A general strategy for selecting high-affinity zinc finger proteins for diverse DNA target sites

Science

Volumn 275, Issue 5300, 1997, Pages 657-661

  Greisman, Harvey A ^a   Pabo, Carl O ^a  

^a HOWARD HUGHES MEDICAL INSTITUTE   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

DNA BINDING PROTEIN; ZINC FINGER PROTEIN;

ARTICLE; BASE PAIRING; BINDING AFFINITY; DISSOCIATION CONSTANT; GENE TARGETING; NUCLEOTIDE SEQUENCE; PRIORITY JOURNAL; PROTEIN DNA INTERACTION; PROTEIN STRUCTURE; TATA BOX;

AMINO ACID SEQUENCE; BASE COMPOSITION; BASE SEQUENCE; BINDING SITES; DNA; DNA-BINDING PROTEINS; GENES, P53; HYDROGEN BONDING; MODELS, MOLECULAR; MOLECULAR SEQUENCE DATA; NUCLEIC ACID CONFORMATION; PEPTIDE LIBRARY; PROTEIN CONFORMATION; PROTEIN ENGINEERING; PROTEIN STRUCTURE, SECONDARY; RECEPTORS, CYTOPLASMIC AND NUCLEAR; TATA BOX; TRANSCRIPTION FACTORS; ZINC FINGERS;

EID: 0031053362     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.275.5300.657     Document Type: Article

References (52)

1. J. Miller, A. D. McLachlan, A. Klug, EMBO J. 4, 1609 (1985).
2. B. A. Christy, L. F. Lau, D. Nathans, Proc. Natl. Acad. Sci. USA 85, 7857 (1988).
3. J. Nardelli, T. Gibson, P. Charnay, Nucleic Acids Res. 20, 4137 (1992); S. K. Thukral, M. L. Morrison, E. T. Young, Mol. Cell. Biol. 12, 2784 (1992); J. R. Desjarlais and J. M. Berg, Proteins 12, 101 (1992); ibid. 13, 272 (1992); Proc. Natl. Acad. Sci. U.S.A. 89, 7345 (1992); ibid. 91, 11099 (1994).
4. J. Nardelli, T. Gibson, P. Charnay, Nucleic Acids Res. 20, 4137 (1992); S. K. Thukral, M. L. Morrison, E. T. Young, Mol. Cell. Biol. 12, 2784 (1992); J. R. Desjarlais and J. M. Berg, Proteins 12, 101 (1992); ibid. 13, 272 (1992); Proc. Natl. Acad. Sci. U.S.A. 89, 7345 (1992); ibid. 91, 11099 (1994).
5. J. Nardelli, T. Gibson, P. Charnay, Nucleic Acids Res. 20, 4137 (1992); S. K. Thukral, M. L. Morrison, E. T. Young, Mol. Cell. Biol. 12, 2784 (1992); J. R. Desjarlais and J. M. Berg, Proteins 12, 101 (1992); ibid. 13, 272 (1992); Proc. Natl. Acad. Sci. U.S.A. 89, 7345 (1992); ibid. 91, 11099 (1994).
6. J. Nardelli, T. Gibson, P. Charnay, Nucleic Acids Res. 20, 4137 (1992); S. K. Thukral, M. L. Morrison, E. T. Young, Mol. Cell. Biol. 12, 2784 (1992); J. R. Desjarlais and J. M. Berg, Proteins 12, 101 (1992); ibid. 13, 272 (1992); Proc. Natl. Acad. Sci. U.S.A. 89, 7345 (1992); ibid. 91, 11099 (1994).
7. J. Nardelli, T. Gibson, P. Charnay, Nucleic Acids Res. 20, 4137 (1992); S. K. Thukral, M. L. Morrison, E. T. Young, Mol. Cell. Biol. 12, 2784 (1992); J. R. Desjarlais and J. M. Berg, Proteins 12, 101 (1992); ibid. 13, 272 (1992); Proc. Natl. Acad. Sci. U.S.A. 89, 7345 (1992); ibid. 91, 11099 (1994).
8. J. Nardelli, T. Gibson, P. Charnay, Nucleic Acids Res. 20, 4137 (1992); S. K. Thukral, M. L. Morrison, E. T. Young, Mol. Cell. Biol. 12, 2784 (1992); J. R. Desjarlais and J. M. Berg, Proteins 12, 101 (1992); ibid. 13, 272 (1992); Proc. Natl. Acad. Sci. U.S.A. 89, 7345 (1992); ibid. 91, 11099 (1994).
9. E. J. Rebar and C. O. Pabo, Science 263, 671 (1994).
10. Y. Choo and A. Klug, Proc. Natl. Acad. Sci. U.S.A. 91, 11163 (1994); ibid., p. 11168.
11. Y. Choo and A. Klug, Proc. Natl. Acad. Sci. U.S.A. 91, 11163 (1994); ibid., p. 11168.
12. A. C. Jamieson, S. H. Kim, J. A. Wells, Biochemistry 33, 5689 (1994); H. Wu, W. P. Yang, C. F. Barbas, Proc. Natl. Acad. Sci. U.S.A. 92, 344 (1995).
13. A. C. Jamieson, S. H. Kim, J. A. Wells, Biochemistry 33, 5689 (1994); H. Wu, W. P. Yang, C. F. Barbas, Proc. Natl. Acad. Sci. U.S.A. 92, 344 (1995).
14. W. E. Taylor et al., Biochemistry 34, 3222 (1995); C. Cheng and E. T. Young, J. Mol. Biol. 251, 1 (1995).
15. W. E. Taylor et al., Biochemistry 34, 3222 (1995); C. Cheng and E. T. Young, J. Mol. Biol. 251, 1 (1995).
16. J. R. Desjarlais and J. M. Berg, Proc. Natl. Acad. Sci. U.S.A. 90, 2256 (1993).
17. Y. Choo, I. Sánchez-García, A. Klug, Nature 372, 642 (1994).
18. N. P. Pavletich and C. O. Pabo, Science 252, 809 (1991).
19. L. Fairall, J. W. R. Schwabe, L. Chapman, J. T. Finch, D. Rhodes, Nature 366, 483 (1993).
20. N. P. Pavletich and C. O. Pabo, Science 261, 1701 (1993).
21. M. Elrod-Erickson, M. A. Rould, L. Nekludova, C. O. Pabo, Structure 4, 1171 (1996).
22. d's have been in the micromolar range (8, 9). Subtle, context-dependent interactions (which provided the motivation for our protocol) may have a critical, cumulative effect when optimizing multifinger proteins; A modest (10-fold) increase in affinity for each finger may yield a substantial (1000-fold) increase in affinity for a three-finger protein.
23. 7) were pooled, and purified DNA from this pool was used to construct the finger 2 library. After the second selection step (Fig. 2B), a similar protocol was used to remove the remaining wild-type finger from the selected pool and to construct the finger 3 library. To accommodate the restriction sites used in these cloning steps (29), we changed residues in the COOH-terminal linker of each randomized finger to TGESR (30) for one round of selections; wild-type residues were restored when the next cassette was added.
24. Our protocol actually was designed so that a sublibrary of successful zinc finger sequences could be carried over from one selection step (Fig. 2, A or B) to the next (15). Preliminary sequencing data to analyze the "evolutionary history" of our selections (29) indicates that a set of finger 1 sequences was carried over into the step in Fig. 2B and that this step then selects for combinations of fingers that work well together (19).
25. E. J. Rebar, H. A. Greisman, C. O. Pabo, Methods Enzymol. 267, 129 (1996).
26. 9 (= 262,144) possible 9-bp sites would be present, with an average concentration of about 40 nM.
27. Each set of proteins exhibits a clear gradient of sequence diversity across the three fingers (Fig. 3), but the finger 1 and finger 2 sequences were more diverse at intermediate stages of the optimization protocol (16, 29). For example, after the first step (Fig. 2A), many of the TATA clones had Asn residues at position -1 or position 6 or in both locations. After the selections indicated in Fig. 2B, most clones had Gln at position -1 and Thr at position 6 of finger 1, and these residues also are present in a homologous natural finger that recognizes the same subsite (25).
28. Based on the Zif268 (Fig. 1B) and Tramtrack (11) structures, our alignments assume that residues at position -1 can contact the 3′ base on the primary strand of the subsite, residues at position 3 can contact the central base, and residues at position 6 can contact the 5′ base. Guanine bases in our sites appear to prefer Arg at positions -1 and 6, but His or Lys at position 3. Adenine bases appear to prefer Asn at position 3, but prefer Gln at position -1 and, to some extent, at position 6.
29. Zinc finger regions were subcloned into pET21 d (Novagen), and the corresponding peptides (with end points as in Fig. 1A) were expressed in E. coli BL21 (DE3) and purified as described (4).
30. d values were calculated from the slopes of Scatchard plots and represent the average of three independent experiments (SD values were all <60%); and (v) mobility shift assays were performed with double-stranded oligonucleotides containing TTT overhangs at the 5′ end of each strand. The sequences of the primary strands within the duplex regions were 5′-AGGGGGGCTATA-AAAGGGGGT-3′ (TATA box), 5′-GCTGTTGGGA-CATGTTCGTGA-3′ (p53 site), 5′-GCCGTCAAGG-GTTCAGTGGGG-3′ (NRE site), and 5′-CCAGTAG-CGGGGGCGTCCTCG-3′ (Zif268 site).
31. 8 (= 65,536), This indicates that our proteins - like Zif268 itself - can effectively specify 7 to 8 bp in the target DNA sites.
32. J. M. Berg, Proc. Natl. Acad. Sci. U.S.A. 89, 11109 (1992); R. E. Klevit, Science 253, 1367 (1991); M. Suzuki, M. Gerstein, N. Yagi, Nucleic Acids Res. 22, 3397 (1994).
33. J. M. Berg, Proc. Natl. Acad. Sci. U.S.A. 89, 11109 (1992); R. E. Klevit, Science 253, 1367 (1991); M. Suzuki, M. Gerstein, N. Yagi, Nucleic Acids Res. 22, 3397 (1994).
34. J. M. Berg, Proc. Natl. Acad. Sci. U.S.A. 89, 11109 (1992); R. E. Klevit, Science 253, 1367 (1991); M. Suzuki, M. Gerstein, N. Yagi, Nucleic Acids Res. 22, 3397 (1994).
35. Several of the subsites recognized by our optimized fingers (Fig. 3) also happen to appear in binding sites for the Tramtrack (11) and Gfi-1 zinc finger proteins [P. A. Zweidler-McKay, H. L. Grimes, M. M. Flubacher, P. N. Tsichlis, Mol. Cell. Biol. 16, 4024 (1996)], and we find remarkable similarities in the amino acid sequences of the corresponding recognition helices. These homologies include, but are not limited to, the canonical base-contacting residues (20) at positions -1,3, and 6. For example, finger 4 of the Gfi-1 protein and finger 1 of our NRE proteins appear to recognize the subsite 3′-ACT-5′, and the Gfi-1 residues at positions -1, 1, 2, 3, 5, and 6 are QKSDKK (underlined residues match the consensus in the selected fingers) (30). Finger 5 of Gfi-1 and finger 1 of the TATA proteins appear to recognize the subsite 3′-AAA-5′, and the corresponding Gfi-1 residues are QSSNIT (30).
36. Given the remarkable homology with Tramtrack (Fig. 3), it seems plausible that the Ser and Asp residues at position 2 in NRE fingers 2 and 3 may make the same contacts that corresponding residues make in Tramtrack fingers 1 and 2 (11). We also anticipate that the Lys at position 1 in finger 1 of the TATA box proteins may make a phosphate contact analogous to the contact made by Tramtrack finger 2.
37. There are several examples of zinc fingers that have appropriate residues (Arg, His, Asn, or Gln) at positions -1,3, and 6, but do not make the expected coded contacts with their 3-bp subsites. Examples include some natural fingers, such as finger 3 of GLI (12) and finger 2 of ADR1 (7), as well as synthetic fingers designed to recognize particular subsites (3). As noted by others (3, 7), context-dependent interactions may explain these effects.
38. J. L. Pomerantz, P. A. Sharp, C. O. Pabo, Science 267, 93 (1995).
39. H. A. Greisman, thesis, Massachusetts Institute of Technology, Cambridge, MA (1997).
40. 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.
41. W. S. El-Deiry et al., Cell 75, 817 (1993); W. S. El-Deiry et al., Cancer Res. 55, 2910 (1995).
42. W. S. El-Deiry et al., Cell 75, 817 (1993); W. S. El-Deiry et al., Cancer Res. 55, 2910 (1995).
43. A. H. Swirnoff and J. Milbrandt, Mol. Cell. Biol. 15, 2275 (1995).
44. E. B. Ziff and R. M. Evans, Cell 15, 1463 (1978).
45. J. A. A. Ladias and S. K. Karathanasis, Science 251, 561 (1991).
46. Phage display was performed in an anaerobic chamber to ensure proper folding of the zinc fingers (4, 17). Five to eight cycles of selection and amplifcation were performed for each finger, and retention efficiencies plateaued at values ranging from ∼0.2 to 3% of input phage (17, 29). Binding reactions for the p53 finger 3 selections contained the nonbiotinylated duplex competitor 5′-CCCTTGGAACATGTTCCTGATCGC-GG-3′ (29). [The p53 target site is pseudosymmetric (Fig. 1C) (31), and we wanted to avoid inadvertently selecting a zinc finger protein that would bind to the opposite strand] The biotinylated sites used in the TATA box selections are shown in Fig. 2, and the sites used for the other selections (29) were designed in a similar manner; we altered the Zif268 subsites when they were no longer needed (Fig. 2, B and C) and removed any cryptic binding sites that resembled the binding site of interest.
47. The pZif12 phagemid display vector (17) encodes a fusion protein that contains (i) Zif268 fingers 1 and 2 [residues 327 to 391 of the intact protein (2)]; (ii) a linker that introduces an amber codon; and (iii) residues 23 to 424 of the M13 gene III protein. The zinc finger region contains a set of restriction sites that were designed to facilitate the multiple cloning steps in our protocol (29).
48. Four of the eight p53 clones had a conservative Ser → Thr mutation at position - 2 in finger 2; in all other clones, residues outside the randomized regions were identical to those in wild-type Zif268.
49. L. Nekludova, personal communication.
50. C. A. Kim and J. M. Berg, Nature Struct. Biol. 3, 940 (1996).
51. In the alternative arrangement, p53 finger 2 spans a 4-bp subsite (3-ACAG-5′) and finger 3 recognizes the adjacent 3′-GGT-5′ subsite. A similar spacing occurs at one point in the GLI-DNA complex (12).
52. We thank E. Rebar for support, encouragement, reagents, and advice that made this project possible; M. Elrod-Erickson for sharing refined coordinates of the Zif268-DNA complex and for advice on purification of zinc finger peptides; L. Nekludova for extensive discussions about modeling studies with these new zinc finger proteins and for help with computer graphics; W. El-Deiry for providing sequences before publication; and J. Pomerantz, S. Wolfe, E. Fraenkel, M. Elrod-Erickson, and E. Rebar for insightful comments on the manuscript. H.A.G. was supported by a predoctoral fellowship from the Howard Hughes Medical Institute and by a Massachusetts Institute of Technology Department of Biology fellowship from the Centocor Corporation; C.O.P. was supported by the Howard Hughes Medical Institute.