Directed evolution of novel binding proteins

• US 5,223,409 A
• Filed: 03/01/1991
• Issued: 06/29/1993
• Est. Priority Date: 09/02/1988
• Status: Expired due to Term

• Current Assignee: Dyax Corporation (Shire plc)
• Original Assignee: Protein Engineering Corp. (Shire plc)
• Inventors: Markland, William, Ladner, Robert C., Ley, Arthur C., Guterman, Sonia K., Roberts, Bruce L., Kent, Rachel B.
• Primary Examiner(s): Hill, Jr., Robert J.
• Assistant Examiner(s): Ulm, John D.
• Application Number:
  US07/664,989
• Time in Patent Office: 851 Days
• Field of Search: 435/69.1, 435/172.3, 435/252.3, 435/320.1, 530/350
• US Class Current: 506/1
• CPC Class Codes: A61K 38/00 Medicinal preparations containing peptides peptides containing beta-lactam rings A61K31/00; cyclic dipeptides not having in their molecule any other peptide link than those which form their ring, e.g. piperazine-2,5-diones, A61K31/00; ergot alkaloids of the cyclic peptide type A61K31/48; containing macromolecular compounds having statistically distributed amino acid units A61K31/74; medicinal preparations containing antigens or antibodies A61K39/00; medicinal preparations characterised by the non-active ingredients, e.g. peptides as drug carriers, A61K47/00
• CPC Class Codes: C07K 1/047 Simultaneous synthesis of different peptide species; Peptide libraries
• CPC Class Codes: C07K 1/107 by chemical modification of precursor peptides
• CPC Class Codes: C07K 14/43522 from scorpions
• CPC Class Codes: C07K 14/805 Haemoglobins; Myoglobins
• CPC Class Codes: C07K 14/8114 Kunitz type inhibitors
• CPC Class Codes: C07K 14/8117 Bovine/basic pancreatic trypsin inhibitor (BPTI, aprotinin)
• CPC Class Codes: C12N 15/1037 Screening libraries presented on the surface of microorganisms, e.g. phage display, E. coli display
• CPC Class Codes: C12N 2795/14111 Inoviridae
• CPC Class Codes: C12N 2795/14122 New viral proteins or individual genes, new structural or functional aspects of known viral proteins or genes
• CPC Class Codes: C12N 2795/14143 viral genome or elements thereof as genetic vector
• CPC Class Codes: C12N 7/00 Viruses; Bacteriophages; Compositions thereof; Preparation or purification thereof preparing medicinal viral antigen or antibody compositions, e.g. virus vaccines, A61K39/00
• CPC Class Codes: C40B 40/02 Libraries contained in or displayed by microorganisms, e.g. bacteria or animal cells; Libraries contained in or displayed by vectors, e.g. plasmids; Libraries containing only microorganisms or vectors

Abstract

In order to obtain a novel binding protein against a chosen target, DNA molecules, each encoding a protein comprising one of a family of similar potential binding domains and a structural signal calling for the display of the protein on the outer surface of a chosen bacterial cell, bacterial spore or phage (genetic package) are introduced into a genetic package. The protein is expressed and the potential binding domain is displayed on the outer surface of the package. The cells or viruses bearing the binding domains which recognize the target molecule are isolated and amplified. The successful binding domains are then characterized. One or more of these successful binding domains is used as a model for the design of a new family of potential binding domains, and the process is repeated until a novel binding domain having a desired affinity for the target molecule is obtained. In one embodiment, the first family of potential binding domains is related to bovine pancreatic trypsin inhibitor, the genetic package is M13 phage, and the protein includes the outer surface transport signal of the M13 gene III protein.

Claims

1. A method of obtaining a nucleic acid encoding a binding protein having a proteinaceous binding domain that binds a predetermined target material, said target being a substance other than an antibody with an exposed antigen-combining site, comprising:

a) preparing a variegated population of amplifiable genetic packages, said genetic packages being selected from the group consisting of cells, spores and viruses, each said genetic package being genetically alterable and having an outer surface including a genetically determined outer surface protein, each package including a first nucleic acid construct coding for a chimeric potential binding protein, each said chimeric protein comprising and each said construct comprising DNA encoding (i) a potential binding domain which is a mutant of a predetermined domain of a predetermined parental protein other than a single chain antibody, and (ii) an outer surface transport signal for obtaining the display of the potential binding domain on the outer surface of the genetic package, the expression of which construct results in the display of said chimeric potential binding protein and its potential binding domain on the outer surface of said genetic package; and

wherein said variegated population of genetic packages collectively display a plurality of different potential binding domains, the differentiation among said plurality of different potential binding domains occurring through the at least partially random variation of one or more predetermined amino acid positions of said parental binding domain to randomly obtain at each said position an amino acid belonging to a predetermined set of two or more amino acids, the amino acids of said set occurring at said position in statistically predetermined expected proportions, said genetic packages being amplifiable in cell culture and separable on the basis of the potential binding domain displayed thereon,

b) causing the expression of said chimeric potential binding proteins and the display of said potential binding domains on the outer surface of said packages;

c) contacting said packages with the predetermined target material such that said potential binding domains and the target material may interact;

d) separating packages displaying a potential binding domain that binds the target material from packages that do not so bind, and

e) recovering at least one package displaying on its outer surface a chimeric binding protein comprising a successful binding domain (SBD) which bound said target, said package comprising nucleic acid encoding said successful binding domain, and amplifying said SBD-encoding nucleic acid in vivo or in vitro.

2. The method of claim 1 wherein the potential binding proteins are intergeneric chimeric proteins.

3. The method of claim 1 wherein said population is characterized by the display of at least 10⁵ different potential binding domains.

4. The method of claim 1 wherein the parental protein is a single domain protein.

5. The method of claim 1 wherein, for any potentially encoded potential binding domain, the probability that it will be displayed by at least one package in said population is at least 50%.

6. The method of claim 5 wherein, for any potentially encoded potential binding domain, the probability that it will be displayed by at least one package in said population is at least 90%.

7. The method of claim 1, wherein the method further comprises (i) isolating from the first nucleic acid construct of a package bearing a successful binding domain, a nucleic acid fragment consisting essentially of DNA encoding said successful binding domain, or (ii) determining enough of the DNA sequence of the first nucleic acid construct of (i) above to deduce the amino acid sequence of the successful binding domain and then preparing a second nucleic acid construct comprising DNA encoding the successful binding domain, said first and second nucleic acid constructs encoding different proteins.

8. The method of claim 7 wherein the second nucleic acid construct encodes a protein consisting essentially of said successful binding domain.

9. The method of claim 1 wherein, in said step (a), the differentiation among said potential binding domains of said variegated population is limited to no more than about 20 predetermined amino acid residues of said sequence and where the permissible variation at said alterable residue positions is also predetermined.

10. The method of claim 9 wherein the initially chosen parental binding domain has a melting point of at least 50°

C.

11. The method of claim 9 wherein the initially chosen parental potential binding domain is selected from the group consisting of (a) binding domain of bovine pancreatic trypsin inhibitor, crambin, Cucurbita maxima trypsin inhibitor III, heatstable enterotoxin of Excherichia coli, α

Conotoxin GI, μ

Conotroxin GIII, ω

Conotoxin GIV, apamin, charybdotoxin, secretory leukocyte protease inhibitor, cystatin, eglin, barley protease inhibitor, ovomucoid, T4 lysozyme, hen egg white lysozyme, ribonuclease, azurin, tumor necrosis factor, and CD4, and (b) domains at least substantially homologous with any of the foregoing domains which have a melting point of at least 50°

C.

12. The method of claim 9, further comprising determining which amino acid residues of said parental binding domain are likely to lie on the surface of said domain and limiting the variegation to codons encoding such amino acids.

13. The method of claim 9, said target material comprising one or more discrete molecules, said parental potential binding domain being characterized as a sequence of amino acids, further comprising identifying an interaction set of amino acids which are on the surface of the parental potential binding domain and which can all simultaneously touch a single molecule of the target material, and obtaining a variegated population wherein for each amino acid residue in said interaction set there is at least one potential binding domain wherein a different amino acid is substituted therefor, said interaction set comprising from about eight to about sixteen residues.

14. The method of claim 9 wherein the initially chosen parental binding domain is not a binding domain of a binding protein for the predetermined target.

15. The method of claim 9 wherein the initially chosen parental binding domain contains no more than 30 residues and at least 2 disulfides.

16. The method of claim 9 wherein the initially chosen parental binding domain contains no more than 60 residues and at least 3 disulfides.

17. The method of claim 9 wherein the initially chosen parental binding domain contains no more than 80 residues and at least 4 disulfides.

18. The method of claim 9, further comprising the steps of:

(f) determining the amino acid sequence of a successful binding domain and (g) preparing a new variegated population of replicable genetic packages according to said step (a), the parental binding domain for the potential binding domains of said new packages being a successful binding domain whose sequence was determined in step (f), and repeating steps (b)-(e) with said new population.

19. The method of claim 18 wherein a potential binding domain is considered successful only if it has at least a certain predetermined degree of affinity for target material, the required degree of affinity being increased for each new variegated population, and wherein each new variegated population dislays its parental binding domain in detectable amounts.

20. The method of claim 1 wherein the replicable genetic package is a bacterial cell and said DNA construct further comprises a periplasmic secretion signal sequence.

21. The method of claim 20 wherein the bacterial cell is selected from the group consisting of strains of Escherichia coli, Salmonella typhimurium, Pseudomonas aeruginosa, Klebsiella pneumonia, Neisseria gonorrhoeae, and Bacillus subtilis.

22. The method of claim 21 wherein the outer surface transport signal is derived from a bacterial outer surface protein selected from the group consisting of the lamB protein, OmpA, OmpC, OmpF, Phospholipase A, and pilin, or an assemblable segment thereof.

23. The method of claim 22 wherein the chimeric surface protein substantially corresponds to LamB with the foreign potential binding domain inserted at 153.

24. The method of claim 22 wherein the chimeric potential binding protein substantially corresponds to the first 153 amino acids of LamB fused in frame to the potential binding domain.

25. The method of claim 1 wherein the replicable genetic package is a bacterial spore.

26. The method of claim 25 wherein the bacterial spore is a Bacillus endospore.

27. The method of claim 26 wherein the replicable genetic package is an endospore of a strain of B. subtilis.

28. The method of claim 27 wherein the outer surface transport signal is the cotA, cotB, cotC or cotD protein or an assemblable segment thereof.

29. A method of producing a binding protein which binds a predetermined target material which comprises:

(i) obtaining, by the method of claim 1, a first nucleic acid construct encoding a chimeric binding protein having a binding domain which binds the predetermined target material, and

(ii) producing either said chimeric binding protein, or a second binding protein having essentially the same binding domain.

30. The method of claim 29 wherein the sequence of the binding domain is determined by sequencing at least a portion of either said first nucleic acid construct or said chimeric binding protein, and the determined sequence is then used to guide production of said second binding protein.

31. The method of claim 30 wherein said second binding protein is produced by expressing a second DNA construct derived at least in part from the first nucleic acid construct.

32. The method of claim 29 wherein the second binding protein is produced by recombinant DNA techniques.

33. The method of claim 29 wherein the second binding protein is produced by nonbiological peptide synthesis techniques.

34. The method of claim 29 wherein the second binding protein consists essentially of said successful binding domain.

35. The method o claim 1 wherein the genetic package is a single-stranded DNA bacteriophage other than bacteriophage lambda and the package is replicated in a bacterial host cell.

36. The method of claim 35 wherein said construct further comprises a cytoplasmic secretion signal sequence which codes for a signal peptide which directs the immediate expression product to the inner membrane of the bacterial host cell infected by said phage where it is processed to remove said signal peptide, yielding a mature chimeric protein comprising the potential binding domain and at least a portion of a coat protein of the phage, said chimeric protein being assembled with wild-type coat protein into the phage coat so that said phage displays the potential binding domain on the surface of its coat.

37. The method of claim 36 wherein the secretion signal sequence is derived from a first gene, and the nucleic acid sequence encoding the outer surface transport signal is derived from a second gene, the first and second genes being different.

38. The method of claim 36 wherein the secretion signal is encoded by a signal sequence selected from the group consisting of the signal sequences of the phoA, bla and geneIII genes.

39. The method of claim 35 wherein the replicable genetic package is a filamentous phage.

40. The method of claim 39 wherein the outer surface transport signal is provided by the major coat protein of a filamentous phage or a assemblable fragment thereof.

41. The method of claim 39 wherein the outer surface transport signal is provided by the gene III protein of a filamentous phage or an assemblable fragment thereof.

42. The method of claim 1 wherein the population of replicable genetic packages of step (a) is obtained by:

i) preparing a variegated population of DNA inserts, said inserts comprising a plurality of variegated codons, which collectively encode a plurality of different potential binding domains, and

ii) incorporating the resulting population of DNA inserts into the chosen replicable genetic packages to produce a variegated population of replicable genetic packages.

43. The method of claim 42 in which at least one variegated codon is a simply variegated codon selected from the group consisting of NNT, NNG, RNG, RMG, VNT, RRS, and SNT.

44. The method of claim 42 wherein none of the variegated codons is a simply variegated codon selected from the group consisting of NNN, NNK and NNS.

45. The method of claim 42 wherein for at least one variegated codon the ratio of amino acids encoded to possible trinucleotide sequences is at least 2:

3.

46. The method of claim 42 wherein for at least one variegated codon there are at least four equally most-favored amino acids.

47. The method of claim 42 wherein the ratio of amino acid sequences encoded to possible polynucleotide sequences is at least 1:

3.

48. The method of claim 42 in which at least one variegated codon is a complexly variegated codon.

49. The method of claim 48 in which the complexly variegated codon is prepared so as to yield a ratio of most favored amino acid to least favored amino acid which is less than 2.6.

50. The method of claim 48 wherein the distribution of nucleotides incorporated at said complexly variegated codon is further chosen to yield the largest value for the quantity.

51. The method of claim 50 wherein the distribution of nucleotides incorporated at said variegated codon is chosen to yield substantially equal abundances of acidic and basic amino acids.

52. The method of claim 50 wherein at least one variegated codon provides at least ten different amino acids at not less than 5% abundance.

53. The method of claim 1, said potential binding domain being a mini-protein sequence of less than about sixty amino acids and having at least one intrachain covalent crosslink between a first amino acid position and a second amino acid position thereof, the amino acids at said first and second positions being invariant in all of the chimeric proteins displayed by said population.

54. The method of claim 53 wherein the crosslink is a disulfide bond and the the amino acids at the first and second amino acid positions are cysteines.

55. The method of claim 54 in which the mini-protein domain has a single disulfide bond and the span of the bond is not more than nine amino acid residues.

56. The method of claim 54 wherein the mini-protein domain has a disulfide bond which bridges a sequence of amino acids which under affinity separation conditions collectively assume a hairpin supersecondary structure.

57. The method of claim 56 wherein the hairpin secondary structure is selected form the group consisting of (a) on α

helix, a turn, and a β

strand;

(b) an α

helix, a turn, and an α

helix, and (c) a β

strand, turn, and a β

strand.

58. The method of claim 54 wherein the mini-protein domain comprises a plurality of intrachain disulfide bonds.

59. The method of claim 58 wherein the mini-protein domain substantially corresponds in sequence to a mini-protein selected from the group consisting of Escherichia coli heat stable toxin I (ST_A), the bee venom apamin, or a squash-seed trypsin inhibitor, the scorpion toxin, charybdotoxin and secretory leukocyte protease inhibitor.

60. The method of claim 58 wherein the mini-protein domain has two disulfide bonds having a connectivity pattern of 1-3, 2-4.

61. The method of claim 60 wherein the mini-protein domain substantially corresponds in sequence to an α

-conotoxin.

62. The method of claim 58 wherein the mini-protein domain has three disulfide bonds having a connectivity patter of 1-4, 2-5, 3-6.

63. The method of claim 62 wherein the mini-protein domain substantially corresponds in sequence to a mu- or omegaconotoxin.

64. In a process for developing novel binding proteins, other than single chain antibodies, with a desired binding activity against a particular target material, other than the antigen-binding domains of antibodies, by mutagenesis of a gene encoding a known protein other than a single chain antibody, the improvement comprising displaying a proteinaceous potential binding domain on the outer surface of an amplifiable genetic package selected form the group consisting of cells, spores and viruses, each said genetic package being genetically alterable, said potential binding domain not being natively associated with the outer surface of aid package, said package containing the gene encoding said binding domain and means for directing said domain to the outer surface of said package, contacting the package with the target material, and determining whether the package displaying the potential binding domain binds to said target material.

65. A variegated population of replicable genetic packages, each package including a nucleic acid construct coding for a chimeric potential binding protein, each said construct comprising DNA encoding (i) a potential binding domain which is a mutant of a predetermined parental binding domain, and (ii) an outer surface transport signal for obtaining the display of the potential binding domain on the outer surface of the genetic package, wherein said initial binding domain is not a single chain antibody and is not identical to or substantially homologous with a binding domain natively associated with said transport signal, and wherein said variegated population of genetic packages collectively display a plurality of different potential binding domains, the differentiation among said plurality of different potential binding domains occurring through the at least partially random variation of one or more predetermined amino acid positions of said parental binding domain to randomly obtain at each said position an amino acid belonging to a predetermined set of two or more amino acids, the amino acids of said set occurring at said position in predetermined expected proportions.

66. A variegated population of DNA molecules encoding chimeric binding proteins, each said chimeric binding protein comprising (i) a binding domain, and (ii) at least a segment of an outer surface protein of a cell or virus, said segment acting to cause the display of the chimeric binding protein or a processed form thereof on the outer surface of the cell or virus, said binding domain being capable of binding to a target material to which said outer surface protein is not capable of preferentially binding, wherein said variegated population of DNA molecules encode chimeric binding proteins which collectively include a plurality of different binding domains, the differentiation among said plurality of different potential binding domains occurring through the at least partially random variation of one or more predetermined amino acid positions thereof to randomly obtain at each said position an amino acid belonging to a predetermined set of two or more amino acids, the amino acids of said set occurring at said position in predetermined expected proportions.