further generations o f inhibitors designed based on this preliminary insight required optimization through numerous crystal structures o f cathepsin K , the actual target, with bound ligands. Finally, we conclude that use o f surrogate enzymes can lead to important insights for the rational design o f novel inhibitors, but optimization requires knowledge o f the structure o f the target molecule preferably bound to inhibitors. This is reminiscent o f studies to identify renin inhibitors where surrogates such as endothiapepsin, rhizopuspepsin and penicillopepsin structures were used to design inhibitors o f renin (7).

References

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Abdel-Meguid, S. S. Medicinal Research Reviews 1993, 13, 731-778. Wlodawer, A., Erickson, J. W. Annu. Rev. Biochem. 1993, 62, 5434-5485. Babine, R. E., Bender, S.L. Chem. Rev. 1997, 97, 1359-1472. Drake, F. H., Dodds, R. A., James, I. E., Connor, J. R., Debouck, C., Richardson, S., Lee-Rykaczewski, E., Coleman, L., Rieman, D., Barthlow, R., Hastings, G., Gowen, M . J. Biol. Chem. 1996, 271, 12511-12516. Bromme, D., Okamoto, K. Biol. Chem. Hoppe-Seyler 1995, 376, 379-384. Bossard, M . J., Tomaszek, T. A., Thompson, S. K., Amegadzie, B. Y., Harming, C. R., Jones, C., Kurdyla, J. T., McNulty, D. E., Drake, F. H., Gowen, M . , Levy, M . A. J. Biol. Chem. 1996, 271, 12517-12524. McQueney, M . S., Amegadzie, B. Y., D'Alessio, K., Hanning, C. R., McLaughlin, M . M., McNulty, D., Carr, S. A., Ijames, C., Kurdyla, J., Jones, C. S. J. Biol. Chem. 1997, 272, 13955-13960. Zhao, B., Janson, C. A., Amegadzie, B. Y., D'Alessio, K., Griffin, C., Hanning, C.R., Jones, C., Kurdyla, J., McQueney, M., Qiu, X., Smith, W. W., AbdelMeguid, S. S. Nat. Struct. Biol. 1997, 4, 109-111. McGrath, M . E., Klaus, J. L., Barnes, M . G., Bromme, D. Nat. Struct. Biol. 1997, 4, 105-109. Delaisse, J. M . , Boyde, A., Maconnachie, E., Ali, N . N . , Sear, C. H. J., Eeckhout, Y., Vaes, G. Bone 1980, 8, 305-313. Delaisse, J. M . , Eeckhout, Y., Vaes, G. Biochem. Biophys. Res. Commun. 1984, 125, 441-441. Gelb, B.D., Shi, G.P., Heller, M., Weremowicz, S., Morton, C., Desnick, R.J., Chapman, H.A. Genomics 1997, 41, 258-262. Drenth, J., Kalk, K.H. & Swen, H.M. Biochemistry 1976 15, 3731-3738. Varughese, K.I., Ahmed, F.R., Carey, P.R., Hasnain, S., Huber, C.P. & Storer, A.C. Biochemistry 1989, 28, 1330-1332. Yamamoto, D., Matsumoto, K., Ohishi, H., Ishida, T., Inoue, M . , Kitamura, K. and Mizuno, H. J. Biol. Chem. 1991, 266, 14771-14777. Yamamoto, A., Tomoo, K., Doi, M , Ohishi, H., Inoue, M . , Ishida, T., Yamamoto, D., Tsuboi, S., Okamoto, H., Okada, Y . Biochemistry. 1992, 31, 11305-11309. Brunger, A. T., Kuriyan, J., Karplus, M . Science 1987, 235, 458-460. Bernstein, F. C., Koetzle, T. F., Williams, G. J., Meyer, E. E. J., Brice, M . D., Rogers, J. R., Kennard, O., Shimanouchi, T., Tasumi, M . J. Mol. Biol. 1977, 112, 535-542. Schechter, I., Berger, A. Biochem. Biophys. Res. Commun. 1967, 27, 157-162. Yamashita, D. S., Smith, W. W., Zhao, B., Janson, C. A., Tomaszek, T. A. Bossard, M . J., Levy, M . A., Marquis, R. W., Oh, H-J., Ru, Y . , Carr, T. J.,


152

21.


22.

Thompson, S. K., Ijames, C. F., Carr, S. A., McQueney, M . , D'Alessio, K. J., Amegadzie, B. Y . , Hanning, C. R., Abdel-Meguid, S. S., DesJarlais, R. L., Gleason, J. G., Veber, D. F. J. Amer. Chem. Soc., in press. Thompson, S. K., Halbert, S. M., Bossard, M . J., Tomaszek, T. A., Levy, M . A., Meek, T. D., Zhao, B., Smith W. W., Janson, C. A., D'Alessio, K. J., McQueney, M . , Abdel-Meguid, S. S., DesJarlais, R. L., Briand, J., Sarkar, S. K., Huddleston, M . , Ijames, C., Carr, S., Games, K. T., Shu, A., Heys, J. R., Bradbeer, J., Zembryki, D., Lee-Rykaczewski, L., James, I. E., Lark, M . , Drake, F. H., Gowen, M . , Gleason J. G., Veber, D.F. Proc. Natl. Acad. Sci. USA, in press. Kraulis, P. J. Appl. Crystallogr. 1991, 24, 946-950.


Rational Drug Design - American Chemical Society

Recommend Documents