Prospective identification of biologically active structures by topomer shape similarity searching

Journal of Medicinal Chemistry

Volumn 42, Issue 19, 1999, Pages 3919-3933

  Cramer, Richard D ^a   Poss, Michael A ^b   Hermsmeier, Mark A ^b   Caulfield, Thomas J ^b   Kowala, Mark C ^b   Valentine, Maria T ^b  

^a TRIPOS INC   (United States)

^b BRISTOL MYERS SQUIBB   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ANGIOTENSIN 2 RECEPTOR ANTAGONIST; ANGIOTENSIN ANTAGONIST;

ARTICLE; DRUG CONFORMATION; DRUG STRUCTURE; DRUG SYNTHESIS; UNITED STATES;

BENZYL ALCOHOLS; BROMIDES; DRUG DESIGN; MODELS, CHEMICAL; MOLECULAR MIMICRY; MOLECULAR STRUCTURE; PEPTIDE LIBRARY; RETROSPECTIVE STUDIES; STEREOISOMERISM; STRUCTURE-ACTIVITY RELATIONSHIP;

EID: 0033598416     PISSN: 00222623     EISSN: None     Source Type: Journal    
DOI: 10.1021/jm990159q     Document Type: Article

References (55)

1. "Drug" is intended herein as a convenient abbreviation referring to any substance having a useful effect on any biological system.
2. Babine, R. E.; Bender, S. L. Chem. Rev. 1997, 47, 1359-1472.
3. (a) Hansch, C. Acc. Chem. Res. 1969, 2, 232.
4. (b) Cramer, R. D., III; Patterson, D. E.; Bunce, J. D. Comparative Molecular Field Analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins. J. Am. Chem. Soc. 1988, 110, 5959-5967. Many excellent discussions of current issues are in
5. (c) 3D-QSAR in Drug Design: Recent Advances and
6. (d) 3D QSAR in Drug Design: Ligand-Protein Interactions and Molecular Similarity; Kubinyi, H., Folkers, G., Martin, Y. C., Eds.; KLUWER/ESCOM, Kluwer Academic Publishers: Dordrecht, The Netherlands, 1997.
7. 6 structures (a good leading reference is Wang, T.; Zhou, J. J. Chem. Inf. Comput. Sci. 1998, 38, 71-77), and "de novo design", generation of (often synthetically challenging) structures under constraints, usually to fit well into an experimentally determined structure of a receptor cavity ( Rotstein, S. H.; Murcko, M. A. GroupBuild: a fragment-based method for de novo drug design. J. Med. Chem. 1993, 36, 1700; DeWitte, R. S.; Shaknovich, E. I. SMoG: de novo design method based on simple, fast, and accurate free energy estimates. 1. Methodology and supporting evidence. J. Am. Chem. Soc. 1996, 118, 11733-11737.).
8. 6 structures (a good leading reference is Wang, T.; Zhou, J. J. Chem. Inf. Comput. Sci. 1998, 38, 71-77), and "de novo design", generation of (often synthetically challenging) structures under constraints, usually to fit well into an experimentally determined structure of a receptor cavity ( Rotstein, S. H.; Murcko, M. A. GroupBuild: a fragment-based method for de novo drug design. J. Med. Chem. 1993, 36, 1700; DeWitte, R. S.; Shaknovich, E. I. SMoG: de novo design method based on simple, fast, and accurate free energy estimates. 1. Methodology and supporting evidence. J. Am. Chem. Soc. 1996, 118, 11733-11737.).
9. 6 structures (a good leading reference is Wang, T.; Zhou, J. J. Chem. Inf. Comput. Sci. 1998, 38, 71-77), and "de novo design", generation of (often synthetically challenging) structures under constraints, usually to fit well into an experimentally determined structure of a receptor cavity ( Rotstein, S. H.; Murcko, M. A. GroupBuild: a fragment-based method for de novo drug design. J. Med. Chem. 1993, 36, 1700; DeWitte, R. S.; Shaknovich, E. I. SMoG: de novo design method based on simple, fast, and accurate free energy estimates. 1. Methodology and supporting evidence. J. Am. Chem. Soc. 1996, 118, 11733-11737.).
10. Apart from the current study, to the best of our knowledge the only other prospective medicinal chemistry experiment reported, where evaluation of a general structure selection methodology was the primary objective, is Kick, E. K.; Roe, D. C.; Skillman, A. G.; Liu, G.; Ewing, T. J. A.; Sun, Y.; Kuntz, I. D.; Ellman, J. A. Structure-Based Design and Combinatorial Chemistry Yield Nanomolar Inhibitors of Cathepsin-D. Chem. Biol. 1997, 4, 297-307.
11. However, for exploring aromatic substituent variations in this situation, the Topliss tree continues to be a valuable tool, which though used manually is based on "classical QSAR" principles. Topliss, J. G. J. Med. Chem. 1972, 15, 1006.
12. 3 tests of randomly chosen molecules) caused by a known concentration of a defined substance.
13. In the absence of any such tendency for "similar" compounds to have similar properties, no rational strategy for compound selection can exist.
14. For example, molecular weight, though very useful in separation, has no "neighborhood behavior" for biological properties. Accordingly, few chemists would "follow up" a "hit" structure having a molecular weight of 375.2 by eliminating from consideration structures having a molecular weight, for example, less than 325.2 or greater than 425.2.
15. Patterson, D. E.; Cramer, R. D.; Ferguson, A. M.; Clark, R. D.; Weinberger, L. E. Neighborhood behavior: a useful concept for validation of molecular diversity descriptors. J. Med. Chem. 1996, 39, 3049-3059.
16. Matter, H. Selecting optimally diverse compounds from structure databases: a validation study of two-dimensional and three-dimensional molecular descriptors. J. Med. Chem. 1997, 40, 1219-1229.
17. Brown, R. D.; Martin, Y. C. Use of structure-activity data to compare structure-based clustering methods and descriptors for use in compound selection. J. Chem. Inf. Comput. Sci. 1996, 36, 572-584.
18. Cramer, R. D.; Clark, R. D.; Patterson, D. E.; Ferguson, A. M. Bioisosterism as a molecular diversity descriptor: steric fields of single topomeric conformers. J. Med. Chem. 1996, 39, 3060-3069.
19. Patani, G. A.; LaVoie, E. J. Bioisosterism: a rational approach in drug design. Chem. Rev. 1996, 96, 3147-3176.
20. Cramer, R. D.; Patterson, D. E.; Clark, R. D.; Soltanshahi, F.; Lawless, M. S. Virtual libraries: a new approach to decision making in molecular discovery research. J. Chem. Inf. Comput. Sci. 1998, 6, 1010-1023.
21. Carr, G. The pharmaceutical industry: the alchemists. The Economist, Feb. 21, 1998, special section, pp 1-5.
22. ChemSpace is a licensed trademark of Tripos, Inc.
23. Ferguson, A. M.; Patterson, D. E.; Garr, C. D.; Underiner, T. L. Designing Chemical Libraries for Lead Discovery. J. Biomol. Screening 1996, 1, 65-73.
24. This compendium of all commercially offered compounds may be obtained from MDL Information Systems, Inc., 140 Catalina Street, San Leandro, CA 94577.
25. The actual query fragment for c contained a -COOH instead of the tetrazole found in the query sartans. As would be expected for such a widely recognized bioisosteric pair (shape difference of 38 units), however, this difference would have had very little effect on the search results. However, the shape differences critical to this work, those used to evaluate the results shown, are based on the corrected structure, the one containing the tetrazole.
26. Rusinko, A., III; Skell, J. M.; Balducci, R.; McGarity, C. M.; Pearlman, R. S. Program available from Tripos, Inc., St. Louis, MO.
27. Because this sum over grid points is a root square sum, not a simple sum, these shape differences increase less rapidly than does a volume difference, such as that produced by the SYBYL MVOLUME command. For example, the shape difference between pentyl and methyl is twice the shape difference between ethyl and methyl, not 4 times.
28. The lattice used in all topomer calculations has Angstrom extents of -4 to 14, -12 to 6, and -8 to 10 along the X, Y, and Z axes respectively, the fragments being rooted at the origin and conformation generation biased to emphasize negative Y and especially positive X space.
29. 3. However, in standard CoMFA, field values at multiple lattice points are combined essentially by a straight summation, rather than by the root-mean-square calculation used here with topomeric fields.
30. The inherent uncertainty in any steric shape difference calculated by these methods is no less than 20-30 units. This much uncertainty arises from the coarseness of the lattice point grid (2.0 Å spacing) over which the field value differences are summed.
31. It must be emphasized that the medicinal chemists did not believe that useful A-II antagonism was very likely to result from 13 of these side chains, based on their own previous experience in synthesizing prospective A-II antagonists by conventional means, as well as the published literature. However, our intent was to simulate a hit follow-up experiment, in which such additional SAR information would be lacking. In this situation, such a repositioning of a critical functional group (here the acidic carboxyl or tetrazole) has been a very common tactic. Such a tactic also facilitates the reuse of experience (the results of validation or "rehearsal" studies) gained with the first synthesis and therefore becomes even more attractive in combinatorial syntheses.
32. Technology available from IRORI, Torrey Pines Science Park, 11149 N. Torrey Pines Rd., La Jolla, CA 92037.
33. 50 values reported for the first seven compounds in column A of Table 3 are, in order: 141, 171, 158, 206, 131, 540, and 324 nM. The four query structures are roughly 100 times more potent.
34. It was not possible to obtain a more direct assessment of the systematic error for this assay. An isolated erroneous value, such as the single -48% in Table 3, was almost certainly a localized failure in correctly performing the assay, caused, for example, by some malfunction of the robot liquid handler.
35. 2value of 60 with 4 degrees of freedom.
36. 2 values, is because a single straight line does not well describe the flat part of the curve, the values at or near 0%, which dominate the leftmost data sub-blocks.
37. Perhaps the reason for observing "neighborhood behavior" among such low Tanimoto coefficients is an unusually small number of common bits set as the sartan 2D fingerprints are generated, because of the almost "undecorated" biphenyl moieties common to all structures.
38. The apparently "significant" observations in blocks D and E of Table 4 actually have the wrong sign (by suggesting that increasing electrotopological index and molecular weight differences are associated with decreasing similarities in biological properties).
39. Furthermore, the shape similar neighbors of weakly active structures were frequently more active. Of course, this result is almost a logical necessity, since the similarity of A to B is the same as the similarity of B to A, so B can lead to A or A to B, regardless of which is the more potent or which is first identified.
40. Even with a tested protocol at hand, synthesis could be attempted for only 44 of 631 ChemSpace-designated a rings and was carried through successfully for about 2/3 of the full combinatorial array. Therefore the original concern, that a less conservative synthetic strategy might not generate enough useful data, seems to have been justified.
41. Andrews-Cramer, K.; Cramer, R. Manuscript in preparation. Note, however, that these "successes" are almost entirely computational, since the only available experimental information on the biological properties of the structures found by shape similarity searching is any other SAR information in the article from which the query was taken.
42. Duncia, J. V.; Carini, D. J.; Chiu, A. T.; Johnson, A. L.; Price, W. A.; Wong, P. C.; Timmermans, P. B. M. W. The discovery of DuP 753, a potent, orally active nonpeptide angiotensin II receptor antagonist. Med. Res. Rev. 1992, 12, 149-191.
43. Prendergast, K.; Adams, K.; Greenlee, W. J.; Nachbar, R. B.; Patchett, A. A.; Underwood. D. J. Derivation of a 3D pharmacophore model for the angiotensin-II site one receptor. J. Comput.-Aided Mol. Des. 1994, 8, 491-511.
44. The current version of the topomeric shape searching software actually generates a shape difference of 186 units between the a rings of irbesartan and losartan. This mistake occurs because the topomeric conformation rules rotate the a rings 180° away from their most similar relative superposition, with their butyl side chains ending up on opposite sides of the a rings. However, such a risk of a "false negative" result in topomeric shape similarity searching is currently being alleviated by generating additional "topomeric" conformations for the occasional query structures which have great asymmetry in shape but low asymmetry in numbers of heavy atoms.
45. Ganellin, C. R. In Medicinal Chemistry - The Role of Organic Chemistry in Drug Research; Ganellin, C. R., Roberts, S. M., Eds.; Academic Press: London, 1993; pp 227-255.
46. Internal study by S. Guessregen of Tripos. Many additional examples of topomer shape similarity among ligands binding to the same or similar receptors have very recently been generated (K. Andrews-Cramer, to be presented at the Fall American Chemical Society National Meeting, 1999, New Orleans).
47. To a first approximation, the number of products increases by almost a factor of 10 for each increase of 10 units in the shape similarity searching radius.
48. Bradbury, R. H.; Allott, C. P.; Dennis, M.; Fisher, E.; Major, J. S.; Masek, B. B.; Oldham, A. A.; Pearce, R. J.; Rankins, N.; Revill, J. R.; Roberts, D. A.; Russell, S. T. New nonpeptide angiotensin II receptor antagonists. 2. Synthesis, biological properties, and structure-activity relationships of 2-alkyl-4-(biphenylylmethoxy)quinoline derivatives. J. Med. Chem. 1992, 35, 4027-4038. Masek, B. B.; Merchant, A.; Matthew, J. B. Molecular shape comparison of angiotensin II receptor antagonists. J. Med. Chem. 1993, 36, 1230-1238. Lin, H.-S.; Rampersand, A. A.; Zimmerman, K.; Steinberg, M. I.; Boyd, D. B. Nonpeptide angiotensin II receptor antagonists: synthetic and computational chemistry of N-[[4-[2(2H-tetrazol-5-yl)-a-cycloalken-1-yl]phenyl]methyl]imidazole derivatives and their in vitro activity. J. Med. Chem. 1992, 35, 2658-2667. Perkins, T. D. J.; Dean, P. M. An exploration of a novel strategy for superimposing several flexible molecules. J. Comput.-Aided Mol. Des. 1993, 7, 155-172. Also see ref 35.
49. Bradbury, R. H.; Allott, C. P.; Dennis, M.; Fisher, E.; Major, J. S.; Masek, B. B.; Oldham, A. A.; Pearce, R. J.; Rankins, N.; Revill, J. R.; Roberts, D. A.; Russell, S. T. New nonpeptide angiotensin II receptor antagonists. 2. Synthesis, biological properties, and structure-activity relationships of 2-alkyl-4-(biphenylylmethoxy)quinoline derivatives. J. Med. Chem. 1992, 35, 4027-4038. Masek, B. B.; Merchant, A.; Matthew, J. B. Molecular shape comparison of angiotensin II receptor antagonists. J. Med. Chem. 1993, 36, 1230-1238. Lin, H.-S.; Rampersand, A. A.; Zimmerman, K.; Steinberg, M. I.; Boyd, D. B. Nonpeptide angiotensin II receptor antagonists: synthetic and computational chemistry of N-[[4-[2(2H-tetrazol-5-yl)-a-cycloalken-1-yl]phenyl]methyl]imidazole derivatives and their in vitro activity. J. Med. Chem. 1992, 35, 2658-2667. Perkins, T. D. J.; Dean, P. M. An exploration of a novel strategy for superimposing several flexible molecules. J. Comput.-Aided Mol. Des. 1993, 7, 155-172. Also see ref 35.
50. Bradbury, R. H.; Allott, C. P.; Dennis, M.; Fisher, E.; Major, J. S.; Masek, B. B.; Oldham, A. A.; Pearce, R. J.; Rankins, N.; Revill, J. R.; Roberts, D. A.; Russell, S. T. New nonpeptide angiotensin II receptor antagonists. 2. Synthesis, biological properties, and structure-activity relationships of 2-alkyl-4-(biphenylylmethoxy)quinoline derivatives. J. Med. Chem. 1992, 35, 4027-4038. Masek, B. B.; Merchant, A.; Matthew, J. B. Molecular shape comparison of angiotensin II receptor antagonists. J. Med. Chem. 1993, 36, 1230-1238. Lin, H.-S.; Rampersand, A. A.; Zimmerman, K.; Steinberg, M. I.; Boyd, D. B. Nonpeptide angiotensin II receptor antagonists: synthetic and computational chemistry of N-[[4-[2(2H-tetrazol-5-yl)-a-cycloalken-1-yl]phenyl]methyl]imidazole derivatives and their in vitro activity. J. Med. Chem. 1992, 35, 2658-2667. Perkins, T. D. J.; Dean, P. M. An exploration of a novel strategy for superimposing several flexible molecules. J. Comput.-Aided Mol. Des. 1993, 7, 155-172. Also see ref 35.
51. Bradbury, R. H.; Allott, C. P.; Dennis, M.; Fisher, E.; Major, J. S.; Masek, B. B.; Oldham, A. A.; Pearce, R. J.; Rankins, N.; Revill, J. R.; Roberts, D. A.; Russell, S. T. New nonpeptide angiotensin II receptor antagonists. 2. Synthesis, biological properties, and structure-activity relationships of 2-alkyl-4-(biphenylylmethoxy)quinoline derivatives. J. Med. Chem. 1992, 35, 4027-4038. Masek, B. B.; Merchant, A.; Matthew, J. B. Molecular shape comparison of angiotensin II receptor antagonists. J. Med. Chem. 1993, 36, 1230-1238. Lin, H.-S.; Rampersand, A. A.; Zimmerman, K.; Steinberg, M. I.; Boyd, D. B. Nonpeptide angiotensin II receptor antagonists: synthetic and computational chemistry of N-[[4-[2(2H-tetrazol-5-yl)-a-cycloalken-1-yl]phenyl]methyl]imidazole derivatives and their in vitro activity. J. Med. Chem. 1992, 35, 2658-2667. Perkins, T. D. J.; Dean, P. M. An exploration of a novel strategy for superimposing several flexible molecules. J. Comput.-Aided Mol. Des. 1993, 7, 155-172. Also see ref 35.
52. For more discussion of this problem, see: Martin, Y. C. in ref 3c, pp 3-23.
53. Carini, D. J.; Wells, G. J.; Duncia, J. V. Eur. Pat. Appl. 323,-841, 1989. Ross, B. C.; Middlemiss, D.; Eldred, C. D.; Montana, J. G.; Shah, P. Eur. Pat. Appl. 446, 062, 1991.
54. Carini, D. J.; Wells, G. J.; Duncia, J. V. Eur. Pat. Appl. 323,-841, 1989. Ross, B. C.; Middlemiss, D.; Eldred, C. D.; Montana, J. G.; Shah, P. Eur. Pat. Appl. 446, 062, 1991.
55. 12 "drug-like" structures routinely take a few hours on standard workstations. (For comparison, this number of structures is approximately 20 000 times greater than the number of structures currently registered by Chemical Abstracts Services.)