CXCL13 antibody for the treatment of autoimmune disorders

Abstract

Background: Homeostatic B Cell-Attracting chemokine 1 (BCA-1) otherwise known as CXCL13 is constitutively expressed in secondary lymphoid organs by follicular dendritic cells (FDC) and macrophages. It is the only known ligand for the CXCR5 receptor, which is expressed on mature B cells, follicular helper T cells (Tfh), Th17 cells and regulatory T (Treg) cells. Aberrant expression of CXCL13 within ectopic germinal centers has been linked to the development of autoimmune disorders (e.g. Rheumatoid Arthritis, Multiple Sclerosis, Systemic Lupus Erythematosis). We, therefore, hypothesized that antibody-mediated disruption of the CXCL13 signaling pathway would interfere with the formation of ectopic lymphoid follicles in the target organs and inhibit autoimmune disease progression. This work describes pre-clinical development of human anti-CXCL13 antibody MAb 5261 and includes therapeutic efficacy data of its mouse counterpart in murine models of autoimmunity.

Results: We developed a human IgG1 monoclonal antibody, MAb 5261 that specifically binds to human, rodent and primate CXCL13 with an affinity of approximately 5 nM and is capable of neutralizing the activity of CXCL13 from these various species in in vitro functional assays. For in vivo studies we have engineered a chimeric antibody to contain the same human heavy and light chain variable genes along with mouse constant regions. Treatment with this antibody led to a reduction in the number of germinal centers in mice immunized with 4-Hydroxy-3-nitrophenylacetyl hapten conjugated to Keyhole Limpet Hemocyanin (NP-KLH) and, in adoptive transfer studies, interfered with the trafficking of B cells to the B cell areas of mouse spleen. Furthermore, this mouse anti-CXCL13 antibody demonstrated efficacy in a mouse model of Rheumatoid arthritis (Collagen-Induced Arthritis (CIA)) and Th17-mediated murine model of Multiple Sclerosis (passively-induced Experimental Autoimmune Encephalomyelitis (EAE)).

Conclusions: We developed a novel therapeutic antibody targeting CXCL13-mediated signaling pathway for the treatment of autoimmune disorders.

Figure 1

Generation of MAb 5261 and its murine counterpart.

Figure 2

Effect on CXCL13- (A-D) and CXCL12- (negative control; E) induced migration of human and murine cells. Human pre-B-697-hCXCR5 cells (A, E), human pre-B-697-hCXCR4 cells (E), human tonsillar cells (B, E), C57BL/6 splenocytes (C, E), and SJL mouse splenocytes (D, E) were seeded into Transwell inserts. Diluted chemokines +/− antibodies were added into the lower chambers and plates were incubated for 2 h at 37°C (50 ug/ml of each antibody were used in CXCL12-induced migration assay). Inserts were removed and Alamar blue was added to the wells and incubated at 37°C overnight. Fluorescence was measured at wavelengths of 530 nm and 590 nm and the percent inhibition of chemokine-induced migration (A-D) or migration index (E) was calculated. Data represents an average of the measurements from at least three independent experiments + SEM. Statistical significance (relative to the isotype controls) was evaluated using one-way ANOVA followed by Bonferroni’s multiple comparison post test **** P < 0.0001; *** P < 0.0003; **P < 0.01.

Figure 3

Inhibition of human CXCL13-mediated internalization of human CXCR5 receptor. Human pre-B-697-huCXCR5 cells were pre-blocked with anti-human Fc for 15 min at 37°C, incubated with a pre-complexed huCXCL13/antibody mix for 2 h at 37°C, and subsequently stained with anti-human CXCR5 antibody for flow cytometric analysis. (A) Cell surface CXCR5 receptor expression. Geometric mean values were determined by FlowJo 7.6 software. (B) EC50 values were calculated from sigmoidal dose response curves (shown on the graph) with R² values equal to 0.9 (MAb 5261) and 0.98 (MAb 5261-muIg). Data points represent an average of measurements obtained from 3 independent experiments.

Figure 4

MAb 5261-muIg-mediated inhibition of B cell migration into the spleens of naïve BALB/c mice. Recipient mice were treated with 30 mg/kg of an Isotype Control Ab or MAb 5261-muIg intraperitoneally twice a week for 2 weeks before adoptive transfer of CFSE-labeled donor B cells (2x10⁷ cells per mouse; i.v.). (A) Percentage of CFSE+ B cells in mouse spleens (n = 3/group) as measured by flow cytometry. Flow data were analyzed using FlowJo 7.6 software. Statistical significance was evaluated using unpaired Student’s t-test (B) IHC on representative spleen sections where CFSE-labeled adoptively transferred cells are green and B cells are labeled red.

Figure 5

Effect of murine anti-CXCL13 antibody on germinal center formation in C57BL/6 mice challenged with NP-KLH. Mice were treated with 30 mg/kg of MAb 5261-muIg antibody and corresponding mouse isotype control, on days −3, 0 (the day of the immunization), 4 and 7 (n = 9/group). (A) Total IgM and IgG1 anti-NP antibody-secreting cells were detected by ELISPOT on wells coated with NP14-BSA. (B) The frequencies of high affinity IgG1 anti-NP-secreting cells were determined by ELISPOT on wells coated with NP4-BSA. (C) Number of germinal centers (n = 9 mice/group). Follicles were hand counted. Number of PNA+ germinal centers in relation to total number of follicles per group was calculated using ImagePro software. (D) Mean germinal center area. Area of each PNA+ germinal center for each group was calculated using ImagePro software by PNA+ pixel area and then converted to μm. The numbers of GC analyzed were: 201 in Control group and 102 in MAb 5261-muIg-treated group. (E) Percentages of activated T cells (CD4+/CD62Llow/CD44+) in spleens and bone marrow. Statistical significance was evaluated by Student’s unpaired t-test. *P = 0.0125; **P = 0.0016; ****P < 0.0001; P = 0.2 and 0.71 for IgG1 and IgM portions of the graph, respectively (A).

Figure 6

Effect of prophylactic and therapeutic treatment with MAb 5261-muIg on CIA progression. (A-C) Prophylactic treatment. (D-F) Therapeutic treatment. (A and D) Arthritic indexes (AI). (B and E) Summary of the histological analysis. Statistical significance was evaluated using unpaired Student’s t-test (n = 10 mice/group); **P < 0.001; ****P < 0.0001. (C and F) Toluidine blue staining of the selected limbs. (C): AI at termination: 11 (control), 4 (MAb 5261-muIg). (F): AI at termination: 10 (control), 6 (MAb 5261-muIg), 4 (Etanercept). B – Bone, C – Cartilage, Syn – Synovial tissue. Thick arrows identify affected joints; yellow arrow – pannus; red arrow – infiltration of inflammatory cells. Results shown are representative of at least three independent experiments.

Figure 7

Effect of MAb 5261-muIg on relapsing-remitting EAE in SJL/J mice. Mice (n = 15 per group) were treated starting on day 7 post-PLP_139–151/CFA immunization with 30 mg/kg of MAb 5261-muIg or an isotype control antibody i.p., once a week for 8 weeks. (A) Average EAE scores during the course of the disease. (B) Group mean scores (GMS). Statistical significance was evaluated using unpaired Student’s t-test. *P < 0.1; ****P < 0.0001.

Figure 8

Effect of MAb 5261-muIg on adoptively transferred EAE in SJL/J mice. Mice were treated i.p. with 30 mg/kg of antibody starting on the day of the adoptive transfer and twice a week thereafter for the total of 4 weeks (n = 10 per group). (A) MAb 5261-muIg in Th17-SJL EAE. (B) MAb 5261-muIg in Th17 vs. Th1 SJL-EAE: Th17 EAE: GMS +/− SE (GMS calculated from three independent experiments); Th1 EAE: GMS +/− SD (Results of one experiment +/− SD). (C) MAb 5261-muIg effect on muCXCL13-induced migration of Th17 and Th1 cells from successful SJL/J passive EAE transfer experiments (concentrations: 5 μg/ml muCXCL13; 50 μg/ml of each antibody). Results are presented as average of triplicate measurements +/− SE (except for Th1 cell migration data which originated from a single measurement). Statistical significance was evaluated using unpaired Student’s t-test (A) and one-way ANOVA followed by Bonferroni’s multiple comparison post test (B and C). **** P < 0.0001; *P < 0.1.