Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
HUMANIZED ANTIBODY AGAINST INTERLEUKIN-20 AND TREATMENT FOR INFLAMMATORY DISEASES
Document Type and Number:
WIPO Patent Application WO/2014/172591
Kind Code:
A9
Abstract:
A humanized antibody, or a scFv, Fab, or F(ab')2 thereof, includes: a heavy chain variable region, or a homologous variant thereof, wherein the heavy chain variable region includes: heavy chain framework regions, CDRH1 that has the sequence of SEQ ID NO:19, CDRH2 that has the sequence of SEQ ID NO:20, and CDRH3 that has the sequence of SEQ ID NO:21, wherein the heavy chain variable region and the homologous variant share at least 90% sequence identity in the heavy chain framework regions; and a light chain variable region, or a homologous variant thereof, that includes: light chain framework regions, CDRL1 that has the sequence of SEQ ID NO:22, CDRL2 that has the sequences of SEQ ID NO:23, and CDRL3 that has the sequences of SEQ ID NO:24, wherein the light chain variable region and the homologous variant share at least 90% sequence identity in the light chain framework regions.

Inventors:
WU CHIA-CHENG (CN)
HUANG CHAO-YANG (CN)
LIN YU-YING (CN)
CHEN YU-JUNG (CN)
LAI JIANN-SHIUN (CN)
Application Number:
PCT/US2014/034579
Publication Date:
May 28, 2015
Filing Date:
April 18, 2014
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
DEV CENTER BIOTECHNOLOGY
DCB USA LLC (US)
International Classes:
C07K16/46
Attorney, Agent or Firm:
LIANG, T. Chyau et al. (909 Fannin Street Suite 350, Houston TX, US)
Download PDF:
Claims:
CLAIMS

What is claimed is:

1. A humanized antibody, or a scFv, Fab, or F(ab')2 fragment thereof, comprising:

a heavy chain variable region, or a homologous variant thereof, comprising: heavy chain framework regions, CDRH1 that comprises the sequence of SEQ ID NO: 19, CDRH2 that comprises the sequence of SEQ ID NO:20, and CDRH3 that comprises the sequence of SEQ ID NO:21 , wherein the heavy chain variable region and the homologous variant thereof share at least 90% sequence identity in the heavy chain framework regions; and

a light chain variable region, or a homologous variant thereof, comprising: light chain framework regions, CDRL 1 that comprises the sequence of SEQ ID NO:22, CDRL2 that comprises the sequences of SEQ ID O:23, and CDRL3 that comprises the sequences of SEQ ID NO:24, wherein the light chain variable region and the homologous variant thereof share at least 90% sequence identity in the light chain framework regions.

2. The humanized antibody, or a scFv, Fab, or F(ab')2 fragment thereof, according to claim

1 , wherein the heavy chain framework regions comprise the sequences of corresponding framework regions in a human immunoglobulin heavy chain of subgroup III (VH3), and/or the light chain framework regions comprise the sequences of corresponding framework regions in a human immunoglobulin light chain of kappa subgroup II (VK2).

3. The humanized antibody, or a scFv, Fab, or F(ab')2 fragment thereof, according to claim

2, wherein the human immunoglobulin heavy chain subgroup III (VH3) variable region comprise the sequences of IGHV3-72*01 (SEQ ID NO: 2) or IGHV3-66*04 (SEQ ID NO: 15), and/or the human immunoglobulin light chain kappa subgroup II (VK2) variable region comprises the sequence of IGKV2D-29*02 (SEQ ID NO: 6) or IGKV1-39*01 (SEQ ID NO: 17).

4. The humanized antibody, or a scFv, Fab, or F(ab')2 fragment thereof, according to any one of claims 1-3, wherein the heavy chain variable region in the humanized antibody comprise the sequence of SEQ ID NO:9, and/or the light chain variable region in the humanized antibody comprise the sequence of SEQ ID NO: 10.

5. A composition comprising the humanized antibody, or a scFv, Fab, or F(ab')2 fragment thereof, according to any one of claims 1-4 for treating or preventing an IL-20 associated disease.

6. The composition of claim 5, wherein the IL-20 associated disease is one selected from the group consisting of an inflammatory, osteoporosis, cancer, stroke, and renal failure.

7. The composition of claim 6, wherein the inflammatory disease is rheumatoid arthritis,.

8. A method for treating or preventing an IL-20 associated disease, comprising

administering to a subject in need thereof an effective amount of the humanized antibody, or a scFv, Fab, or F(ab')2 fragment thereof, according to any one of claims 1-4.

9. The method of claim 8, wherein the IL-20 associated disease is one selected from the group consisting of an inflammatory disease, osteoporosis, cancer, stroke, and renal failure.

10. The method of claim 9, wherein the inflammatory disease is rheumatoid arthritis.

Description:
HUMANIZED ANTIBODY AGAINST INTERLEUKIN-20 AND

TREATMENT FOR INFLAMMATORY DISEASES

BACKGROUND OF THE INVENTION

Cross Reference to Related Applications

[0001] This application claims the priority of U.S. patent application No. 13/865,671, filed on April 18, 2013.

Field of the Invention

[0002] The field of the present invention is humanized antibodies against interleukin-20.

Background

[0003] Interleukin-20 (hereinafter "IL-20") belongs to the IL-10 family. IL-20 is a pleiotropic cytokine with chemoattractive, angiogenic, and osteoclastogenesis characteristics and plays an important role in various inflammation diseases and other disorders. Evidence suggesting IL-20 involvement in various pathological conditions include: up-regulation of both IL-20 and its receptors in atherosclerosis lesions, in synovial fluid of rheumatoid arthritis (RA) patients, and in human atherosclerotic artery. Furthermore, IL-20 is shown to promote atherosclerosis in apolipoprotein E-deficient mice, to induce neutrophil chemotaxis, to induce rheumatoid arthritis (RA) synovial fibroblasts (RASF) migration, and to promote angiogenesis. See e.g., U.S. Patent No. 7,786,274, issued to Chang.

[0004] Inhibition of IL-20 functions has been shown to prevent or reduce the developments of these diseases or disorders. For example, IL-20 receptor I (as a decoy receptor) has been shown to inhibit collagen-induced arthritis. Therefore, anti-IL-20 reagents (such as antibodies against IL-20) can be promising therapeutic options for the treatments and managements of IL-20-induced diseases.

[0005] U.S. Patent No. 7,786,274, issued to Chang, discloses a monoclonal antibody against IL-20 (mAb 7E). This monoclonal antibody (mAb 7E) is a mouse mAb, generated by immunizing BALB/cJ mice with recombinant hIL-20. This mAb 7E has been shown to be useful in treating various IL-20 associated inflammatory diseases, such as rheumatoid arthritis (RA), psoriasis, psoriatic arthritis, bacteria-induced gastric ulcer, and acute renal failure. [0006] For Example, U.S. Patent No. 7,611,705 shows that mAb 7E can be used to treat

IL-20 associated inflammatory diseases. U.S. Patent Nos. 7,837,994 and 8,454,956 show that mAb 7E can be used to treat osteoporosis. U.S. Patent No. 8,012,478 shows that mAb 7E can be used to treat ischemic stroke.

[0007] These prior art results show that anti-IL-20 antibodies (e.g., mAb 7E) can be useful therapeutic agents for the treatments and control of various IL-20 associated diseases. However, these antibodies are mouse antibodies, which may elicit immune responses when used as therapeutic agents. Such immune responses pose major limitations for the use of such antibodies in clinics. Recently, U.S. Patent No. 8,597,647, issued to Chang et aL, disclosed a humanized analog of mAb 7E, which was produced by grafting all six CDR sequences from mAb 7E into a human heavy chain and a light chain variable sequences. The resultant humanized antibodies have lower affinities for IL-20, as compared with the original mAb 7E.

[0008] All these prior art results indicate that anti-IL-20 antibodies can be useful.

However, there is still a need for better anti-IL-20 antibodies.

SUMMARY OF THE INVENTION

[0009] Embodiments of the present invention relate to anti-IL-20 antibodies or fragments thereof. The antibodies may be monoclonal antibodies, humanized antibodies, human antibodies, chimeric antibodies, or fragments thereof. Fragments of such antibodies may include Fab, scFv, F(ab')2, etc. Antibodies of the invention are generated based on human immunoglobulin heavy chain and light chain sequences, the CDR sequences of which are mutated to those of a known anti-IL-20 antibody from other animals (e.g., mouse) that bind tightly to IL-20 (e.g., mAb 7E disclosed in U.S. Patent No. 7,786,274). Such initial humanized antibodies may be further optimized to afford antibodies with substantially higher affinities for IL-20 and/or lower immunogenicities when used as therapeutics, as compared with the original mouse anti-IL-20 antibody.

[0010] One aspect of the invention relates to humanized antibodies, or a scFv, Fab, or

F(ab')2 fragments thereof. In accordance with one embodiment of the invention, a humanized antibody, or a scFv, Fab, or F(ab') 2 fragment thereof, include a heavy chain variable region, or a homologous variant thereof, comprising: heavy chain framework regions, CDRHl that has the sequence of SEQ ID NO:19, CDRH2 that has the sequence of SEQ ID NO:20, and CDRH3 that has the sequence of SEQ ID NO:21, wherein the heavy chain variable region and the homologous variant thereof share at least 90% sequence identity in the heavy chain framework regions; and a light chain variable region, or a homologous variant thereof, comprising: light chain framework regions, CDRL1 that has the sequence of SEQ ID NO:22, CD L2 that has the sequences of SEQ ID NO:23, and CDRL3 that has the sequences of SEQ ID NO: 24, wherein the light chain variable region and the homologous variant thereof share at least 90% sequence identity in the light chain framework regions.

[0011] In any of the embodiments described above, the heavy chain framework regions may comprise the sequences of corresponding framework regions in a human immunoglobulin heavy chain of subgroup III (VH3), and/or the light chain framework regions may comprise the sequences of corresponding framework regions in a human immunoglobulin light chain of kappa subgroup II (VK2).

[0012] In any of the above embodiments, the human immunoglobulin heavy chain subgroup III (VH3) variable region may comprise the sequences of IGHV3-72*01 (SEQ ID NO: 2) or IGHV3-66*04 (SEQ ID NO: 15); and/or the human immunoglobulin light chain kappa subgroup II (VK2) variable region may comprises the sequence of IGKV2D-29*02 (SEQ ID NO: 6) or IGKV1-39*01 (SEQ ID NO: 17).

[0013] In any of the above embodiments, the heavy chain variable region in the humanized antibody may comprise the sequence of SEQ ID NO:4, and/or the light chain variable region in the humanized antibody may comprise the sequence of SEQ ID NO:8. In accordance with one embodiment of the invention, an antibody is FLB5M5 that comprises a heavy chain variable region having the sequence of SEQ ID NO: 4 and a light chain variable region having the sequence of SEQ ID NO: 8. The FLB5M5 antibody may contain the constant regions from human immunoglobulin. For example, FLB5M5 may have a full heavy chain having the sequence shown in SEQ ID NO:9, and a full light chain having the sequence shown in SEQ ID NO: 10, and the corresponding coding sequences are shown in SEQ ID NO: 11 and SEQ ID NO: 12, respectively.

[0014] In any of the above embodiment of the present invention, the anti-IL-20 antibodies may include humanized monoclonal antibodies, such as FLB5M5 (also referred to as hFLB5M5). FLB5 5 is a humanized monoclonal antibody with three mutated amino acids m the complementarity determining regions (CDRs), relative to its parental mouse anti-IL-20 monoclonal antibody 7E (mAb 7E), and five mutated amino acids of the light-chain framework region, relative to the amino acids of the light-chain framework region of human V 2. FLB5M5 not only retains binding specificity toward IL-20 but also has an unexpected better binding affinity than mAb 7E for IL-20. FLB5M5 is also less immunogenic than mAb 7E to the human host in clinical application.

[0015] Another aspect of the invention relates to method for treating or preventing an IL-

20 associated disease. A method in accordance with one embodiment of the invention may comprise administering to a subject in need thereof an effective amount of the humanized antibody, or a scFv, Fab, or F(ab') 2 fragment thereof, according to any one of the above- described embodiments.

[0016] In accordance with any of the above described embodiments, the IL-20 associated disease may be one or more selected from the group consisting of an inflammatory disease, osteoporosis, cancer, stroke, and renal failure.

BRIEF DESCRIPTION OF TFTE DRAWINGS

[0017] FIG. 1 shows sequence comparison of the amino acids of mAb 7E (SEQ ID

NO:l), human homolog (IGHV3-72*01) (SEQ ID NO:2), HH12 (SEQ ID NO:3), and FLB5M5 (SEQ ID NO:4), in the heavy chain variable regions (VH regions).

[00 8] FIG. 2 shows sequence comparison of the amino acids of mAb 7E (SEQ ID

NO:5), human homolog (IGKV2D-29*02) (SEQ ID NO:6), HH12 (SEQ ID NO:7), and FLB5M5 (SEQ ID NO:8), in the light chain variable regions (V L regions).

[001 9] FIG. 3 shows sequence comparison of the amino acids of mAb 7E (SEQ ID

NO:l), human homolog (IGHV3-66*04) (SEQ ID NO:15), and DD10B5 (SEQ ID NO: 16), in the heavy chain variable regions (V H regions). DD10B5 is a humanized antibody having a heavy chain variable region based on IGHV3-66*04.

[0020] FIG. 4 shows sequence comparison of the amino acids of mAb 7E (SEQ ID

NO:5), human homolog (IGKV1-39*01) (SEQ ID NO: 17), and DD10B5 (SEQ ID NO:18), in the light chain variable regions (V L regions). DD10B5 is a humanized antibody having a heavy chain variable region based on IGKV1 -39*01. [0021 ] FIG. 5 shows summary of sequence comparisons between the amino acids of mAb 7E and the various human homologs shown in FIGs. 1-4.

[0022] FIG. 6 shows results of inhibition of IL-20-induced proliferation by various antibodies: mAb E7, FLB5, and FLB5M5.

[0023] FIG. 7A shows results from prevention of arthritis (measured with arthritis score

(AS)) by treatments with different doses (1 mg/kg, 3 mg/kg, or 9 mg/kg) of hFLB5M5 and compared with treatment using Enbrel (6 mg/kg) as a positive control, in a rat rheumatoid arthritis model.

[0024] FIG. 7B shows results from prevention of arthritis (measured with hind-paw thickness) by treatments with different doses (1 mg/kg, 3 mg/kg, or 9 mg/kg) of hFLB5M5 and compared with treatment using Enbrel (6 mg/kg) as a positive control, in a rat rheumatoid arthritis model.

[0025] FIG. 7C shows results from prevention of arthritis (measured with body weight) by treatments with different doses (1 mg/kg, 3 mg kg, or 9 mg/kg) of hFLB5M5 and compared with treatment using Enbrel (6 mg/kg) as a positive control, in a rat rheumatoid arthritis model.

[0026] FIG. 8 shows results of alanine substitutions for three important residues in

CDRH3 and CDRL3 of HH12 and FLB5M5. The results show that W100 in CDRH3 is important because alanine substitution at this position substantially reduce the binding affinities of both antibodies (FIH12 and FLB5M5). In CDRL3, F94 seems to be more important than Q90.

Definition

[0027] Unless otherwise defined, scientific and technical terms used herein shall have the meanings that are commonly understood by those of ordinary skill in the art. Furthermore, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular. A list of abbreviations used herein is as follows: Ab: antibody; BaF3: murine precursor B cells; CDRs: complementarity determining regions; C H : heavy-chain constant domain; CL: light-chain constant domain; HEPES: 4-(2-hydroxyethyl)-l- piperazineethanesulfonic acid; IL-20: interleukin-20; LTM: look-through mutagenesis; PBS: phosphate buffered saline; PCR: polymerase chain reaction; PTH: hind paw thickness; AS: arthritis score; RPMI medium: Roswell Park Memorial Institute medium; scFv: single-chain variable fragment; SD rats: Sprague-Dawley rats; V L : light-chain variable domain; V H : heavy- chain variable domain.

[0028] Standard techniques are used for recombinant DNA, oligonucleotide synthesis, and tissue culture and transformation (e.g., electroporation, lipofection). Enzymatic reactions and purification techniques are performed according to manufacturer's specifications or as commonly accomplished in the art or as described herein. These techniques and procedures are generally performed according to conventional methods well known in the art and as described in various general and more specific references that are cited and discussed throughout the present specification. See e.g., Sambrook et al., Molecular Cloning: A Laboratory Manual (3rd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (2001).

[0029] As used herein, the term "antibody" refers generally and broadly to immunoglobulins, monoclonal antibodies, and polyclonal antibodies, as well as active fragments thereof. The fragment may be active in that it binds to the cognate antigen (i.e., IL-20 or fragment of IL-20), or it may be active in that it is biologically functional. The antibodies of the invention may be chimeric, humanized, or human, using techniques known in the art.

[0030] As used herein, the term "humanized antibody" refers to an antibody in which minimal portions of a non-human antibody are introduced into an otherwise human antibody.

[0031 ] As used herein, the term "human antibody" refers to an antibody in which substantially every part of the protein is substantially non-immunogenic in humans, with only minor sequence changes or variations.

[0032] An antigen binding site is generally formed by the heavy chain variable region

(VH) and the light chain variable region (VL) immunoglobulin domains, with the antigen- binding interface formed by six surface polypeptide loops, termed complimentarity determining regions (CDRs). There are three CDRs each in VH (HCDR1, HCDR2, HCDR3) and VL (LCDR1, LCDR2, LCDR3), together with framework regions (FRs).

[0033] The term "CDR region" or "CDR" means the hypervariable regions of the heavy or light chains of the immunoglobulin as defined by Kabat et al., 1991 (Kabat, E. A. et al., (1991) Sequences of Proteins of Immunological Interest, 5th Edition. US Department of Health and Human Services, Public Service, NIH, Washington), and later editions. An antibody typically contains 3 heavy chain CDRs and 3 light chain CDRs. [0034] The term a "set of CDRs" referred to herein comprises CDR1, CDR2 and CDR3.

Thus, a set of HCDRs refers to HCDR1 , HCDR2 and HCDR3 (HCDR refers to a variable heavy chain CDR), and a set of LCDRs refers to LCDR1, LCDR2 and LCDR3 (LCDR refers to a variable light chain CDR). Unless otherwise stated, a "set of CDRs" includes HCDRs and LCDRs.

[0035] It has been shown that fragments of a whole antibody can also bind antigens.

Examples of binding fragments include: (i) an Fab fragment consisting of VL, VH, CL and CHI domains (Ward, E. S. et al., (1989) Nature 341, 544-546); (ii) an Fd fragment consisting of the VH and CHI domains (McCafferty et al., (1990) Nature, 348, 552-554); (iii) an Fv fragment consisting of the VL and VH domains of a single antibody (Holt et al., (2003) Trends in Biotechnology 21, 484-490); (iv) a dAb fragment (Ward, E. S. et al., Nature 341, 544-546 (1989), McCafferty et al., (1990) Nature, 348, 552-554, Holt et al, (2003) Trends in Biotechnology 21, 484-490], which consists of a VH or a VL domain; (v) isolated CDR regions; (vi) F(ab') 2 fragments, a bivalent fragment comprising two linked Fab fragments (vii) single chain Fv molecules (scFv), wherein a VH domain and a VL domain are linked by a peptide linker which allows the two domains to associate to form an antigen binding site (Bird et al., (1988) Science, 242, 423-426, Huston et al, (1988) PNAS USA, 85, 5879-5883); (viii) bispecific single chain Fv dimers (PCT US92/09965) and (ix) "diabodies", multivalent or multispecific fragments constructed by gene fusion (WO94/13804; Holliger, P. (1993) et al., Proc. Natl. Acad. Sci. USA 90 6444-6448).

[0036] Fv, scFv or diabody molecules may be stabilized by incorporation of disulfide bridges linking the VH and VL domains (Reiter, Y. et al., Nature Biotech, 14, 1239-1245, 1996). Minibodies comprising a scFv joined to a CH3 domain may also be made (Hu, S. et al., (1996) Cancer Res., 56, 3055-3061). Other examples of binding fragments are Fab', which differs from Fab fragments by the addition of a few residues at the carboxyl terminus of the heavy chain CHI domain, including one or more cysteines from the antibody hinge region, and Fab'-SH, which is a Fab' fragment in which the cysteine residue(s) of the constant domains bear a free thiol group.

[0037] Digestion of antibodies with the enzyme, papain, results in two identical antigen- binding fragments, known also as "Fab" fragments, and a "Fc" fragment, having no antigen- binding activity but having the ability to crystallize. Digestion of antibodies with the enzyme, pepsin, results in the a F(ab')2 fragment in which the two arms of the antibody molecule remain linked and comprise two-antigen binding sites. The F(ab') 2 fragment has the ability to crosslink antigen.

[0038] "Fv" when used herein refers to the minimum fragment of an antibody that retains both antigen-recognition and antigen-binding sites. "Fab" when used herein refers to a fragment of an antibody that comprises the constant domain of the light chain and the CHI domain of the heavy chain. The term "mAb" refers to monoclonal antibody.

DETAILED DESCRIPTION OF THE INVENTION

[0039] Embodiments of the invention relate to humanized anti-IL-20 antibodies or fragments thereof. The antibodies of the invention are monoclonal antibodies generated using molecular biology techniques. Fragments of these antibodies may include Fab, scFv, F(ab') 2 , chimeric antibodies, etc. As compared with mAb 7E disclosed in U.S. Patent No. 7,786,274, antibodies of the invention are exhibit higher affinities for IL-20 and/or lower immunogenicities when used as therapeutics.

[0040] Because a major limitation in the use of non-human antibodies in therapy is that these antibodies may elicit undesired immune responses, inventors of the present invention set out to design better anti-IL-20 antibodies by starting from human immunoglobulin sequences. Specifically, antibodies of the invention are produced starting with human framework sequences. These human antibody framework sequences are then modified to have IL-20 binding affinities. The modifications involve changing the sequences in the complementarity-determining regions (CDRs). Any methods known in the art may be used to produce the IL-20 binding sites, including random mutagenesis and screening, phage display and panning, or grafting of CDR sequences from known anti-IL-20 antibodies.

[0041] In accordance with embodiments of the invention, for example, selected immunoglobulin sequences may be endowed with IL-20 binding affinities by incorporating CDR sequences from known anti-IL-20 antibodies, such as mAb 7E. Using mAb 7E as an example, the following describes some specific examples for the production of antibodies having human framework sequences but containing mAb 7E CDR sequences. However, one skilled in the art would appreciated that similar antibodies may be produced using CDR sequences from other anti-IL-20 antibodies. The immediate products from the CDR grafting are chimeric antibodies, which contain human framework sequences and CDR sequences from another sources (e.g., mouse sequences).

[0042] Using the CDR grafting approach, it would be better to start with human antibody framework sequences that are homologous to the framework sequences of the antibody that will provide the CDR sequences. Such homologous sequences may be identified using consensus or homology searches with a known anti-IL-20 antibody framework sequence as a query sequence. A sequence with high homology is likely to provide the same framework structures to preserve the binding affinities of the target CDR sequences. Briefly, variable light (VL) and variable heavy (VH) domain framework sequences from human immunoglobulin subgroups having high homology with a target antibody (e.g., mAb 7E) may be identified as starting points.

[0043] In accordance with embodiments of the invention, the heavy chain framework sequences may be from VH subgroup III, and the light chain framework sequences may be from VL kappa subgroup II (VK2). For example, in accordance with one embodiment of the invention, the VH subgroup II sequence that is highly homologous to the framework regions of mAb 7E is found to be IGHV3-72*01, and the V 2 sequence that is highly homologous to the corresponding framework regions of mAb 7E is found to be IGKV2D-29*02. The sequence comparison between IGHV3-72*01 and the heavy chain variable region of mAb 7E is shown in FIG. 1, while the sequence comparison between IGKV2D-29*02 and the light chain variable region of mAb 7E is shown in FIG. 2. In FIG. 1 and FIG. 2, the sequence differences in the framework regions (i.e., non-CDR regions) are shown as boxed residues.

[0044] Based on the IGHV3-72*01 and the IGKV2D-29*02 sequences, the CDR sequences (i.e., CDR-1, CDR-2, and CDR-3) in each of which may be converted to the corresponding CDR sequences of the target anti-IL-20 antibody (e.g., mAb 7E). These conversions would generate a chimeric heavy chain and a chimeric light chain shown in FIG. 1 and FIG. 2, respectively. These chimeric heavy chain and light chain may be used to construct an scFv or Fab fragment, using techniques known in the art. Alternatively, the human framework sequences and the mouse CDR sequences may be assembled into an scFv or Fv fragment in a single step using overlap PCR, as described in the following example.

[0045] In yet another alternative approach, an scFv or Fab fragment may be constructed first, based on the identified human framework sequences for the VH and VL regions (e.g., IGHV3-72*01 and the IGKV2D-29*02 sequences). Then, such an scFv or Fab fragment may then be modified into a humanized anti-IL-20 antibody, by mutating the CDR sequences in the Fab fragment into known CDR sequences of a target anti-IL-20 antibody (e.g., mAb 7E) or to sequences substantially the same as the known CDR sequences of the target anti-IL-20 antibody.

[0046] As noted above, the mutations or CDR grafting may be accomplished with any methods known in the art, for example by site directed mutagenesis or by PCR incorporation of mutated residues/fragments (e.g., overlap PCR).

[0047] As an example, an scFv fragment based on IGHV3-72*01 as a heavy chain variable sequence and IGKV2D-29*02 as a light chain variable sequence and containing all six CDR sequences from mAb 7E is a humanized anti-IL-20 antibody, HH12, the heavy chain and light chain variable regions of which are shown in FIG. 1 and FIG. 2, respectively.

[0048] In accordance with embodiments of the invention, the initial chimeric antibodies

(such as HH12) may be further modified to improve their affinities for IL-20. These modifications may be referred to as affinity maturation, which may be accomplished with techniques known in the art, such as random mutagenesis and screening, phage display and panning, etc. These affinity maturations may involve residues in the CDR regions and/or non- CDR regions.

[0049] As an example, an scFv fragment, FLB5M5 (also referred to as hFLB5M5), was obtained by affinity maturation of FHT12. FLB5M5 contains three amino acid mutations in the complementarity determining regions (CDRs), as compared to the starting HH12 antibody or the parental mouse anti-IL-20 monoclonal antibody 7E (mAb 7E). These three amino acid mutations are indicated with circles in FIG. 2.

[0050] In addition, FLB5M5 contains five amino acid mutations in the light-chain framework region, as compared to the amino acids of the light-chain framework region of human VK2. The five amino acid mutations in the framework regions are illustrated as underlined residues in FIG. 2. The mutations in the non-CDR regions (i.e., the framework regions) may be by random mutagenesis. Alternatively, one may selectively replace the amino acid residues in the framework regions that differ from the corresponding amino acids in the "model" mouse antibody (e.g., mAb 7E) with those amino acids in the original "model" antibody. The rationale for putting the different amino acids back to what were in the original antibody (referred to as "back mutation") is that such amino acids in the framework regions may indirectly impact the binding of the antigen due to minor conformational changes in the paratope regions. The five amino acid mutations in the framework regions in FLB5M5 result from such "back mutations."

[0051] FLB5M5 represents an example of a product of affinity maturation by mutating amino acids in the light chain variable regions. One skilled in the art would appreciate that other affinity matured antibodies may be similarly obtained.

[0052] Even though FLB5M5 is an artificially generated antibody, it was surprisingly found that it not only exhibits the binding specificity for IL-20, but also has a better binding affinity than that of natural mAb 7E. In addition, because FLB5M5 contains mostly human immunoglobulin sequences, it is expected that this humanized monoclonal antibody would elicit less immune response in a human host. Thus, FLB5M5 and similar humanized antibodies are promising therapeutics for clinical applications.

[0053] In an attempt to understand the underlying mechanism for the enhanced affinity of

FLB5M5, a structure-activity relationship (SAR) study was performed. Such SAR studies may be conducted using any techniques known in the art, such as alanine scanning.

[0054] From such an SAR study, it was found that a tyrosine residue (Y32) located in the light-chain variable domain CDR1 plays an important role for the increased IL-20 affinity in FLB5M5. In addition, it was found that the combination of this tyrosine with other amino acid residues mutation enhance the binding affinity.

[0055] Some embodiments of the present invention relate to methods for the production of humanized antibodies of the invention. A method for the production of a humanized antibody may include the steps of: (1) obtaining a sequence of a mouse monoclonal antibody (e.g., mAb 7E) as a model, based on which a humanized antibody is to be designed; (2) selecting human framework donor antibody sequences based on homology with the "model" mouse monoclonal antibody; and (3) replacing the sequences of the CDR regions in the framework donor antibody sequences with the corresponding CDR sequences from the mouse monoclonal antibody. In accordance with some embodiments, the method may further include: affinity maturation steps, which may include mutations of the amino acids in the CDR regions and/or replacing amino acids in the framework regions in order to enhance the antibody affinities. The mutations in the framework regions may be by replacing amino acids with those from the "model" mouse monoclonal antibody.

[0056] For example, in preparing a humanized antibody, the amino acid sequence of a model mouse monoclonal antibody (e.g., mAb 7E) may be obtained from the database or determined using methods known in the art. The amino acid sequence of mAb 7E may then be used to search for human germ-line VL and VH sequences with the highest degree of homology with the mAb 7E framework regions. The homology search can be performed using any methods known in the art, such as BLAST (basic local alignment search tool) available at the National Center for Biotechnology Information (NCBI).

[0057] The search for human homologs of mAb 7E found IGHV3-72*01 to be the homolog for the V H of mAb 7E and IGKV2D-29*02 to be the homolog for V L of mAb 7E. The sequence alignments between mAb 7E and these two sequences are shown in FIG. 1 (V H regions) and FIG. 2 (V L regions). In FIG. 1 and FIG. 2, the amino acids in the framework regions that are different between mAb 7E and the human homolog sequences (IGHV3-72*01 and IGKV2D-29*02) are shown in boxes. As can be seen from the sequence alignments, the differences in the framework regions are relatively minor, suggesting that these homolog sequences could be good candidates for antibody humanization.

[0058] Therefore, these two human homolog sequences are used as base sequences for the production of a humanized antibody. In one example, the CDR regions of the human homologs (IGHV3-72*01 for the V H and IGKV2D-29*02 for V L ) were replaced with the corresponding CDR regions of mAb 7E by methods known in the art, such as by site-directed mutagenesis or by CDR grafting. CDR grafting, for example, may be accomplished by obtaining the oligonucleotide sequences of the cDR regions (e.g., by PCR or chemical synthesis) followed by ligation and/or cloning into suitable vectors to join with the remaining sequences of the framework regions. The CDR replacements produced a humanized anti-IL-20 antibody, HH12. As shown in FIG. 1 and FIG. 2, FIH12 contains all six CDR sequences from the mouse mAB 7E, while retaining all human framework sequences.

[0059] Once HH12 was prepared, it was further modified to optimize the binding affinity

(i.e., affinity maturation) for IL-20. In addition, further mutations may be performed to minimize its immunogenic effect in human hosts. First, the binding affinity of HH12 was optimized by mutations in the CDR regions. The mutations may be accomplished with random mutations at the selected sites (e.g., residues in the CDRLl, CDRL2, or CDRL3) followed by selection for the tight binders. Multiple methods of introducing mutations in the CDRs are known in the art, including radiation, chemical mutagens, and error-prone PCR. After such mutations, the mutants are screened for their binding affinities. Those with tighter bindings are selected and may be subjected to further runs of affinity maturations. This optimization process produced a new antibody that contains three mutations shown with circles in FIG. 2.

[0060] The antibody with optimized CDR sequences may be further enhanced by putting some of the framework amino acids in the human sequences back to the amino acids in the mouse sequences. Even though framework regions are not directly involved in antigen binding, they may indirectly impact the binding due to conformational effects. As shown in FIG. 2, five amino acids (underlined in FIG. 2) in the framework regions of FLB5M5 were mutated back to those found in the mouse mAB 7E sequences. These modifications can be performed in either order or simultaneously, i.e. modifying the framework first and then the CDRs or modifying the CDRs first and then the framework.

[0061] After these mutations/optimizations, a better antibody FLB5M5 was identified. In

FLB5M5, five amino acids from the HH12 light-chain framework region were replaced with amino acids from the mouse monoclonal antibody 7E (i.e., I2F, Q3V, Y36T, Q45K, and L46H), and three amino acids in light-chain CDR1 and CDR2 were mutated (i.e., S32Y, L50Q, and D55N).

[0062] FIG. 1 and FIG. 2 summarize the alignments of the V H and V L , respectively, amino acid sequences among mAb 7E, the human homologs (IGHV3-72*01 for the V H and IGKV2D-29*02 for V L ), HH12, and FLB5M5. In FIG. 2, amino acids that differ between FLB5M5 and HH12 in the CDR region are enclosed in rectangles and those in the framework regions are underlined.

[0063] Three mutations in the CDRs (circled residues in the CDR regions in FIG. 2) were found to be beneficial from the affinity maturation, with one mutation (from serine to tyrosine) in light chain CDR-1 found to be especially beneficial.

[0064] Table 1 shows IL-20 binding characteristics (k a : association rate; kj. dissociation rate; and K D : dissociation constant (i.e., inverse of affinity)) for several mutants in the CD regions arising from affinity maturations. These binding parameters may be obtained using any methods known in the art, such as ELISA or BIACore binding assays described below.

Table 1. Comparison of binding kinetics for various specimens with mutations in the CDRs

CDRH3 CDRL1 CDRL2 CDRL3 k a kd D

Specimen

96 98 27E 50 93 (M "1 s "1 ) is "1 ) (M)

HH12 S R S L H 1.34E+6 3.58E-3 2.68E-9

G3L S R Y L H 8.12E+5 1.07E-3 1.31E-9

FLB67 s Q Y Q H 8.88E+5 7.13E-4 8.03E-10

FLB35 D R Y L L 9.88E+5 4.31E-4 4.37E 0

FLB35(S) D R s L L 7.03E+5 2.83E-3 4.03E-9

[0065] From the data in Table 1, it can be concluded that mutations in CDRH3 did not produce much effects in the IL-20 binding affinity. On the other hand, mutation of serine-32 to tyrosine-32 in CDRL1 (i.e., 27E position in CDRL1), significantly improved the IL-20 binding affinities.

[0066] Back mutation identified five amino acids within the light-chain framework regions of HH12 that can increase the binding affinity to IL-20. In FIG. 2, these amino acids are underlined.

[0067] Finally, a combination of three mutations in the CDRs and five mutations within the light-chain framework region were all expressed as one humanized anti-IL-20 antibody, FLB5M5.

[0068] Anti-IL-20 antibodies (e.g., mAb 7E) have been shown to be effective in treating or preventing various IL-20 associated diseases, including osteoporosis (U.S. Patent Nos. 7,837,994 and 8,454,956), ischemic stroke (U.S. Patent No. 8,012,478), rheumatoid arthritis (U.S. Patent No. 7,786,274), etc. Therefore, antibodies of the invention, which are better (as compared to mAb 7E) anti-IL-20 antibodies, should also be effective in treating or preventing these IL-20 associated diseases or disorders.

[0069] Some embodiments of the invention relate to use of antibodies of the invention in treating or preventing IL-20 associated diseases or disorder, such as inflammatory diseases. In accordance with embodiments of the invention, a subject in need of treatment or prevention of an IL-20 associated disease is given a therapeutically effective amount of an antibody of the invention. The therapeutically effective amount is an amount sufficient to produce the desired clinical effects. One skilled in the art would appreciate that a pharmaceutically effective amount would depend on several factors, including patient's age, weight, condition, administration route, dosage form, and the antibody. However, finding an effective amount is routine practice that does not require inventive efforts, and one skilled in the art can find such effective amounts without undue experimentation.

[0070] Embodiments of the invention will be further illustrated with the following examples. These specific examples are for illustration only and are not intended to limit the scope of the invention. One skilled in the art would appreciate that various modifications and variations are possible without departing from the scope of the invention.

EXAMPLES

Selection of human V region framework sequences:

[0071 ] Human germ-line V L and V H sequences with the highest degree of homology with the mAb 7E framework regions were identified from the IMGT database (the International immuno genetics Information System®). The homology searches may be performed with BLAST or similar methods. The mAb 7E variable region sequences used as query sequences are available from the literature, such as U.S. Patent No. 7,786,274.

[0072] Human heavy chain framework sequences in the VH subgroup III (VH3) have been used in many humanized antibodies with success, and human light chain framework sequences of the VL subgroup II (VK2) are also shown to be good candidates. Therefore, the framework sequences of VH3 and V 2 subgroups were selected for the search for VH and V L frameworks, respectively. These searches identified IGHV3-72*01 and IGKV2D-29*02, respectively, as the VH and VL sequences most homologous to the corresponding heavy chain and light chain framework sequences in mAb 7E.

[0073] As shown in FIG. 1, the sequences of IGHV3-72*01 heavy chain framework regions differ from those in mAb 7E by 19 amino acids (the boxed residues), which corresponds to a 23.45% (19/ 81 total residues in the framework regions) variation. As shown in FIG. 2, the sequences of IGKV2D-29*02 light chain framework regions differ from those in mAb 7E by 10 amino acids (the boxed residues), which corresponds to a 13.16% (10/76 total residues in the framework regions) variation. [0074] Even with these degrees of variations in the framework regions, an scFv (HH12) generated by grafting CDR sequences from mAb 7E into the IGHV3 -72*01 and IGKV2D-29*02 sequences has a relatively good affinity for IL-20 (K D = 2.68> 10 "9 M) (for comparison, mAb 7E,

l ϋ

K D = 7.81x 10 " M) (see Table 2 below). These results suggest that the framework regions can tolerate a relatively high degree of variations without impacting the CDR region conformations.

[0075] These two sequences, IGHV3-72*01 and IGKV2D-29*02, will be used as examples for the construction of humanized antibodies against 11-20. However, one skilled in the art would appreciate that other similar sequences with high degrees of homologies may also be used. Examples of other sequences, for example, may include human IGHV3-66*04 (Fig. 3) and human IGKVl -39*01 (FIG. 4). The homologies of these examples are shown in the table in FIG. 5.

[0076] One skilled in the art would appreciate that these specific examples are for illustration only, and are not intended to limit the scope of the invention. Using homology search methods and human immunoglobulin libraries to search for homologous sequences for a model antibody (in which the framework regions will be replaced with human sequences) involves only routine techniques and one skilled in the art would be able to identify suitable homologous sequences for the desired purposes. Any similar homologous human sequences may be used. Preferably, these human sequences have high homologies in frame work regions of the model antibody (e.g., 7E). In addition, variant sequences (i.e., "homologous variants") that are substantially homologous with a naturally occurring human heavy chain or light chain variable sequence may also be used. For example, the homologous variants may have 90% or higher (e.g., 93%, 95%, 98%, or 99%) identities with a human heavy chain or light chain variable sequence.

HH12 scFv Construction and Display:

[0077] Based on the selected human heavy chain and light chain variable region homologs (e.g., IGHV3-72*01 and IGKV2D-29*02), an anti-IL-20 antibody or its fragments (e.g., scFv, Fab, etc.) may be constructed by grafting known CDR sequences from a known anti- IL-20 antibody (e.g., mAb 7E) into the homologous human heavy chain and light chain variable sequences. The following example uses CDR sequences from mAb 7E. However, one skilled in the art would appreciate that CDR sequences from any anti-IL-20 antibodies may be used.

[0078] In this particular example, an scFv fragment was generated. The scFv fragment, which is referred to as HH12, consists of complete human heavy-chain and light-chain framework sequences from IGHV3-72*01 and IGKV2D-29*02, respectively, and the six complete murine CDR sequences from mAb 7E. In this example, the HH12 scFv construct was assembled by overlap extension PCR. First, oligonucleotides having overlap sequences were prepared. Then, an equimolar mixture of the oligonucleotides (e.g., at a final concentration of 0.4 μΜ) was PCR-assembled using 0.5 μΐ of Pfx50 DNA polymerase and 5 μΐ of Pfx50 buffer (Invitrogen), according to the manufacturer's procedures.

[0079] After the assembly of the Hill 2 scFv construct using overlap extension PCR, a second PCR was performed using oligonucleotide primers to incorporate 5' Sfil and 3 r Notl restriction sites. This produced a full-length HH12 scFv construct with the desired restriction sites at the 5' and 3' ends for directional subcloning into a modified phage display vector pCANTAB5e (Amersham Pharmacia Biotech). This vector may be used for the production of HH12 scFv.

Back mutation:

[0080] For the expression of full-length antibodies in free style 293 cells, plasmid pTCAE8.3 was used for subcloning of the variable regions. This plasmid contains a DNA fragment encoding human κ CL region and human Y C h region. For mutations in the variable regions, individual oligonucleotides were synthesized to encode the desired mutations using PCR. These oligonucleotides include sufficient overlaps for PCR priming from the ΕΓΗ12 template. Next, the PCR products were gel-purified. Then, equimolar aliquots of the purified PCR products were combined for megaprime PCR to regenerate the full-length V L (or V H , if the mutation is in the heavy chain). The full-length V L fragment was then subcloned into pTCAE8.3 vector for the expression of full-length antibodies containing the desired mutations. Look-through mutagenesis (LTM) Library Construction:

[0001] Look-through mutagenesis (LTM) is a multidimensional mutagenesis that allows one to simultaneously assess and optimize combinatorial mutations of selected amino acids in a target peptide segment. LTM has been successfully used to optimize CDR sequences. The mutagenesis process typically focuses on one or more CDR domains and explores the contributions, including synergistic contributions, of amino acid side chains in the antigen binding. See, Rajpal et al., "A general method for greatly improving the affinity of antibodies by using combinatorial libraries " P.N.A.S., 102(24), pp. 8466-8471, 2005.

[0081] To prepare the LTM library, individual oligonucleotides were synthesized to encode each amino acid substitution for each CDR position. Each oligonucleotide contains sufficient overlaps for PCR priming from the HH12 template. PCRs containing LTM oligonucleotide mixtures corresponding to individual CDRs were used to amplify LTM- substituted CDR fragments. Next, these PCR fragments were gel-purified, and equimolar aliquots of the purified fragments were combined for megaprime PCR to regenerate full-length variable regions. These full-length variable regions were inserted into the pCANTABSe phagemid vector. The ligated DNA (the phagemid) was electroporated into E. coli TGI cells to generate a library stock.

Preparation of phage and selection of phage antibody libraries

[0082] For selection of phage (LTM) library with biopanning, the above-described library stock of E. coli was grown to the log phase. The phage particles displaying scFv were rescued by infection with M13K07 helper phage (NEB) and amplified overnight in 2YTA (2YT containing 100 pg/mL ampicillin and 25 g/mL kanamycin) at 30°C. The phage was precipitated with PEG NaCl (20% PEG 8000/2.5M NaCl), and then resuspended in PBS. The library was selected using biotinylated IL-20 and streptavidin-coated paramagnetic beads M280 (Dynal), as described below.

[0083] For selection of the LTM library, the biotinylated IL-20 at concentrations of

4.0x l0 "8 M, 1.0 x lO "9 M, 1.0 x lO "11 M, 1.0 x lO 42 M, and 1.0 xlO "13 M were used for selection rounds 1, 2, 3, 4, and 5, respectively. The mixture of the phages and the biotinylated IL-20 antigen was gently rotated for one hour at room temperature, and the phages bound to the biotinylatedIL-20 antigen were captured using 50~100 μΐ of streptavidin-coated M280 magnetic beads for five minutes. After capture of the phages, the beads were washed a total of ten times (4xPBST (PBS containing 0.05% Tween 20), 2*PBSM (containing 2% skimmed milk powder), 4* BS) using a Dynal magnetic particle concentrator. The third, fourth, and fifth washes were performed in competition with 1.4 μΜ IL-20 (non-biotinylated). Bound phages were eluted from the beads by sequential incubation with 1ml of lOOmM triethylamine (TEA) for 30 minutes. Eluents were combined and neutralized with 0.5 ml of 1 M Tris HC1 (pH 7.4) and half of the eluent was used to infect log phase E. coli TGI. The bindings, capture, wash, elution, and reinfection of E. coil were repeated for the desired number of cycles.

Expression and Affinity Measurements of HH12 Variants

[0084] The genes encoding the V H and V L chains of HH12 and its mutants were inserted into the pTCAE8.3 expression vector to produce scFv constructs. Free style 293 cells (InvitrogenO were transfected with the constaicts. The vector was transfected into the host cells by lipofectamine 2000 in accordance with the attached instruction manual (manufactured by Invitrogen). After the full-length scFv was purified from the pooled supernatants, competition ELISA and BIACore assays were used to detect and assess epitope specificities and binding affinities.

Combinatorial beneficial clone

[0085] From the above experiments, the mutant antibodies that can still bind to IL-20 would be identified, while mutants that lost binding affinities to IL-20 would likely not be identified. Analysis of the results reveals that three mutations in HH12 light chain CDR regions produced an antibody with an enhanced binding to IL-20. These three mutations include S32Y (in CDRL-1), L50Q (in CDRL-2), and D55N (in CDRL-2), which are shown as circled residues in FIG. 2.

[0086] These results also suggest that mutations at other residues in the CDR regions probably produced antibodies with relatively lower affinities such that they were not selected by phage panning, which tends to favor the tightest binder in the phage population. Therefore, it is likely that these other amino acids in the CDRs are relatively good fits for the binding interactions with IL-20 and no significant improvements can be realized by substitutions at these other locations.

[0087] In addition to the three "improved" substitutions in the CDR regions, the best antibody, FLB5M5, identified from the phage display-biopanning process also contains five beneficial mutations in the HH12 framework regions. These five amino acids are found to have been mutated from the residues in HH12 back to the corresponding residues in mAb 7E (i.e., back mutations). These five amino acids are: F2, V3, L36, K45, and H46, which are all in the light chain variable sequence, as shown in FIG. 2. The fact that back mutation produced better binders suggests that these residues in the framework regions contribute indirectly to the binding to IL-20 They probably contribute to the maintenance of proper conformations in the CDR regions.

[0088] The antibody, FLB5M5, identified from panning of the LTM library was found to bind tighter to IL-20 (as compared with mAb 7E), as shown in Table 2, and has a better therapeutic efficacy, as shown in FIG. 3. This finding (i.e., the humanized antibody is better than the natural antibody) is surprising because mAb 7E is a natural antibody produced by an immune system. Immune systems are known for their efficiency in producing "optimized" antibodies to bind with antigens. Therefore, humanization of an antibody is typically expected to produce a worse antibody in terms of binding with the target antigen.

[0089] The heavy chain variable region sequence (SEQ ID NO:4) and the light chain variable region sequence (SEQ ID NO: 8) are shown in FIG. 1 and FIG. 2, respectively. An example of the entire heavy chain sequence (including the constant regions from human immunoglobulin) and the entire light chain sequence (including the constant region) are shown as SEQ ID NO: 9 and SEQ ID NO: 10, respectively, while the corresponding polynucleotide sequences that encode the SEQ ID NO:9 and SEQ ID NO:10 are shown as SEQ IS NO:l l and SEQ ID NO: 12, respectively.

Important residues in the CDR regions by alanine scanning [0090] The above described phage panning of LTM library can identify optimized (i.e., tighter binders) antibodies. However, this phage panning approach gives only implication of the importance of certain amino acids in the CDR regions, as inferred from the lack of tighter binders when substitutions occurred at those locations. In order to positively assess the contributions of various amino acids side chains in the CDR regions to the binding of IL-20, alanine scanning was performed to replace individual amino acids in the CDR regions with alanine. If the original amino acid side chains are important for the binding, such alanine substitutions would reduce the binding affinities. Results from such an exemplary structure- activity relationship (SAR) study are shown in FIG. 5.

[0091] FIG. 8 shows the binding assay results of antibodies generated by alanine substitutions. Three residues in CDRH3 and CDRL3 of HH12 and FLB5M5 are found to be important. Specifically, W100 in CDRH3 is important because alanine substitution at this position substantially reduce the binding affinities of both antibodies (HH12 and FLB5M5). In CDRL3, F94 seems to be more important than Q90.

[0092] It was also found that alanine substitutions at other locations did not produce significant changes in the bindings to IL-20.

IL-20 binding activity

[0093] The human IL-20-binding kinetics of each purified anti-IL-20 antibody was estimated by surface plasmon resonance measurements using the BIAcore T100 biosensor system. The anti-IL-20 Ab was captured on an anti-human IgG immobilized CM5 sensor chip. The immobilized level of anti-human IgG was about 9,000-10,000 RU and the capture level of anti-IL-20 antibody was about 350-400 RU. Binding was carried out at constant flow rates of 30μί/ηιίη of IL-20 at various dilutions in HEPES buffered saline (BIA certified) for 60 seconds. Dissociations were carried out by passing through HEPES buffer for 480 seconds. Regeneration of the surface was carried out by infecting 10 mM Glycine pH 2.0/1.5 (50:50) for 40 seconds.

[0094] The IL-20 affinity of each of the anti-IL-20 antibodies were calculated from an affinity binding curve fit using the predefined model (1:1 binding) provided by Biacore T100 evaluation software 2.0. The binding affinity data for mAb 7E, HH12, and FLB5M5 are summarized in Table 2. As can be seen from the data in Table 2, even though HH12 has a lower affinity to IL-20 than mAb 7E, FLB5M5 has a higher affinity (i.e., lower K D ) than mAb 7E.

Table 2. Comparison of binding affinities to IL-20 of mAb 7E, HH12, and FLB5M5

Specimen k a (M l s "1 ) kd (s 1 ) D (M)

7E 9Λ ΙΕ+5 7.11E-4 7.81E-10

HH12 1.34E+6 3.58E-3 2.68E-9

FTB5M5 1.33E+6 3.82E-4 2.88E-10

[0095] The fact that an artificial antibody, FLB5M5, binds several folds tighter to IL-20 than the binding between mAb 7E and IL-20 is truly unexpected. mAb 7E is a natural antibody generated by a mouse immune system. By cooperative actions of various cytokines and helper cells, the immune system is known for its efficiency in producing optimized antibodies to bind with antigens. Therefore, humanization of a natural antibody often produces "less-optimal" antibodies, i.e., humanized antibodies typically have lower affinities than the original antibodies (see e.g., U.S. Patent No. 8,597,647).

IL20R2/IL22R BaF3 Proliferation assay

[0096] IL-20 is a pleitropic cytokine and has been shown to be involved in various inflammatory diseases and other disorders, and inhibition of IL-20 function has been shown to prevent or reduce various IL-20 associated diseases and disorder. Antibodies of the present invention bind IL-20 with very high affinities. Therefore, they should also be useful as therapeutics for the treatment and prevention of IL-20 associated diseases or disorders.

[0097] Some embodiments of the invention relate to methods for treating or preventing a disease associated with IL-20, which for example may include an inflammatory disease (e.g., rheumatoid arthritis), osteoporosis, cancer, stroke, and renal failure, estrogen deficiency (e.g., menopause), androgen deficiency (e.g., andropause), or cancer-induced osteolysis. A method in accordance with one embodiment of the invention may include administering to a subject in need of such treatment an effective amount of an antibody of the invention.

[0098] In this example, antibodies of the invention were tested for their abilities to inhibit or reduce IL-20 induced cell proliferation. Briefly, Ba/F3 cells stably transfected with full- length human IL-20 receptor complexes IL-22R and IL-20R2 to produce BaF-3(lL-20R2/IL22R) cells as the targets. Ba/F3 cells are murine precursor B cells of early lymphoblastoid cell lineage; they depend on IL-3 for viability and proliferation. These cells may be cultured in RPMI medium containing 10% fetal bovine serum and 1 ng/mL IL-3. Human recombinant IL- 20 can induce the proliferation of these cells. This IL-20 induced cell proliferation system can be used to assess the abilities of the antibodies of the invention to inhibit the IL-20 functions, due to their bindings to IL-20.

[0099] In the proliferation assay, BaF-3(IL-20R2/IL22R) cells were seeded in the wells of microtiter plates at 10 4 cells per well in the medium without IL-3 for 2 h at 37 °C in a 5% C0 2 incubator. Then, BaF-3(IL-20R2/IL22R) cells were cultured with pre-incubated 300 pM human cytokine IL20 and an increasing amount of test antibodies (three-fold dilutions from 1000 to 0.15 nM) for another 72h. In this example, three antibodies were tested: mAb 7E, FLB5, and FLB5M5. FLB5 is an antibody that contains the three amino acid changes in the CDR regions shown in FIG. 2, as compared with HH12. However, FLB5 contains the same framework regions as those in HH12. In other words, FLB5 does not have the five "back mutated" amino acids in the framework regions, as compared with FLB5M5.

[00100] After the incubation, AlamarBIue (Promega), a colorimetric/fluorigenic growth indicator, was added to the cultures, and the cells were incubated for another 6 hrs. The microtiter plates were then read on a fluorometer with 530 nm excitation and 580 nm emission. The fluorescent readings were analyzed using SigmaPlot Software to find the half maximal response (EC 50 ) for the anti-human IL20 antibodies.

[00101] FIG. 6 shows results from the proliferation assay. As shown in FIG. 6, antibody FLB5 is as good as or slightly better than mAb 7E in inliibiting IL-20-induced proliferation, while FLB5M5 is significantly better than mAb 7E. The fact that a modified antibody has a tighter binding than a naturally produced antibody is unexpected.

[00102] The EC 50 values of FLB5M5 as compared with that of niAB 7E are shown in Table 3. These data show that FLB5M5 is several times better than mAb 7E in inhibiting IL-20- induced cell proliferation. Again, the fact that this artificially created antibody is more effective than the natural antibody mAb 7E is unexpected.

Table 3. Comparison of neutralizing activity toward IL-20 of mAb 7E and FLB5M5

Specimen EC 50 (nM)

7E 134.63

FLB5M5 20.76

Collagen-induced arthritis rat model:

[00103] As noted above, IL-20 is involved in imflammatory diseases and antibodies against IL-20 have been shown to be effective in reducing the inflammatory diseases such as theumatoid arthritis. Antibodies of the invention should also be useful in treating arthritis. This was test using a collagen-induced arthritis model in rats.

[00104] In this animal model, six-week-old male Sprague-Dawley (SD) rats were immunized with type II collagen on day 0 and day 7 to induce arthritis. For example, the rats were immunized with an emulsion containing equal parts of Freund's complete adjuvant, 4 mg/ml of heat-killed mycobacterium tuberculosis and bovine type II collagen solubilized at 2 mg/ml in 0.05 M acetic acid. Each rat was injected intradermally in its dorsum with 200 μΐ of the emulsion on day 0. On day 7, the rat received a booster dose subcutaneously in the base of its tail with 100 μΐ of the same emulsion. Rats injected with buffers without the collagen (blank) were as a negative control (no arthritis).

[00105] After the onset of arthritis, which typically occurs on day 10-13, the treatments were started. Rats were administrated twice per week for a total of three injections - i.e., animals of all treatment groups received a single bolus subcutaneous injection in the back on days 10, 14, and 18. The treatments were conducted at doses of 1 mg/kg (mpg), 3 mg/kg, and 9 mg kg of antibody, hFLB5M5. Embrel (at 6 mg/kg) was used as a positive treatment control. The body weight (FIG. 4C), hind-paws thickness (PTH) (FIG. 4B and arthritic score (AS) (FIG. 4 A) were assessed to evaluate the therapeutic efficacies at three doses of the antibody. [00106] FIG. 7A shows results of the tests, as assessed with the arthritis scores (AS). As shown in FIG. 7A, administration of FLB5M5 produced substantial reductions in the AS, indicating that this antibody is effective in reducing the symptoms of arthritis. The FLB5M5 antibody is effective in reducing the AS even at 1 mg kg, while at 3 mg kg and 9 mg/kg, this antibody produced effects similar to the positive control (Embrel at 6 mg/kg). These results clearly show that FLB5M5 can be used to treat or prevent arthritis.

[00107] FIG. 7B shows results of hind-paw thickness assays. As shown in FIG. IB, administration of FLB5M5 produced substantial reductions in hind-paw thickness, indicating that this antibody is effective in reducing the symptoms of arthritis. The FLB5M5 antibody is effective in reducing hind-paw thickness even at 1 mg/kg, while at 3 mg/kg and 9 mg/kg, this antibody produced effects similar to the positive control (Embrel at 6 mg/kg). These results clearly show that FLB5M5 can be used to treat or prevent arthritis.

[00108] FIG. 7C shows results of body weight assays. As shown in FIG. 7C, administration of FLB5M5 produced substantial reductions in body weight, indicating that this antibody is effective in reducing the symptoms of arthritis. The FLB5M5 antibody is effective in reducing body weight even at 1 mg/kg, while at 3 mg/kg and 9 mg/kg, this antibody produced effects similar to the positive control (Embrel at 6 mg/kg). These results clearly show that FLB5M5 can be used to treat or prevent arthritis.

[00109] In addition to arthritis, antibodies against IL-20 are effective in the treatments of diseases associated with the IL-20 mediated signaling pathway including, but are not limited to osteoporosis, rheumatoid arthritis, cancer, stroke, or renal failure. In accordance with embodiments of the invention, the diseases may include osteoporosis, which can be caused by an inflammatory disease (e.g., rheumatoid arthritis), osteoporosis, cancer, stroke, and renal failure, estrogen deficiency (e.g., menopause), androgen deficiency (e.g., andropause), or cancer-induced osteolysis. That anti-IL-20 antibodies can be used to treat these diseases is known in the art, see e.g., U.S. Patent No. 8,597,647, issued to Chang et al.

[00110] While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims.