Epitope-directed antibody selection by photocrosslinking

Abstract

Provided is a method for screening antibodies against a specific antigen epitope, including: Incubating antigens with incorporation of photocrosslinking amino acids(designated as mutant antigen) with antibody library under light irradiation with suitable wavelength and energy, and selecting antibodies that covalently crosslink with the mutant antigen; then the antibodies selected are subjected to affinity maturation against wild-type antigen, and then epitope-directed antibodies obtained.

Claims

1. A method for selecting antibodies against a specific epitope of a target antigen, wherein the method comprises the following steps: (i) providing a support, wherein the support is immobilized with a mutant antigen formed by incorporation of amino acids with photocrosslinking activity or a derivative thereof in or near a target epitope of the target antigen; (ii) providing conditions that enables the contact of the mutant antigen with antibodies in antibody display library, and applying light irradiation with suitable wavelength and energy to allow the mutant antigen to covalently cross-link with displayed antibodies in the library that binds to or near the target epitope to form antigen-antibody complexes; (iii) performing elution under certain condition, wherein this condition enables the displayed antibodies that do not covalently cross-link with the mutant antigen to be washed away from the support, while the displayed antibodies that form the covalent cross-link remain on the support; (iv) releasing the displayed antibodies that covalently cross-link with the mutant antigen from the support; and optionally (v) further selecting the displayed antibodies capable of binding to the target antigen from the displayed antibodies obtained in the step (iv).

2. The method according to claim 1, wherein the amino acids with the photocrosslinking activity or its derivative thereof is incorporated by genetic codon expansion, or the amino acids with photocrosslinking activity are a natural amino acid or a noncanonical amino acid; or the noncanonical amino acid photocrosslinking activity is selected from: p-benzoyl-L-phenylalanine (pBpa) or p-azido-L-phenylalanine (pAzF); preferably, the light irradiation conditions suitable for pBpa cross-linking are: 365 nM, and 6 W.

3. The method according to claim 1, wherein the step (v) comprises repeated one or more rounds of the steps (i)-(iv).

4. The method according to claim 1, wherein the method further comprises sequencing the antibodies selected in the step (v).

5. The method according to claim 1, wherein the epitope comprises one or more amino acid residues; or the epitope is a linear epitope or a conformational epitope.

6. (canceled)

7. The method according to claim 1, wherein the antibody display library is selected from: IgG antibodies or antibody fragments such as Fab library, single chain Fv (scFv) library, or nanobody library; or, the antibody display library is selected from: fully human antibody libraries, humanized antibody libraries, mouse immune antibody library, alpaca immune nano-body library, and synthetic or semi-synthetic antibody library designed based on antibody sequences of different species; or, the antibody display library is a phage display antibody library.

8. (canceled)

9. The method according to claim 1, wherein a display carrier of the antibody display library is selected from: phages, bacteria, yeast, or mammalian cells.

10. (canceled)

11. The method according to claim 1, wherein the mutant antigen is a soluble protein, or a transmembrane protein expressed on a phospholipid membrane structure, wherein the mutant antigen can be directly immobilized on the support or indirectly immobilized on the support by membrane with phospholipid membrane structure.

12. (canceled)

13. (canceled)

14. (canceled)

15. The method according to claim 1, wherein the elution conditions of the step (iii) are selected from: a) competitive elution with a buffer containing the target antigen; b) acidic elution with a low-pH buffer; and c) alkaline elution with a high-pH buffer.

16. The method according to claim 1, wherein the releasing of the step (iv) is by enzymatic digestion.

17. The method according to claim 1, comprising using the antibody library obtained in the step (iv) or antibody library obtained in the step (iii) as the antibody panning pool.

18. The method according to claim 17, wherein the mutant antigen can be a soluble protein or a transmembrane protein expressed on a phospholipid membrane structure, which can be directly immobilized on the support or indirectly immobilized on the support by a membrane with the phospholipid membrane structure.

19. An antibody panning pool is obtained by the method according to claim 17.

20. A method for selecting an antibody in an antibody library, wherein the method comprises the following steps: (i) providing a support, wherein the support is immobilized with a mutant antigen with incorporation of amino acids with photocrosslinking activity or a derivative thereof in or near a target epitope of the target antigen; (ii) providing conditions that enable the contact of the mutant antigen with an antibody display library that allows the antigen to bind to the antibody, and applying light irradiation with suitable wavelength and energy to allow the mutant antigen to covalently cross-link to displayed antibodies in the library that binds to or near the target epitope to form antigen-antibody complexes; and (iii) performing elution under a certain condition, wherein this condition enables the displayed antibodies that do not covalently cross-link to the mutant antigen to be washed away from the support, while the displayed antibodies that form covalent cross-linking with the mutant antigen remains on the support, thereby selecting the antibodies that bind to the target epitope from those that do not bind to the target epitope in the library.

21. The method according to claim 20, wherein the mutant antigen is a soluble protein or a transmembrane protein expressed on a phospholipid membrane structure, which can be directly immobilized on the support or indirectly immobilized on the support by a membrane with the phospholipid membrane structure.

22. A method for selecting an antibody against a specific epitope of an antigen, wherein the method comprises: (i) providing conditions that allows contact of mutant antigen with incorporation of amino acids with photocrosslinking activity or a derivative thereof in or near a target epitope with a labeled antibody display library to allows the antigen binding to the antibody, and applying light irradiation with suitable wavelength and energy to allow the mutant antigen to covalently cross-link to displayed antibodies in the library that binds to or near the target epitope to form antigen-antibody complex; (ii) selecting the antigen-antibody complex that forms covalent cross-link with the mutant antigen, and releasing the displayed antibodies that form covalent cross-linking with the mutant antigen; and optionally (iii) further selecting the displayed antibodies capable of binding to the target antigen from the displayed antibodies obtained in the step (ii).

23. The method according to claim 22, wherein the mutant antigen contacts with the antibody display library in solution.

24. The method according to claim 23, wherein the mutant antigen is expressed on the cell surface.

25. The method according to claim 24, wherein cells expressing the mutant antigen are selected by flow cytometry.

26. The method according to claim 22, wherein the mutant antigen is a transmembrane protein.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings constituting the present application are used to provide a further understanding of the present disclosure. The exemplary embodiments and descriptions are used to explain the present disclosure, and do not constitute improper limitation of the present disclosure. In the drawings:

[0054] FIG. 1 shows the SDS-PAGE of wild-type (WT) IL1β and pBpa mutants.

[0055] FIG. 2 (FIG. 2A and FIG. 2B) shows the of enzyme-linked immunosorbent assay (ELISA) results of WT IL1β and pBpa mutants with Canakinumab (2A) or Gevokizumab (2B).

[0056] FIG. 3 shows the Western blot detection of WT IL1β and pBpa with Canakinumab or Gevokizumab, herein “−” represents no UV irradiation treatment; and “+” represents UV irradiation treatment.

[0057] FIG. 4 (FIG. 4A and FIG. 4B) shows the affinity detection of phage target with WT IL1β, 64pBpa(A) and 63pBpa(B), herein *p<0.05, **p<0.01, ***p<0.001, and .sup.nsp≥0.05.

[0058] FIG. 5 (FIG. 5A and FIG. 5B) is the ELISA result of a monoclonal phage and IL1β with alanine mutation at different sites, herein p value is the comparison between the alanine mutant group and the wild-type group, *p<0.05, **p<0.01,***p<0.001, and .sup.nsp≥0.05.

[0059] FIG. 6 (FIG. 6A and FIG. 6B) shows the affinity of phage and antigen, herein 6A is ELISA results of hC5a-35 phages with WT hC5a and 18pBpa; 6B is ELISA results of E02 phages with WT hC5a and alanine mutant, *p<0.05.

[0060] FIG. 7 (FIG. 7A and FIG. 7B) shows the affinity detection of hC5a-35-E02 phage with hC5a(7A) and mC5a(7B), *p<0.05, and ns means no statistical difference.

[0061] FIG. 8 (FIG. 8A and FIG. 8B) shows the affinity of hC5a-35 phage with WT hC5a, and hC5a mutants in which pBpa is incorporated at different positions.

[0062] FIG. 9 shows the Western Blotting detection of E02-scFv-Fc fusion protein binding to 18pBpa.

[0063] FIG. 10 shows the ELISA results of #8, #62, #125, #137 monoclonal phage libraries with BSA, WT GFP, and pBpa GFP.

[0064] FIG. 11 shows the expression of A2A and pBpa-A2A on the surface of Hela cells detected by flow cytometry.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] The present disclosure is described in detail herein by using the following definitions and reference to the examples. The contents of all patents and publications mentioned herein, including all sequences disclosed in these patents and publications are expressly incorporated herein by reference.

[0066] As used herein, the term “amino acid with photocrosslinking activity” refers to an amino acid that can covalently crosslink with an amino acid residue of adjacent proteins under suitable light irradiation conditions. The “amino acid with photocrosslinking activity” may include natural amino acids or ncAAs. The “photocrosslinking activity” includes, but is not limited to, sensitivity to ultraviolet (UV) light. Non-limiting examples of the “amino acid with photocrosslinking activity” include pBpa, pAzF and the like.

[0067] As used herein, the term “noncanonical amino acid” refers to an amino acid that is not one of 20 classic amino acids or selenocysteine or pyrrolysine. Other terms that can be used synonymously with the term “noncanonical amino acid” are a “non-naturally encoded amino acid”, an “unnatural amino acid”, and a “non-naturally occurring amino acid”. The term “noncanonical amino acid” also includes, but is not limited to, amino acids that are modified (e.g., post-translational modification) by naturally-encoded amino acids (including but not limited to 20 common amino acids or pyrrolysine and selenocysteine), but they themselves are not naturally incorporated into the growing polypeptide chain through the translation complex. The “noncanonical amino acid” may include a variety of functional groups or active groups, which can provide additional functions and/or activities.

[0068] As used herein, the term “photocrosslinking” means that under the suitable light irradiation condition, the groups of amino acids with photocrosslinking activity will covalently cross-link with groups of amino acid residues of adjacent proteins to form complex.

[0069] As used herein, “antibody display carrier” or “antibody library display carrier” is not limited to a particular vector. Although the present disclosure is exemplified with reference to phage display, the “antibody display library” of the present disclosure can also be identified by other display and enrichment techniques. Antibody fragments have been displayed on the surface of filamentous bacteriophage encoding antibody genes (Hoogenboom and Winter, J Mol Biol, 222:381-388 (1992); McCafferty et al., Nature 348(6301):552-554 (1990); Griffiths et al. EMBO J, 13(14):3245-3260(1994)). For a review of techniques for selecting and screening antibody display libraries, referring to, for example, Hoogenboom, Nature Biotechnol. 23(9): 1105-1116 (2005). In addition, it is known in the art to display heterologous proteins and its fragment on the surface of Escherichia coli (Agterberg et al., Gene 88:37-45 (1990); Charbit et al., Gene 70:181-189 (1988); Francisco et al., Proc. Natl. Acad. Sci. USA 89: 2713-2717 (1992)), and yeast such as Saccharomyces cerevisiae (Boder and Wittrup, Nat. Biotechnol. 15: 553-557 (1997); Kieke et al., Protein Eng. 10: 1303-1310 (1997)). Other known display techniques include ribosome or mRNA display (Mattheakis et al., Proc. Natl. Acad. Sci. USA 91:9022-9026 (1994); Hanes and Pluckthun, Proc. Natl. Acad. Sci. USA 94:4937-4942 (1997)), DNA display (Yonezawa et al., Nucl. Acid Res. 31(19):e118 (2003)); microbial cell display such as bacterial display (Georgiou et al., Nature Biotech. 15:29-34 (1997)), display on mammalian cells, spore display (Isticato et al, J. Bacteriol. 183: 6294-6301 (2001); Cheng et al, Appl. Environ. Microbiol. 71: 3337-3341 (2005) and U.S. 60/865,574 filed on Nov. 13, 2006), viral display such as retroviral display (Urban et al, Nucleic Acids Res. 33:e35 (2005), protein-DNA ligation based display (Odegrip et al, Proc. Acad. Natl. Sci. USA 101:2806-2810 (2004)); Reiersen et al, Nucleic Acids Res. 33:e10 (2005)), and microbead display (Sepp et al, FEBS Lett. 532:455-458 (2002)).

[0070] The “antibody phage display library” refers to a phage display library that displays antibodies or antibody fragments. The library can be monovalent, displaying one single-chain antibody or antibody fragment on average per phage particle, or multivalent, displaying an average of two or more antibodies or antibody fragments per viral particle. The term “antibody fragment” includes, but is not limited to, single-chain Fv (scFv) fragments and Fab fragments.

[0071] The “displayed antibodies” refers to antibodies or antibody fragments that are displayed on the surface of library carrier (for example, phages, bacteria, yeast, or mammalian cells) and can be exposed to the antigen used for screening. Thus, in the present disclosure, when referring to the “displayed antibodies”, it generally refers to the antibody together with the carrier on which it is displayed, unless otherwise understood by those skilled in the art according to the context. For example, when the phage display library is used, eluting the “displayed antibodies” from the support means eluting the phage displaying the corresponding antibody from the support.

[0072] The “display library” refers to the general term for the population of molecules presented on the display carrier. For antibody display libraries, the population includes, but is not limited to, antibodies, antibody fragments, nanobodies, scFv, Fab, VH, VL, protein/antibody fusion molecules, and the like.

Effect of Present Disclosure

[0073] Compared with conventional affinity-based antibody screening and separating methods, the method of the present disclosure can minimize the enrichment/screening bias due to the differences in the initial affinity of different antibodies in the library for the target antigen, and improve the diversity of the antibody pool during the selecting process.

Embodiment

[0074] It should be noted that, in the case no conflict, the embodiments in the present application are merely illustrative, and are not intended to limit the present disclosure in any manner.

Embodiment 1: IL1β Incorporated with pBpa on Epitope Capable of Crosslinking with Antibody

[0075] 1. Selection of Epitope

[0076] IL1β is an important cytokine that mediates inflammatory responses and various physiological activities. Canakinumab is a monoclonal antibody that blocks the interaction of human IL1β with IL1 receptor, and was approved by Food and Drug Administration (FDA) for clinical use. On the basis of the crystal structure of the Canakinumab-IL1β complex (Protein Data Bank (PDB) database ID number: 4G6J), the antigen epitope that Canakinumab binds includes Ser21, Glu25, Lys27, Glu64, Lys65, Asn66 and Asn129 (M. Blech et al, Journal of molecular biology 2013, 425:94). Among these, residues Glu64, Lys65, and Asn66 residues of IL1β are in a flexible loop and form extensive interactions with CDR3 of VH and CDR1 and CDR3 of VL. Another neutralizing antibody, Gevokizumab (PDB database ID number: 4G6M), not only binds to a distinct epitope, but also blocks the interaction between IL1β and IL1R, IL1 receptor accessory protein (IL1 RAcP) (M. Blech et al, Journal of molecular biology 2013, 94:425; D. Wang et al., Nature immunology 2010, 11(10):905-911). IL1β residues 63-66 are selected as the target epitope to develop and validate the new selection method. Residues Ala2, Leu7 of the antigen are greater that 15A away from loop 63-66 and does not form any direct interactions with Canakinumab or Gevokizumab, and are therefore selected as negative controls of the undesired binding site.

[0077] 2. Incorporation of pBpa into IL1β

[0078] 2.1 Construction of Wild-Type(WT) hIL1β and its Mutants

[0079] There are numerous previous reports that the Methanocaldococcus jannaschii (Mj) tyrosine tRNA synthetase (MjTyrRS) mutant and MjtRNA.sub.CUA pair w can be efficiently incorporated in the wild-type proteins through genetic codon expansion method (Wang L et al. Chem Biol 2001, 8(9): 883-90; Jiantao Guo et al. Angew Chem Int Ed Engl 2009, 48(48): 9148-9151; Wang L et al. Science 2001, 292(5516): 498-500; Chin J W. Nature 2017, 550(7674): 53-60]. pBpa may also function efficiently in this system and be incorporated into proteins (Hino N, et al. Nat Methods. 2005; 2(3): 201-6. Dorman G et al. Chem Rev 2016, 116(24): 15284-15398; Joiner M C et al, Protein Sci. 2019 June; 28(6): 1163-1170). In this example, recombinant plasmids expressing IL1β wild-type or mutant with the succinate codon TAG incorporation at 63, 64, 65, 66, 2 or 7 (designated as 63pBpa, 64pBpa, 65pBpa, 66pBpa, 2bpa or 7pBpa) is constructed.

[0080] In this example, all plasmids were generated by the Gibson assembly method (Daniel G. Gibson et al. Nucleic Acids Res 2009, 37(20):6984-6990). The open reading frames of IL1β fused with a 6×Histidine at the C terminus were amplified from pUC57-IL1β with primers IL1β-WT-F/R (see Table 1), and inserted into a linearized pET28a vector (Novagen, 69864-3) vector (digested with Nco I and Nhe I). Site-directed mutagenesis was performed to generate plasmids pET28a-IL1β-2TAG, pET28a-IL1β-7TAG, pET28a-IL1β-63TAG, pET28a-IL1β-64TAG, pET28a-IL1β-65TAG, and pET28a-IL1β-66TAG using according to the manufacturer's instructions (Yeasen Biotech, 10911) by using method of seamless and rapid cloning. pBpa was incorporated into IL-1β of these plasmids (J&K Scientific, 204322, CAS104504-45-2) through genetic codon expansion method. The same method was applied to the construction of pET28a-IL1β-2Ala, pET28a-IL1β-7Ala, pET28a-IL1β-63Ala, pET28a-IL1β-64Ala, pET28a-IL1β-65Ala, pET28a-IL1β-66Ala and pET28a-IL1β-63-66Ala, which was used to express alanine mutants of IL1β.

TABLE-US-00001 TABLE 1 Primers and gene sequences Primers Sequences pBpa mutant primers 2TAG-F TAAGAAGGAGATATACCATGTAGC CTGTGCGGAGCCT 7TAG-F: TAAGAAGGAGATATACCATGGCGC CTGTGCGGTAGCTGAACTGTACC 63TAG-F CATTGGGCCTTTAGGAAAAGAATC 63TAG-R GATTCTTTTCCTAAAGGCCCAATG 64TAG-F TGGGCCTTAAGTAGAAGAATCTGT 64TAG-R ACAGATTCTTCTACTTAAGGCCCA 65TAG-F GCCTTAAGGAATAGAATCTGTACC 65TAG-R GGTACAGATTCTATTCCTTAAGGC 66TAG-F TTAAGGAAAAGTAGCTGTACCTGA 66TAG-R TCAGGTACAGCTACTTTTCCTTAA Alanine mutant primers K63A-F CATTGGGCCTTGCGGAAAAGAATC K63A-R GATTCTTTTCCGCAAGGCCCAATG E64A-F TGGGCCTTAAGGCAAAGAATCTGT E64A-R ACAGATTCTTTGCCTTAAGGCCCA K65A-F GCCTTAAGGAAGCGAATCTGTACC K65A-R GGTACAGATTCGCTTCCTTAAGGC N66A-F TTAAGGAAAAGGCTCTGTACCTGA N66A-R TCAGGTACAGAGCCTTTTCCTTAA K63A-N66A-F CATTGGGCCTTGCGGCAGCGGCTC TGTACCTGAG K63A-N66A-R CTCAGGTACAGAGCCGCTGCCGCA AGGCCCAATG scFv phage construction primers Canakinumab-sc GGCCCAGGCGGCCGAGATTGTCCTT Fv-F ACCCAGAGTCC Canakinumab-sc GGCCGGCCTGGCCACTAGTAAGGGT Fv-R TGGGGCGGATGCACTCCCACTGCT GACGGTTACC Gevokizumab-sc GGCCCAGGCGGCCGACATACAGATG Fv-F ACCCAATCCAC Gevokizumab-sc GGCCGGCCTGGCCACTAGTAAGGGT Fv-R TGGGGCGGATGCACTCCCGGATG AGACCGTCACG 64UV63-scFv-Fc- AATTCGGCGGCCCAGGCGGCCGAG F CTCACACTCACGCAGTCT 64UV63-scFv-Fc- AGATGCCAGGCCGGCCTGGCCACT R AGTGAGGGTTGGGGCGGA pBpa mutant construction primers hC5a-18TAG-F ATATAAACATTCAGTATAGAAGAA ATGTTGTTACGATG hC5a-18TAG-R CATCGTAACAACATTTCTTCTATA CTGAATGTTTATAT Alanine mutant construction primers V18A-F ACGCTGCAAAAGAAGATAGAAGAA ATAGCTGCTAAATATAAACATTCA GTAGCGAAGAAATG V19A-F ACGCTGCAAAAGAAGATAGAAGAA ATAGCTGCTAAATATAAACATTCA GTAGTGGCGAAATG V18A-K19A-F ACGCTGCAAAAGAAGATAGAAGAA ATAGCTGCTAAATATAAACATTCA GTAGCGGCGAAATG

TABLE-US-00002 TABLE 2 Sequence table Nucleic acid sequence Amino acid sequence Gene (SEQ IDNO.) (SEQ ID NO.)) IL 1β WT with  1  2 6xHis tag Canakinumab HC  3  4 Canakinumab LC  5  6 Gevokizumab HC  7  8 Gevokizumab LC  9 10 Canakinumab scFv 11 12 Gevokizumab scFv 13 14 64UV63 scFv 15 16 hC5a with 6xHis tag 17 18 hC5a-35 scfv 19 20 E02-scfv 21 22 HCDR1 23 HCDR2 24 HCDR3 25 LCDR1 26 LCDR2 27 LCDR3 28 hC5a-35 VH 29 hC5a-35 VL 30 E02 VH 31

[0081] 2.2 Expression of WT hIL1β and its Mutants

[0082] Plasmid encoding the IL1β mutant and pEVOL-pBpa RS vector (Young T S et al. Mol Biol 2010, 395(2):361-74) (MjTyrRS-tRNA.sub.CUA pair containing Y32G, V103L, E107P, D158T and I159S mutations) were co-transformed into Escherichia coli BL21 (DE3). Only cells containing the double plasmids expressed the full-length protein in the presence of pBpa, which were purified by Ni-NTA column chromatography followed by size exclusion chromatography (SEC). The cells were cultured to OD600=0.6 in 2×YT medium, then, 1 mM pBpa, 0.5 mM isopropyl-β-d-thiogalactoside (IPTG) and 0.2% arabinose were added, and cultured overnight at 37° C. The yield of these mutant proteins was from 8 to 43 mg/L. Proteins were analyzed by SDS-PAGE (FIG. 1) and electrospray ionization mass spectrometry (ESI-MS) (data was not shown) to confirm the incorporation of pBpa. IL1β wild-type and mutant proteins migrated as a single band at approximately 19 kDa on SDS-PAGE gel, and exhibited the expected mass that was consistent with its amino acid sequence.

[0083] 3. Binding Ability of WT IL1β and its Mutants to Antibody

[0084] ELISA results show that compared with the binding of wild-type IL1β to Canakinumab, the affinity of the IL1β mutant to Canakinumab was reduced, and it was concentration-dependent (FIG. 2A). In contrast, the binding of Gevokizumab to wild-type IL1β and mutant IL1β was not significantly different (FIG. 2B), so Gevokizumab was used as a negative antibody control in the following experiments.

[0085] 4. Verification of Photocrosslinking Between IL1β with Incorporation of pBpa on Epitope and Antibody

[0086] IL1β WT and mutants (0.5 mg/ml) were incubated with Canakinumab (1 mg/ml), respectively, and exposed to long UV irradiation (6 W, and 365 nm) for 10 hours according to the method used in other protein cross-linking researches (Sato S et al. al, Biochemistry. 2011; 50(2):250-7. Results of Western Blot (FIG. 3) showed that mutant strains 63pBpa, 64pBpa and 66pBpa form covalently crosslinked products with light chain of Canakinumab, while 2pBpa and 7pBpa did not form the covalently crosslinked products. Mass spectrometry data also supported these results (not shown), confirming that pBpa in IL1β can cross-link with adjacent binding antibodies under UV irradiation, in contrast, no cross-linking of Gevokizumab (the binding of Gevokizumab to IL1β was away from loop 63-66 or loop 2-7 of IL1β, and the distance between the two proteins was greater than 13 Å) was observed under the same conditions.

[0087] The results above showed that: when pBpa was incorporated into the epitope and its vicinity, the spatial distance of photocrosslinking is exactly the same as the spatial distance of antigen-antibody binding. Under UV irradiation, the antibody could covalently cross-link with pBpa-incorporated antigen to form new complex, while non-epitope with too large spatial distance will not undergo photocrosslinking. The pBpa-incorporated IL1β on the epitope can undergo photocrosslinking reaction with antibodies, laying the foundation for screening epitope-directed antibodies by specific photocrosslinking.

Embodiment 2

[0088] Epitope-Directed Screening of Fully Human Antibody Phage Library

[0089] 1. Construction of a Fully Human Antibody Phage Library

[0090] According to published methods (Barbas C F et al, Proc Natl Acad Sci USA 1991, 88(18):7978-7982; Barbas III CF, Dennis R B, Gregg J S, In Phage Display: A Laboratory Manual. (CSH Press, 2001)), a multivalent single-chain antibody pill phage display library was previously constructed by using B cells from human peripheral blood mononuclear cells (PBMC), and the library had an antibody sequence diversity of about 10.sup.9 cfu.

[0091] 2. Screening of Fully Human Antibody Phage Library by Specific Photocrosslinking

[0092] The ELISA plate (Corning Costar, 2592) was coated with 100 μl 0.1 mg/ml mutant proteins (diluted with Dulbeccos Phosphate-Buffered Saline (DPBS)) and incubated overnight at 4° C. On the next day, the coating solution was removed, and 200 μl of blocking solution (3% skimmed milk, DPBST, 0.25% Tween 20) was used for blocking at 37° C. for 2 h. After removing the blocking solution, 10.sup.10 pfu of phage was added and incubated with the mutant proteins 63pBpa and 64pBpa for a period of time, and then UV irradiation (6 W, and 365 nm) was applied for 15 min-2 h. After washing with routine washing solution, three rounds of competitive washing (DPBS, 0.25% Tween 20, pH 7.4, and 0.1 mg/ml IL1β WT) was performed. After washing with routine washing solution, three rounds of low-pH washing (300 mM NaCl, 3% Tween 20, 100 mM glycine, and pH 2.0) was performed to remove non-covalently bound phage, followed by three rounds of PBS washing. After washing steps, the covalently cross-linked phage-antigen complex was released from the well by trypsin digestion. The collected phases from each well were incubated with E. coliXL1-Blue strain to infect cells. Colony Forming Unit (CFU) was counted, and finally the positive clones(hits) were picked. As expected, the output CFUs from panning of the both mutants were very low. Nonetheless, the output CFU is 3-4-fold higher compared to the group without UV irradiation (designated as non-UV-treated group), suggesting that a substantial portion of the output phage pool was covalently cross-linked with 63pBpa or 64pBpa (Table 3). In contrast, panning against WT IL1β using the same phage library and the same method exhibited a output UV/non-UV output ratio of 1.2 (close to 1), indicating that no significant cross-linking happened without incorporated pBpa. In addition, monoclonal phages displaying the scFv of Canakinumab and Gevokizumab were generated and selected following the same protocol, respectively. The UV/non-UV output ratio of the Canakinumab-scFV phages was 3.8. In contrast, the negative control antibody Gevokizumab-scFv phages exhibited a ratio of 1.1, indicating no significant number of the phage cross-linked with IL1β.

TABLE-US-00003 TABLE 3 Output CFU ratios of different phage libraries for WT IL1β and pBpa mutants with or without UV treatment UV- Non-UV- UV-treated treated treated output CFU/ group group Non-UV- Screened output output treated output Sample antigen CFU CFU CFU Fully human WT L1β  776  648 1.2 antibody phage 64pBpa  268  86 3.1 library 63pBpa  981  261 3.7 Canakinumab 63pBpa 6400 1696 3.8 scFv phage Gevokizumab 63pBpa 8650 7900 1.1 scFv phage

[0093] 3. Selection and Analysis of Clones

[0094] 55 colonies from the hit pool of 63pBpa or 64pBpa were randomly picked and their sequences were analyzed. Results showed that the sequences are diverse with low homology. In these, 15 (7 and 8 from the hit pool of 63pBpa and 64pBpa, respectively) distinct sequences were selected to generate monoclonal phages. The binding affinities of these phages to for WT IL1β and mutant 63pBpa and 64pBpa were analyzed by the ELISA. As shown in FIGS. 4A and 4B, more than half of phages selected from selection on 63pBpa or 64pBpa were cross-reactive with WT IL1β wild-type, although some of them showed reduced affinities. Two monoclonal phages (designated as 63UV7 and 64UV63 respectively) with significant affinities to both WT-IL1β and 63pBpa or 64pBpa, respectively, were picked and then incubated with 63pBpa or 64pBpa and processed with the panning procedure. After elution, the output CFUs of UV and non-UV-treated groups were counted and compared. The CFU of UV-treated group was 4-6-fold higher than that of the non-UV-treated group (Table 4), demonstrating that these scFv-displaying monoclonal phages bind to the desired epitope, and could cross-link with 64pBpa or 63pBpa upon UV irradiation. As a control, these phages did not show much difference on the CFUs between UV- and non-UV-treated groups against WY IL1 (Table 4). In addition, Lys63Ala, Glu64Ala, Lys65Ala, and Asn66Ala single mutants and Lys63Ala-Asn66Ala quadruple mutant of IL1β were also generated. Phages 64UV63 and 63UV7 showed significantly lower affinities to some of these alanine mutants compared to the WT, 63pBpa or 64pBpa (FIG. 5), indicating that they bind to the target epitope.

[0095] In conclusion, it is feasible and efficient to screen epitope-directed antibodies by the fully human antibody phage library in the specific photocrosslinking method.

TABLE-US-00004 TABLE 4 Output CFU ratios of targets selected from fully human antibody phage library against WT IL1β, 63pBpa and 64pBpa with or without UV treatment UV- Non-UV- UV-treated treated treated output CFUp/ group group Non-UV- Screened output output treated output Sample antigen CFU CFU CFU 63UV7 phage WTIL1β  184  197 0.9 63pBpa  201  46 4.4 64UV63 phage WT IL1β  128  117 1.1 64pBpa  140  22 6.4 Canakizumab 63pBpa 7900 2120 3.7 scFv phage Gevokizumab 63pBpa 4000 3040 1.3 scFv phage

Embodiment 3

[0096] Epitope-Directed Selecting of Mouse Immune Antibody Phage Library (Also Applicable to Humanized Antibody Transgenic Mice)

[0097] 1. Screening of Mouse Immune Antibody Phage Library by Specific Photocrosslinking

[0098] A large number of researches show that mouse immunization is a popular approach to generate antibodies with high affinity and selectivity against an antigen. The epitope-directed antibody selection method to the phage library produced from mouse immunization approaches. Mice were routinely immunized for three times with WT IL1β. Once the antibody titer in serum was confirmed, their spleens were collected and used to generate phage display libraries (Barbas C F et al, Proc Natl Acad Sci USA 1991, 88(18):7978-7982; Barbas III C F, Dennis R B, Gregg J S, In Phage Display: A Laboratory Manual. (CSH Press, 2001)). These libraries were then applied to epitope-directed panning using a similar method described in Embodiment 2, except that two rounds of panning were applied against 64pBpa to further enrich the hits. The output CFUs from the UV-treated group were about 9 times higher than those of the non-UV-treated group. In contrast, the UV/non-UV ratios of the hit pool and the selected hits were all around 1 when WT IL1β was used as the antigen (Table 5).

TABLE-US-00005 TABLE 5 Output CFU ratios of monoclonal phages targeting 64pBpa and derived from mouse immune phage libraries against 64pBpa, wild-type IL1β under UV or non-UV treatment UV-treated Non-UV- output CFU/ Coated UV-treated treated Non-UV-treated Sample antigen output CFU output CFU output CFU Fully human 64pBpa 3500  380 9.2 antibody phage WT IL1β  476  432 1.1 i64UV9 phage 64pBpa 7500 1400 5.4 WT IL1β 1400 1500 0.9 i64UV120 64pBpa 6600 1900 3.5 phage WT IL1β 2500 2300 1.1 i64UV5 phage 64pBpa  296  312 0.9 WT IL1β  453  402 1.1 i64UV40 phage 64pBpa 7900 7400 1.1 WT IL1β 8300 8000 1.0 i64UV104 64pBpa 1300  980 1.3 phage WT IL1β 1600 1200 1.3 Note: irepresents that the phage was derived from the immunized mouse phage library

[0099] 2. Selection and Analysis of Clones

[0100] 47 colonies from the hit pool were randomly picked and their sequences were analyzed. 7 sequence families were identified based on the homology. One representative clone was selected from each family, and generated monoclonal phages (except for one clone that yielded very low phage titer after production). Then, the output CFUs of these selected phages after one round of panning with or without the UV irradiation were examined. As shown in Table 2, 2 of the 6 selected monoclonal phages exhibited a UV/non-UV ratio larger than 3 against 64 pBpa, indicating their ability to cross-link with the target epitope. As a control, these phages did not show significant difference on the CFUs between UV and non-UV-treated groups against WT IL1β (Table 5).

[0101] In conclusion, it is feasible and efficient to select the epitope-directed antibodies by the mouse (also applicable to humanized antibody transgenic mice) immune antibody phage library through the specific photocrosslinking method.

Embodiment 4

[0102] Epitope-Directed Selecting of Antibodies Specific to Human Complement 5a (hC5a)

[0103] In order to show the general applicability of this method and prove the versatility of this method on other therapeutic targets, it is applied to antibody screening against hC5a antigen. Astherapeutic antibodies often require binding to a specific epitope in an antigen to exert their functions, this method could potentially facilitate antibody drug development (Daniel Ricklin et al. Nat Immunol 2010, 11(9):785-797; Cook W J et al. Acta Crystallographica 2010, 66: 190-197; Toth M J et al. Protein Sci 1994, 3(8): 1159-68). hC5a is a potential target for treatment of various diseases and syndromes such as anti-neutrophil cytoplasmic antibody-associated vasculitis (ANCA), atypical hemolytic uremic syndrome, systemic lupus erythematous, rheumatoid arthritis, and ischemia/reperfusion injury (Morgan B P et al. Nat Rev Drug Discov 2015, 14(12):857-77). In order to develop a therapeutic antibody against hC5a, it is desirable not only efficiently block the binding of hC5a to a hC5a receptor (C5aR), but also be highly selective to hC5a versus human C5 (hC5). By analyzing the crystal structures of hC5a (PDB: 3HQA) versus hC5 (PDB: 3CU7), an epitope (Ser16, Val17, Val18, Lys19 and Lys20), which is involved for the interaction with hC5a receptor (Huber-Lang M S, et al. J Immunol 2003, 170(12): 6115-24; Colley C S et al. MAbs 2018, 10(1): 104-117), but is buried inside the surface of hC5, was identified. Therefore, antibodies that bind to this epitope of hC5a are less likely to bind to hC5. Furthermore, the sequence of this epitope is highly conserved among human, monkey and rodent, which indicates that antibodies binding to this epitope are very likely cross-reactive among species.

[0104] In theme present disclosure, a Val18pBpa mutant of hC5a (designated as 18pBpa) was generated, characterized, and used for epitope-directed antibody selection. Panning was performed against the fully human phage displayed antibody library. After two rounds of screening against 18pBpa according to the selection procedure described above, the output ratio of UV/non-UV was greater than 13 (Table 6), suggesting that a significant portion of the output phage pool was covalently cross-linked with the antigen.

[0105] 25 colonies from the hit pool are selected and their sequences were analyzed. Sequences with correct scFv sequence assemblies are clustered on the basis of homology. Hit hC5a-35scfv (SEQ ID NO.) was selected from the cluster with the most abundant homologous sequences. Although it only showed modest affinities to hC5a and low affinity to 18pBpa (FIG. 6A), its UV/non-UV output CFU ratio was larger than 3 (Table 6). After affinity maturation on WT-hC5a using a secondary phage displayed antibody library generated by random mutagenesis based on the hC5a-35scfv sequence, a strong binder hC5a-35-E02 phage (E02) with the high affinity to hC5a was identified, and its UV/non-UV output ratio was significantly increased to 8.6 (Table 7). In order to verify whether this clone is a positive clone for the antigen epitope, it is verified by alanine scanning. Val18Ala and Lys19Ala single mutants and Val18Ala-Lys19Ala double mutants of hC5a were expressed and purified. Compared to WT-hC5a, E02 showed significantly lowered affinities for these alanine mutants, indicating that it binds to the target epitope (FIG. 6B). Furthermore, as expected, E02 showed much lower affinity to hC5 than hC5a (FIG. 7A), but similar affinity to mC5a (FIG. 7B). Next, E02 scFV-Fc fusion protein E02-scFv-Fc were also expressed and purified, and its binding affinities to hC5a, Val18Ala, Lys19Ala and Val18Ala-Lys19Ala, hC5 and mC5a were measured. The binding affinity profile corresponds to the results of E02 phages (FIGS. 8A and 8B). Western blot results also showed that 18pBpa formed a covalently linker product with E02-scFv-Fc fusion protein, while WT-hC5a did not (FIG. 9).

[0106] In conclusion, the selection method for epitope-directed antibodies using specific photocrosslinking method has broad application value and universality.

TABLE-US-00006 TABLE 6 Output CFU ratios of fully human antibody phage antibody library against WT hC5a, 18pBpa mutant with or without UV treatment UV-treated output CFU/ Non-UV- Non-UV- Screened UV-treated treated treated output Sample antigen output CFU output CFU CFU Fully human 18pBpa 370 28 13.2 library phage antibody WT hC5a  42 34  1.2

TABLE-US-00007 TABLE 7 Output CFU ratio of monoclonal phage against wild WT 1hC5a, 8pBpa mutant under UV or non-UV treatment UV-treated UV- Non-UV- output CFU/ treated treated Non-UV- Coated output output treated Sample antigen CFU CFU output CFU hC5a-35 18pBpa 296  89 3.3 phage WT hC5a 116 103 1.1 E02 phage 18pBpa 670  78 8.6 WT hC5a 101  97 1.0

Embodiment 5: Epitope-Directed Selecting of Antibodies Against Membrane Surface Protein

[0107] In order to prove the applicability of this method for screening antibodies against transmembrane proteins, we first applied to screen antibodies against antigens expressed on cell membranes. pCDNA3.1-WT GFP-GPI and pCDNA3.1-(TAC151TAG)GFP-GPI eukaryotic expression plasmids were constructed and transfected into 293T cells, respectively. After 48 hours, the results of flow cytometry showed that GFP-GPI was expressed on the membrane (data not provided). Then, the pcDNA3.1-(TAC151TAG)GFP eukaryotic expression plasmid was transfected into Hela cells, and 48 hours after transfection, the Hela cells were blocked with 5% milk for 1 hour, and incubated with a human natural antibody phage library (the titer is 10.sup.12/mL) on ice for 30 min. 365 nm UV irradiation was applied for 30 min, elution with acidic eluent for 30 min was applied to remove non-covalently cross-linked phages The covalently cross-linker phages were released by 200 ug/mL trypsin at 37° C. for 30 min and subjected to infect XL-blue for 1 h, followed by counting of the number of single clones. The output ratio of UV/non-UV was 5.8 (see Table 8), indicating that the cross-linked phages generated by UV irradiation were enriched. Single colonies in the UV-treated group were amplified with pSEX-F and pSEX-R as upstream and downstream primers respectively, and colonies PCR products larger than 750 bp were subjected to phage packaging (low titer). At the same time, soluble WT GFP and pBpa-GFP were also expressed and purified in 293F cells. After coated in 96-well ELISA plate, and blocked with 5% milk for 1 h at a room temperature, then phage above was added and incubated at room temperature (anti-M13-HRP as a secondary antibody). After tetramethylbenzidine (TMB) color development, OD650 was read detected by a microplate reader. The single clone was sequenced. The sequencing results of #8, #62, #125, and #137 showed the sequence characteristics of scFv. The 4 single clones were separately packaged with phage and precipitated with 5×PEG to increase the titer. the phage ELISA was verified with 10 uL phage (BSA, WT GFP, and pBpa GFP were used as the antigens, respectively), and the results were shown in FIG. 10.

[0108] The Hela cells transfected with pcDNA3.1-(TAC151TAG) GFP were blocked with 5% milk for 1 hour, incubated with #125 monoclonal phage library on ice for 30 min. 365 nm UV irradiation was applied for 30 min, followed by elution by acidic eluent for 30 min. After eluted with 0.1% sodium dodecyl sulfate (SDS) for 10 min and 20 min, respectively, 200 ug/mL trypsin were applied at 37° C. for 30 min, and subjected to infected XL-Blue for 1 h followed by counting of single clones. The UV/non-UV ratio was shown in Table 9, indicating that the #125 phage displaying scFv was able to cross-link with the GFP epitope expressed on the cell membrane surface.

TABLE-US-00008 TABLE 8 Output CFU ratio of natural human antibody phage library against pBpa-GFP with or without UV treatment UV-treated UV- Non-UV- output CFU/ treated treated Non-UV- Coated output output treated output Sample antigen CFU CFU CFU Natural human pBpa 146 25 5.8 antibody phage library

TABLE-US-00009 TABLE 9 Output CFU ratio of #125 monoclonal phage library against pBpa-GFP under different elution times of 0.1 % SDS with or without UV treatment SDS UV-treated Non-UV- UV-treated output elution output treated output CFU/non-UV-treated time CFU CFU output CFU 10 min 194 5 38.8 20 min 153 0 ∞

[0109] Next, this screening method was further applied to screen antibodies against multi-transmembrane protein A2A. pCDNA3.4-WT A2A and pCDNA3.4-(TTT168TAG)A2A eukaryotic expression plasmids (with a flag-tag at a N-terminal and a His-tag at a C-terminal) were constructed, and transfected into Hela cells, respectively. Flow cytometry after 48 h of transfection showed the expression of both WT A2A and mutant A2A on the membrane (see FIG. 11). After that, the Hela cells expressing the mutant A2A (pBpa-A2A) were used for epitope-directed antibody screening: Hela cells expressing the mutant A2A (pBpa-A2A) were blocked with 5% milk for 1 hour and incubated with a human natural antibody phage library (the titer is 10.sup.12/mL) on ice for 30 min, followed by 365 nm UV irradiation for 30 min; After eluting with an acid eluent for 30 min, followed by eluted with 0.1% SDS for 5 min, 10 min, 20 min, 25 min, and 30 min, respectively. Covalently bound phages were released by 200 ug/mL trypsin at 37° C. for 30 min and subjected to infected with XL-Blue for 1 h, followed by the number of clones counting. The UV/non-UV ratio (see Table 10) was greater than 3 when eluted with 0.1% SDS for more than 20 min, indicating that these scFv-displaying phages were capable of cross-linking with target epitope. Single clones in UV-treated group eluted with 0.1% SDS for 25 min and 30 min, respectively were selected and amplified with primers pSEX-F and pSEX-R. PCR products larger than 750 bp were picked. 20 single clones were randomly selected for sequencing. Sequence alignment showed that monoclonal phages all exhibited typical scFv sequence features with diversity in CDR region. Phylogenetic tree analysis also showed the amino acid homology of these clones.

TABLE-US-00010 TABLE 10 Output CUF ratios of natural human antibody phage library against pBpa-A2A under different elution times of 0.1% SDS with or without UV treatment Non-UV- UV-treated SDS UV-treated treated output CFU/ elution output output Non-UV-treated time CFU CFU output CFU  5 min 198 192 1.03 10 min  97  60 1.62 20 min  34  10 3.4 25 min  43  9 4.78 30 min  34  10 3.4