Treatment of arthritis by macrophage depletion and immunomodulation: testing an apoptosis-mediated therapy in a humanized death receptor mouse model

Abstract

Objective: To determine the therapeutic efficacy and immunomodulatory effect of an anti-human death receptor 5 (DR5) antibody, TRA-8, in eliminating macrophage subsets in a mouse model of type II collagen-induced arthritis (CIA).

Methods: A human/mouse-chimeric DR5-transgenic mouse, under the regulation of a mouse 3-kb promoter and a loxP-flanked STOP cassette, was generated and crossed with an ubiquitous Cre (Ubc.Cre) mouse and a lysozyme M-Cre (LysM.Cre)-transgenic mouse to achieve inducible or macrophage-specific expression. Chicken type II collagen was used to induce CIA in mice, which were then treated with an anti-human DR5 antibody, TRA-8. Clinical scores, histopathologic severity, macrophage apoptosis and depletion, and T cell subset development were evaluated.

Results: In human/mouse DR5-transgenic Ubc.Cre mice with CIA, transgenic DR5 was most highly expressed on CD11b+ macrophages, with lower expression on CD4+ T cells. In human/mouse DR5-transgenic LysM.Cre mice, transgenic DR5 was restrictively expressed on macrophages. Both in vivo near-infrared imaging of caspase activity and TUNEL staining demonstrated that TRA-8 rapidly induced apoptosis of macrophages in inflamed synovium. Depletion of pathogenic macrophages by TRA-8 led to significantly reduced clinical scores for arthritis; decreased macrophage infiltration, synovial hyperplasia, osteoclast formation, joint destruction, cathepsin activity, and inflammatory cytokine expression in joints; reduced numbers of Th17 cells; and an increased number of Treg cells in draining lymph nodes.

Conclusion: The anti-human DR5 antibody TRA-8 was efficacious in reducing the severity of arthritis via targeted depletion of macrophages and immunomodulation. Our data provide preclinical evidence that TRA-8 is a potential novel biologic agent for rheumatoid arthritis therapy.

Figure 1

Human, mouse and chimeric DR5 and their expression and function in inducing apoptosis by TRA-8. A, Amino acid alignment of human and mouse DR5. Junction between human and mouse DR5 in the chimeric molecule, transmembrane, and death domain are as indicated. B, Schematic diagram representing the human, mouse and chimeric DR5 (left), and human, mouse and chimeric DR5 constructs generated with different promoters and regulatory elements (right). C, FACS analysis of cell surface expression of DR5 recognized by an anti-human DR5 antibody in NIH3T3 cells transiently transfected with the indicated constructs. Percentage of the hu/mo DR5⁺ cells was shown. N ≥ 3. * P < 0.05 versus result from construct 1. D, TRA-8 mediated killing in NIH3T3 transiently transfected with the indicated constructs. Cell viability was determined using the ATPLite assay. For panels C and D, the data represent the mean ± s.e.m. (n ≥ 3). Hu, human; mo, mouse; chim, chimeric. * P < 0.05 and ** P < 0.01 versus isotype control.

Figure 2

Hu/mo-chimeric Tg DR5 expression and function in Ubc.Cre DR5 Tg mice. A, Real-time PCR analysis of the chimeric DR5 expression in the indicated tissues obtained from the Ubc.Cre DR5 double-positive transgenic mice after induction of Cre expression by tamoxifen. Expression is represented as the ratio of copy numbers of chimeric Dr5 or endogenous mouse Dr5 to those of Gapdh. The correlation R² and correlation P value are shown in the upper-right of the panel. B, Transgenic chimeric DR5 cell surface expression on different immune cells was analyzed by FACS (left panels). The chimeric DR5 expression is representative of three experiments (right panels). C, The percentage of CD11b^high spleen macrophages from mice with and without TRA-8 treatment (0.2 mg, i.v., twice/week for 2 months) was determined by FACS. Data represent the mean ± s.e.m. (n ≥ 3). D, Proliferation of LPS-stimulated spleen cells from DR5 Tg⁻ Ubc.Cre and hu/mo DR5 Tg⁺ Ubc.Cre mice treated with TRA-8 or isotype control antibody was determined using the [³H]-thymidine incorporation assay (n ≥ 3). * P < 0.05 versus isotype control.

Figure 3

TRA-8 treatment prevents the development and attenuates the severity of CIA in hu/mo-chimeric DR5 Tg⁺ Ubc.Cre mice. CIA was induced in the indicated DR5 Tg⁺ (A) and DR5 Tg⁻ mice (B) as described in detail in the methods section. Chicken type II collagen emulsified in CFA and IFA was administered intradermally on day 0 and day 32 respectively (arrows). Early and late TRA-8 treatment (0.2 mg/mouse, once per week) was initiated on day 0 and day 30 respectively (arrow heads) and continued until sacrifice. Clinical scores (0–3 per paw; n=6 per group) were assessed daily until the mice were sacrificed. Data are presented as mean arthritis score ± s.e.m.. ** P < 0.01 versus TRA-8-treated groups at the same time point. C and D, Representative H&E and Mac-3 immunohistochemical staining of the knee joint sections from a TRA-8 treated hu/mo DR5 Tg⁺ Ubc.Cre (C) or a TRA-8 treated DR5 Tg⁻ Ubc.Cre (D) mouse. The magnification of the objective lens used to acquire the indicated images is shown in the left. E, erosion; H, hyperplasia; M, macrophages, (Scale bar, 100 µm).

Figure 4

Expression of hu/mo-chimeric DR5 in macrophages and selective depletion by TRA-8 in hu/mo DR5 Tg⁺ LysM.Cre mice with CIA. CIA was induced as described in the methods section. A group of mice were treated with TRA-8 (0.2 mg/mouse, i.v., twice/week) until analysis at day 60 post primary cCII injection. A, Flow cytometry analysis of the expression of Tg DR5 as indicated by cell surface binding to an anti-human DR5 antibody on CD11b⁺ cells from the draining LN (upper panels). Chimeric Tg DR5 expression on CD11b⁺ cells by immunofluoresence staining of CD11b and hu/mo-chimeric DR5 on cells prepared by cytospin without cell permeabilization (lower panels). Chimeric Tg DR5 was detected using the Tyramide Signal Amplification (TSA) technique. B, Flow cytometry analysis of the expression of Tg DR5 by cell surface binding to an anti-human DR5 antibody on the indicated cell types from the draining LN. C and D, TRA-8 treatment depleted CD11b⁺ macrophages and CD11b⁺Ly6C⁺ inflammatory macrophages from draining LN from the indicated groups of mice by dot plot (left) and bar graph (right) analysis. Data represent the mean ± s.e.m. (n ≥ 3). * P < 0.05 and ** P < 0.01 between the indicated comparisons.

Figure 5

Apoptosis induced by TRA-8 in the joints of hu/mo-chimeric DR5 Tg⁺ LysM. Cre mice with CIA. A, In vivo imaging of TRA-8 induced apoptosis in joints. B, Quantitation of the imagines. DR5 Tg⁺ LysM.Cre and control LysM.Cre mice were induced to develop CIA. At 8 week after induction, baseline levels of the caspase activity were measured using the caspase-targeted activity based probe AB50-Cy5 (left panel). Mice were then treated with TRA-8 (0.2 mg, day 0 and 3) and apoptosis imaging using AB50-Cy5 was performed on the same mice on day 6 after initiating TRA-8 treatment (right panel). C, Immediately after the second AB50-Cy5 imaging, the joints were removed, fixed and processed for H.&E. (left), Mac-3 staining (middle, counterstained with methyl green), and TUNEL (right, counterstained with hematoxyline). D, Quantitative analysis of the TUNEL⁺ cells of the indicated cell types in synovium. The values on the Y-axis represent the percentages of TUNEL-positive cells of the indicated cell types. Five randomly chosen fields of synovium were evaluated for each section. E, erosion; H, hyperplasia; M, macrophages, and TU, TUNEL. Scale bar, 100 µm. Data represent mean ± s.e.m. (n ≥ 3). ** P < 0.01 between the indicated comparisons.

Figure 6

TRA-8 treatment ameliorates the severity of CIA in hu/mo-chimeric DR5 Tg⁺ LysM.Cre mice. A, CIA was induced in hu/mo DR5 Tg⁺ LysM.Cre (upper panel) and control DR5 Tg⁻ LysM.Cre mice (lower panel). Arrows indicate the intradermal injection of chicken type II collagen on day 0 and 30. TRA-8 treatment was initiated on day 28 (arrow heads). Clinical scores (n = 6 per group) were assessed until the mice were sacrificed on day 60. Data are presented as the mean clinical score ± s.e.m. * P < 0.05 and ** P < 0.01 versus TRA-8 treated group at the indicated time point. B, Cathepsin activity in joints was measured by in vivo imaging using the NIRF-probe ProSense 750. C, Quantitative analysis of ProSense 750 intensity. Data are presented as the mean ± s.e.m. * P < 0.05 between the indicated comparisons. D, Histological assessment of representative knee joints from TRA-8-treated DR5 Tg⁺ LysM.Cre DR5 (left panels) and DR5 Tg⁻ LysM.Cre mice (right panels), which included H&E, Mac-3, and TRAP staining as indicated (scale bar: 100 µm). Both groups were treated with TRA-8 weekly for one month. E, erosion; H, hyperplasia; M, macrophages, and TR, TRAP.

Figure 7

TRA-8 treatment decreases the expression of pro-inflammatory cytokines and exhibits immunomodulatory effects in hu/mo DR5 Tg⁺ LysM.Cre mice. CIA was induced in hu/mo DR5 Tg⁺ LysM.Cre and control DR5 Tg⁻ LysM.Cre mice as described in the methods section. TRA-8 treatment was initiated on day 28 until analysis at day 60 post primary cCII injection. A, Sera levels of IL-6 and IL-17A of indicated mice with TRA-8 treatment were analyzed by ELISA. B, Absolute copy numbers of Tnfa, Il6, Il17, Il23(p19), Irf5 and Foxp3 of synovium of the indicated mice were determined by qRT-PCR and represented as copy number ×10⁵/Gapdh. C, Percentage of IL-23⁺ CD11b⁺ macrophages from the draining LN of the indicated mice treated with TRA-8 on day 60 (CD11b⁺ gated) was determined by flow cytometry. D, Percentage of Th17 (IL-17⁺) and Th1 (IFN-γ⁺) cells (CD4⁺ gated) and Tregs (CD4⁺, Foxp3⁺) from the draining LN of the indicated mice treated with TRA-8 on day 60 were analyzed by flow cytometry. Data are presented as the mean ± s.e.m. * P < 0.05 and ** P < 0.01 between the indicated comparisons.