Convergent solutions to binding at a protein-protein interface

Science

Volumn 287, Issue 5456, 2000, Pages 1279-1283

  DeLano, Warren L ^a,b   Ultsch, Mark H ^c   De Vos, Abraham M ^c   Wells, James A ^a,b  

^a UNIVERSITY OF CALIFORNIA   (United States)

^b Sunesis Pharmaceuticals Inc   (United States)

^c GENENTECH INC   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

IMMUNOGLOBULIN FC FRAGMENT; IMMUNOGLOBULIN G;

ARTICLE; BINDING SITE; CRYSTAL STRUCTURE; HORMONE RECEPTOR INTERACTION; HUMAN; MUTAGENESIS; PEPTIDE ANALYSIS; PRIORITY JOURNAL; PROTEIN PROTEIN INTERACTION;

AMINO ACID SEQUENCE; AMINO ACID SUBSTITUTION; BINDING SITES, ANTIBODY; CRYSTALLOGRAPHY, X-RAY; DIMERIZATION; EVOLUTION, MOLECULAR; HUMANS; HYDROGEN BONDING; IMMUNOGLOBULIN FC FRAGMENTS; IMMUNOGLOBULIN G; LIGANDS; MODELS, MOLECULAR; MOLECULAR SEQUENCE DATA; PEPTIDE LIBRARY; PEPTIDES; PROTEIN BINDING; PROTEIN CONFORMATION; PROTEIN STRUCTURE, SECONDARY; PROTEIN STRUCTURE, TERTIARY; RECEPTORS, FC; RHEUMATOID FACTOR; STAPHYLOCOCCAL PROTEIN A;

EID: 0034681465     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.287.5456.1279     Document Type: Article

References (37)

1. J. A. Wells and A. M. de Vos, Annu. Rev. Biochem. 65, 609 (1996).
2. B. C. Cunningham and J. A. Wells, Proc. Natl. Acad. Sci. U.S.A. 88, 3407 (1991).
3. Fc/Protein A [Protein Data Bank (PDB) accession code 1FC2, 2.9 Å]: J. Deisenhofer, Biochemistry 20, 2361 (1981); Fc/Protein G (1FCC, 3.5 Å): A. E. Sauer-Eriksson, G. J. Kelywegt, M. Uhlen, T. A. Jones, Structure 3, 265 (1995); Fc/rheumatoid factor (1ADQ, 3.15 Å): A. L. Corper et al., Nature Struct. Biol. 4, 374 (1997); Fc/neonatal Fc receptor (1FRT,4.5 Å): W. P. Burmeister, A. H. Huber, P. J. Bjorkman, Nature 372, 379 (1994).
4. Fc/Protein A [Protein Data Bank (PDB) accession code 1FC2, 2.9 Å]: J. Deisenhofer, Biochemistry 20, 2361 (1981); Fc/Protein G (1FCC, 3.5 Å): A. E. Sauer-Eriksson, G. J. Kelywegt, M. Uhlen, T. A. Jones, Structure 3, 265 (1995); Fc/rheumatoid factor (1ADQ, 3.15 Å): A. L. Corper et al., Nature Struct. Biol. 4, 374 (1997); Fc/neonatal Fc receptor (1FRT,4.5 Å): W. P. Burmeister, A. H. Huber, P. J. Bjorkman, Nature 372, 379 (1994).
5. Fc/Protein A [Protein Data Bank (PDB) accession code 1FC2, 2.9 Å]: J. Deisenhofer, Biochemistry 20, 2361 (1981); Fc/Protein G (1FCC, 3.5 Å): A. E. Sauer-Eriksson, G. J. Kelywegt, M. Uhlen, T. A. Jones, Structure 3, 265 (1995); Fc/rheumatoid factor (1ADQ, 3.15 Å): A. L. Corper et al., Nature Struct. Biol. 4, 374 (1997); Fc/neonatal Fc receptor (1FRT,4.5 Å): W. P. Burmeister, A. H. Huber, P. J. Bjorkman, Nature 372, 379 (1994).
6. Fc/Protein A [Protein Data Bank (PDB) accession code 1FC2, 2.9 Å]: J. Deisenhofer, Biochemistry 20, 2361 (1981); Fc/Protein G (1FCC, 3.5 Å): A. E. Sauer-Eriksson, G. J. Kelywegt, M. Uhlen, T. A. Jones, Structure 3, 265 (1995); Fc/rheumatoid factor (1ADQ, 3.15 Å): A. L. Corper et al., Nature Struct. Biol. 4, 374 (1997); Fc/neonatal Fc receptor (1FRT,4.5 Å): W. P. Burmeister, A. H. Huber, P. J. Bjorkman, Nature 372, 379 (1994).
7. N. C. Wrighton et al., Science 273, 458 (1996); O. Livnah et al., Science 273, 464 (1996).
8. N. C. Wrighton et al., Science 273, 458 (1996); O. Livnah et al., Science 273, 464 (1996).
9. k, where X is a random amino acid from an NNS codon, i + j + k = 18, and j = 4 to 10), a GGGSGGG linker, and the M13 gene VIII starting at the first residue of the mature protein.
10. Selections were done as described (5) with the following modifications: Microtiter wells were coated with IgG-Fc (5 μg/ml); Casein Blocker Buffer (Pierce) was used in place of 0.1% bovine serum albumin to better prevent nonspecific binding; elution of phage was effected with either 75 mM dithiothreitol or 0.2 mM glycine (pH 2.0) with equivalent results. IgG-Fc was obtained by papain cleavage of CD4-IgG, immunoadhesin protein [D. J. Capon et al., Nature 337, 525 (1989)]. Cleaved material was purified over Protein A-Sepharose followed by Superdex-75 (Pharmacia) and then quantified by absorbance at 280 nm.
11. Single-letter abbreviations for the amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.
12. B. Nilsson et al., Protein Eng. 1, 107 (1987).
13. Peptides were synthesized on solid phase with standard 9-fluorenylmethoxycarbonyl protocols and purified by reversed-phase chromatography. Masses were confirmed by electrospray mass spectrometry, and peptides were quantified by ultraviolet absorbance at 280 nm.
14. i.
15. The DNA sequence of the peptide was moved to a monovalent phage display format by cassette mutagenesis to give a construct with the STII signal sequence, the peptide KEASCSYWLGELVWCVAGVE, a GGGPGGG linker, and the M13 gene III protein starting at residue 253.
16. 8. These libraries were independently screened for binding to IgG-Fc for six rounds and then sequenced.
17. Three additional libraries were constructed by using the degeneracy of the genetic code to recombine the preferred amino acids at each position into one peptide. The DNA sequences for these libraries contained the following mixtures of bases (IUPAC codes): DRG GWA GMA RRC TGC KCT TRS CAC MTG GGC GAG CTG GTC TGG TGC RVC RVM BKC GAS KDW, DRS VWG SVG RRC TGC KCC TRS YRS MTG GGC GAG CTG GTC TGG TGC RNC VVS NBS GWS KDM, and DNS NNS NNS VNS TGC BVG TDS HRS MDS GGC GAG STC KKG WRG TGC RNM NNS NNS NNS NNM. These libraries were also sorted against IgG-Fc for six rounds and then sequenced.
18. d = 16 nM in 25 mM MES (pH 6.0), 0.05% Tween-20.
19. S. R. Fahnestock et al., in Bacterial Immunoglobulin-Binding Proteins (Academic Press, New York, 1990), vol. 1, chap. 11; R. Karlsson, L. Jendeberg, B. Nilsson, J. Nilsson, P. Nygren, J. Immunol. Methods 183, 43 (1995).
20. S. R. Fahnestock et al., in Bacterial Immunoglobulin- Binding Proteins (Academic Press, New York, 1990), vol. 1, chap. 11; R. Karlsson, L. Jendeberg, B. Nilsson, J. Nilsson, P. Nygren, J. Immunol. Methods 183, 43 (1995).
21. 1 residues 237 to 443 with eight sugars per monomer.
22. 1 residues 237 to 443 with eight sugars per monomer.
23. 1 residues 237 to 443 with eight sugars per monomer.
24. Surface area and geometric measurements were made with the Crystallography and NMR System (CNS) [A. T. Brünger et al., Acta Crystallogr. D. 54, 905 (1998)]. A solvent probe radius of 1.4 Å was used, and surface area changes were computed by subtracting complexed from uncomplexed solvent-accessible surface areas. Contact regions were defined as the set of atoms that lie within 5.0 Å of any nonhydrogen atom on the opposing molecule.
25. Protein A domain numbering is from H. Gouda et al., Biochemistry 31, 9665 (1992).
26. 2. Patches were randomly distributed across all of the available structures (PDB codes: 1FC1, 1FC2, 1FCC, 1ADQ, and 1DN2) and were of a random globular shape. To ensure even sampling, probabilities were weighted so that each solvent-exposed atom was included in an equal number of surface patches (∼10,000 patches per atom) The properties of each site were computed and then compared with those of the consensus binding patch on Fc.
27. L. Young, R. L. Jernigan, D. G. Covell, Protein Sci. 3, 717 (1994); S. Jones and J. M. Thornton, Proc. Natl. Acad. Sci. U.S.A. 93, 13 (1996); C. Tsai, S. L. Lin, H. J. Wolfson, R. Nussinov, Protein Sci. 6, 53 (1997); L. L. Conte, C. Chothia, J. Janin, J. Mol. Biol. 265, 2177 (1999).
28. L. Young, R. L. Jernigan, D. G. Covell, Protein Sci. 3, 717 (1994); S. Jones and J. M. Thornton, Proc. Natl. Acad. Sci. U.S.A. 93, 13 (1996); C. Tsai, S. L. Lin, H. J. Wolfson, R. Nussinov, Protein Sci. 6, 53 (1997); L. L. Conte, C. Chothia, J. Janin, J. Mol. Biol. 265, 2177 (1999).
29. L. Young, R. L. Jernigan, D. G. Covell, Protein Sci. 3, 717 (1994); S. Jones and J. M. Thornton, Proc. Natl. Acad. Sci. U.S.A. 93, 13 (1996); C. Tsai, S. L. Lin, H. J. Wolfson, R. Nussinov, Protein Sci. 6, 53 (1997); L. L. Conte, C. Chothia, J. Janin, J. Mol. Biol. 265, 2177 (1999).
30. L. Young, R. L. Jernigan, D. G. Covell, Protein Sci. 3, 717 (1994); S. Jones and J. M. Thornton, Proc. Natl. Acad. Sci. U.S.A. 93, 13 (1996); C. Tsai, S. L. Lin, H. J. Wolfson, R. Nussinov, Protein Sci. 6, 53 (1997); L. L. Conte, C. Chothia, J. Janin, J. Mol. Biol. 265, 2177 (1999).
31. Patches from (79) were ranked separately by polarity and solvent-accessible surface fraction. For each atom in the Fc dimer, the average rank of alt patches involving the atom was then computed The average atomic ranks for polarity and accessibility were then combined linearly (〈accessibility〉 - 〈polarity〉) to give a composite score incorporating both properties.
32. C. Medesan, D. Matesoi, C. Radu, V. Ghetie, E. S. Ward, J. Immunol. 158, 2211 (1997).
33. W. L. Delano, in preparation.
34. Single alanine mutants of the peptide and Protein A Z-domain were displayed morovalently on gene III, assayed by enzyme-linked immunosorbent assay, and directly compared with unmutated control peptide as described [B. C. Cunningham, D. G. Lowe, B. Li, B. D. Bennett, J. A. Wells, EMBO J. 13, 2508 (1994)].
35. 50, median effective concentration). Additional mutagenesis data are available in L. Jendeberg et al., J. Mol. Recog. 8, 270 (1995).
36. B. K. Kay, A. V. Kurakin, R. Hyde-DeRuysher, Drug Discovery Today 8, 370 (1998).
37. We thank B. C. Cunningham, J. K. Tong, and M. Dennis for assistance in the initial selection experiments against Fc; A. Braisted for training in solid phase peptide synthesis; C. Wiesmann for help with crystallographic refinement; and the SSRL for use of their facility in data collection