Autoantibody production significantly decreased with APRIL/BLyS blockade in murine chronic rejection kidney transplant model

Chronic antibody mediated rejection (cAMR) remains a significant barrier to achieving long-term graft survival in kidney transplantation, which results from alloantibody production from B lymphocytes and plasma cells. APRIL (A proliferation-inducing ligand) and BLyS (B lymphocyte stimulator) are critical survival factors for B lymphocytes and plasma cells. Here we describe the results of APRIL/BLyS blockade in a murine cAMR kidney transplant model. c57/B6 mice underwent kidney transplantation with Bm12 kidneys (minor MHC mismatch), a well-described model for chronic rejection where animals cannot make donor specific antibody but rather make antinuclear antibody (ANA). Following transplantation, animals received TACI-Ig (to block APRIL and BLyS) or no treatment. Animals were continued on treatment until harvest 4 weeks following transplant. Serum was analyzed for circulating anti-nuclear autoantibodies using HEp-2 indirect immunofluorescence. Spleen and transplanted kidneys were analyzed via H&E. ANA production was significantly decreased in APRIL/BLyS blockade treated animals (p<0.0001). No significant difference in autoantibody production was found between syngeneic transplant control (B6 to B6) and APRIL/BLyS blockade treated animals (p = 0.90). Additionally, disruption of splenic germinal center architecture was noted in the APRIL/BLyS blockade treated animals. Despite the significant decrease in autoantibody production and germinal center disruption, no significant difference in lymphocyte infiltration was noted in the transplanted kidney. APRIL/BLyS blockade resulted in a significant decrease of autoantibody production and disrupted splenic germinal center formation in a chronic kidney transplant model, however in this model no difference in kidney transplant pathology was seen, which may have to do with the absence of any T cell centric immunosuppression. Regardless, these findings suggest that APRIL/BLyS blockade may play a role in decreasing antibody formation long-term in kidney transplantation. Future investigations will use APRIL/BLyS blockade in conjunction with T lymphocyte depleting agents to determine its efficacy in chronic rejection.

Introduction Antibody mediated rejection (AMR) is widely recognized as a common cause for late kidney allograft failure. [1][2][3] Mature B lymphocytes and plasma cells (terminally differentiated B lymphocytes) play a critical role not only in the development of AMR through donor specific antibody (DSA) production, but they also function as effector cells in T lymphocyte activation, which may result in cellular rejection. [4,5] Current strategies used to treat AMR include anti-CD20 antibodies (rituximab), antibody removal (plasmapheresis, intravenous immunoglobulin), and proteasome inhibitors to target plasma cells, which are the producers of alloantibody. [6,7] Despite these efforts, chronic antibody mediated rejection (cAMR) remains a significant barrier to achieving long-term graft survival in kidney transplantation. As a result, novel and effective strategies to treat antibody mediated rejection remain an unmet need. [8] Due to the multiple functions of B lymphocytes as both effector and alloantibody-producing cells, it is our hypothesis that B lymphocytes will need to be targeted at various stages of development in order to successfully prevent DSA production and rejection. APRIL (A proliferation-inducing ligand) (TNFSF 13a) and BLyS (B lymphocyte stimulator) (TNFSF 13b) are critical survival factors for plasma cells and B lymphocytes, respectively. APRIL plays an important role in plasmablast and plasma cell survival. [9,10] BLyS acts as an important costimulator of B lymphocyte survival and proliferation. [4,11,12] APRIL and BLyS have been individually investigated as potential therapeutic targets in oncology and transplant clinical trials, respectively. A recent randomized-controlled trial investigated the efficacy of anti-BLyS antibody (belimumab, GlaxoSmithKline) in addition to standard-of-care immunosuppression to reduce naïve B cells. Although this endpoint was not met, belimumab reduced memory B cells, which suggests it may have a role in long-term desensitization strategies in patients with pre-existing DSA. [13] B cell maturation antigen (BCMA), which APRIL and BLyS bind to on the surface of memory B and plasma cells, has been targeted in multiple myeloma clinical trials using anti-BCMA antibody with promising results. [14,15] We have previously reported the significant B lymphocyte depletion and DSA reduction that occurs with APRIL/BLyS blockade in allosensitized mice. [16] However, APRIL/BLyS blockade did not result in any difference in acute rejection rates, which may be due to a need for extended treatment. Therefore, our goal of the current study was to determine the effect of targeting both APRIL and BLyS in a chronic rejection kidney transplant model. Previously, MHC class II-mismatched Bm12 (B6.H-2 bm12 ) to B6 (c57/BL6) has been described as a murine model of chronic heart rejection but has not been applied to murine kidney transplant models. [17][18][19] Here we describe the results of APRIL/BLyS blockade in a novel murine chronic rejection kidney transplant model.

Animals
C57BL/6 (H-2 b ) and B6.H-2 bm12 (H-2 bm12 ) mice were purchased from Jackson Laboratories (Bar Harbor, ME) and housed in the University of Wisconsin Laboratory Animal Facility. All procedures were performed in accordance with the Animal Care and Use Policies at University of Wisconsin. Animal health including animal deaths, room temperature, 12-hour light/dark cycles, and cage cleaning among other sanitation duties were performed daily by WIMR housing staff. Food and water were available ad libitum. This research was prospectively approved by School of Medicine and Public Health Institutional Animal Care and Use Committee at the University of Wisconsin (M005204). Animals that underwent transplantation were monitored daily post-transplant. Animal health was evaluated by activity level, weight gain or loss, hunched posture, and other signs of distress. C57BL/6 animals were transplanted with a Bm12 kidney with simultaneous nephrectomy as previously described. [20] Animals were then randomized to no treatment or APRIL/BLyS blockade (treated with TACI-Ig (Transmembrane activator and calcium modulator and cyclophilin ligand interactor-Immunoglobulin) (100 μg TACI-Ig in PBS, i.p. injection 3x/week for 28d)) post-transplantation. TACI-Ig blocked both APRIL and BLyS. Animals were anesthetized with isoflurane during surgery or injections and sacrificed via cardiac puncture. Buprenorphine (1mg/kg SQ) was used administered posttransplant and every 72 hours for the first week following surgery. C57BL/6 animals transplanted with a c57BL/6 kidney were used as syngeneic transplant control. The contralateral native kidney was removed on day 1 post-transplant, leaving the animal to rely fully on the transplanted kidney. Spleen, blood, urine, and kidney were collected at 28d post-transplant for immediate utilization, storage in 10% formalin for immunohistochemistry (IHC), or were processed to single cells and cryopreserved in liquid nitrogen.

Quantification of circulating autoantibodies and biochemical
Serum samples were collected from mice 1 month post-transplant at the time of harvest. Circulating autoantibody levels were determined using NOVA Lite HEp-2 indirect immunofluorescent assay. Briefly, the serum was diluted to 1:40 with PBS, incubated with antigen substrate slides and unreacted antibodies are washed off. The substrate was then incubated with antimouse IgG FITC and the unbound reagent was washed off. A fluorescent microscope was used to image autoantibody positive samples. Quantification was performed using a custom macro written for ImageJ software (NIH, imagej.nih.gov/ij/). Three to five non-overlapping pictures of representative images were taken from animals for ImageJ analysis.
Plasma cell gating. Cells were gated to remove non-singlets, then through a large gate to ensure capture of the typically larger plasma cells, and were subsequently visualized in an IgD versus CD45R gate. IgD -CD45Rcells were then visualized as CD27 versus IgM. Plasma cells were defined as CD3 -IgD -CD45R -CD27 + IgM -CD38 + . Normalized cell counts for plasma cells were calculated based on the large gate, rather than the lymphocyte gate as was used for other lymphocyte populations.

Histology
Kidneys were collected at time of euthanasia and 1/3 of the spleen was preserved in 10% formalin for at least 24 hours, processed, paraffin embedded, and cut into 5 μm sections. After deparaffinizing and rehydrating, spleen sections were stained with anti-PAX5 (Abcam, ab140341 polyclonal). The ImmPRESS HRP reagent and ImmPACT DAB substrate were used to detect anti-PAX5 primary antibody as a brown pigment. Kidney sections were washed in distilled water, counter-stained with hematoxylin and eosin (H&E) or periodic acid-Schiff (PAS) stain and dehydrated through an ethanol series. Slides were imaged on a Nikon Eclipse E600 supplied with an Olympus DP70 camera. Automated quantification was performed using a custom macro written for ImageJ software (NIH, imagej.nih.gov/ij/). All H&E and PAS slides were reviewed by a transplant pathologist (WZ) blinded to study groups and scored for rejection according to Banff 2013 guidelines. [23] Statistics Statistics were performed using the statistical packages that are part of Prism 7 for Windows, v 7.0b. ANOVA, T-tests, and chi-square were primarily used. P values of 0.05 or less were considered significant. Statistical calculations to determine power were determined prior to implementation of this experiment.

APRIL/BLyS blockade significantly decreased anti-nuclear autoantibody production
Animals were harvested 4 weeks following transplant and their tissues were collected to evaluate for changes in ANA production, chronic kidney rejection, and B and T lymphocyte populations. This model has previously been established in preclinical cardiac transplant models as a method to study chronic rejection without immunosuppression because all allografts will develop vasculopathy. [24] ANA was used as a marker for antibody production because the minor MHC mismatch between Bm12 and C57/Bl6 (3 amino acids) does not result in alloantibody production, but rather autoantibody. [25] Post-transplant treatment with APRIL/BLyS blockade resulted in a significant reduction of serum ANA (p<0.0001) (Fig 1). Serum ANA detected in the APRIL/BLyS blockade treated group was not significantly different from that seen in the syngeneic kidney transplant group.

Germinal centers in spleen were significantly disrupted in groups treated with APRIL/BLyS blockade
Splenic germinal centers were evaluated via immunohistochemistry using PAX5 antibody in order to further characterize changes in B lymphocyte populations. PAX5 has previously been established as a marker for B lymphocytes. [26] Germinal centers demonstrated normal architecture in syngeneic and untreated groups but was completely disrupted in APRIL/BLyS blockade treated animals (Fig 5). The depletion of B lymphocytes in splenic germinal centers was also confirmed by an overall decrease in anti-PAX5 staining in APRIL/BLyS blockade treated groups compared to other groups (p<0.0001).

APRIL/BLyS blockade results in significantly decreased regulatory T cells (Tregs) but increases effector T cell populations
After characterizing B lymphocytes with APRIL/BLyS blockade, T cell populations were assessed in order to determine if the changes seen in B lymphocyte populations resulted in T cell alterations. Interestingly, both CD3 + CD4 + and CD3 + CD8 + T cell populations were significantly increased in the APRIL/BLyS blockade treated group compared to both syngeneic transplant (p<0.009) and untreated transplant groups (p<0.005). Conversely, Tregs in untreated and treated transplant groups were significantly lower compared to syngeneic transplant (p<0.05) (Fig 6).

Despite changes in ANA production and B lymphocytes, APRIL/BLyS blockade did not prevent rejection or preserve kidney function
Finally, we evaluated the transplanted kidneys 4 weeks post-transplant. Histology was reviewed by a transplant pathologist (blinded) and scored for acute and chronic peritubular capillaritis (ptc), glomerulitis (g), tubulitis (t), vasculitis (v), interstitial inflammation (i), mi (microcirculation inflammation), and arteriolar hyalinosis (ah) according to Banff 2013. Overall, no difference in rates of cellular rejection were seen between untreated and treated transplant groups (Fig 7). The overwhelming level of cellular rejection noted in the treated group is  consistent with the significant increase in T cells and decrease seen in Tregs. In the untreated group, 16.7% (N = 1) developed active AMR whereas no animals in the APRIL/BLyS blockade treated group developed AMR. Significant tubulitis, interstitial inflammation, and vasculitis was noted in both untreated and treated transplant groups. Syngeneic transplant group did not develop any ACR or AMR (Table 1).
Renal function was measured in each group using a urine protein to creatinine (UPC) ratio. Consistent with the rejection seen on histology, untreated and treated transplant groups had worse renal function compared to syngeneic transplant, although not significantly different (Fig. 7D). In the syngeneic transplant group, 66.7% (N = 2) and 33.3% (N = 1) of animals had none (UPC <0.5) or mild proteinuria, respectively. APRIL/BLyS blockade treated group demonstrated slightly worse renal function than the untreated group. Seventy-five percent (N = 3) in the untreated group had no proteinuria compared to 33.3% (N = 1) in the treated group. No animals in the untreated group had moderate proteinuria (UPC 1.5-2.0), however 33.3% (N = 1) in the treated group developed moderate proteinuria (Fig 7D).

Discussion
Chronic rejection remains a significant cause of late kidney allograft loss; therefore, novel therapies are required to treat this ongoing problem. Bm12 (H-2 bm12 ) to c57/B6 (H-2 b ) minor MHC mismatch has previously been described as a murine chronic heart rejection model with severe cardiac allograft vasculopathy and fibrosis evident within 4 weeks. [27][28][29] However, this chronic rejection model has yet to be described in the kidney transplant literature. Therefore, we investigated the effect of targeting the B lymphocyte survival factors APRIL and BLyS on chronic antibody mediated rejection in a murine kidney transplant model. APRIL/BLyS blockade via TACI-Ig resulted in significant changes in both B and T lymphocyte populations. Specifically, follicular, T2 marginal zone, mature B lymphocytes and plasma cells were depleted with APRIL/BLyS blockade. These changes in mature B lymphocyte populations were further confirmed by the disruption of splenic germinal center architecture in treated animals. Importantly, animals treated with APRIL/BLyS blockade demonstrated a significant decrease in anti-nuclear autoantibody that was reduced to syngeneic transplant levels. This autoantibody decrease is consistent with the significant depletion of plasma cells seen in APRIL/BLyS blockade treated animals, which are the primary source of antibody production. Immature B lymphocyte populations such as newly formed and T1 B lymphocytes, which are less dependent on APRIL and BLyS for survival, were not altered with APRIL/BLyS blockade. In addition to not altering immature B lymphocytes, long-lived memory B cells were not depleted.
Despite depleted mature B lymphocytes, these changes did not result in any difference in rejection seen in untreated versus treated groups. Significant cellular rejection was seen in APRIL/BLyS blockade treated animals, which corresponds to increased effector T lymphocyte and decreased regulatory T lymphocyte populations. No AMR developed in APRIL/BLyS blockade treated group whereas 16.7% (N = 1) of untreated animals demonstrated active antibody mediated rejection on biopsy. No differences in renal function, as measured by urine protein to creatinine ratio, were seen between untreated and treated groups.
Interestingly, CD3 + CD4 + and CD3 + CD8 + T lymphocytes significantly increased compared to both untreated animals and syngeneic transplant control. The increase in effector T lymphocytes demonstrated here may be due to the depletion of regulatory B lymphocytes, which would normally downregulate inflammation and autoimmunity through IL-10. [30]   Autoantibody production decreased with APRIL/BLyS blockade in murine kidney transplant model APRIL and BLyS have been extensively studied as therapeutic targets to reduce autoantibody in several preclinical models of autoimmune diseases including systemic lupus erythematous (SLE), IgA nephropathy, Sjorgren's syndrome and rheumatoid arthritis. [31][32][33][34][35] In a preclinical murine SLE model, dual APRIL and BLyS inhibition, but not BLyS blockade alone, reduced plasma cells and IgM levels. [36] Furthermore, APRIL/BLyS blockade with TACI-Fc completely prevented SLE disease progression including renal injury and arrested autoantibody production and onset of nephritis. [37] As indicated above, decreased B lymphocytes and autoantibody production through APRIL/BLyS blockade has been established in preclinical autoimmune disease models. However, the novelty of our current study is that we report decreased autoantibody production in a preclinical kidney transplant model using APRIL/ BLyS blockade. Although changes in B lymphocyte populations and decreased autoantibody production did not translate to improved renal function or decreased rejection, these initial findings are important to note and may indicate modifications that need to be made to the chronic rejection model presented here. It is possible that if the post-transplant period was extended beyond 4 weeks, then significant differences in AMR would be seen between untreated and treated groups as was seen in circulating autoantibody levels. Additionally, future investigations will use APRIL/BLyS blockade in conjunction with T lymphocyte depleting agents to determine its efficacy in chronic rejection. By including a T lymphocyte depleting agent Conclusions APRIL/BLyS blockade successfully reduced autoantibody production in a novel chronic kidney transplant rejection model. Decreased autoantibody production is likely due to depleted mature B lymphocyte populations including plasma cells. The findings presented here are further supported by the autoimmune literature in which APRIL/BLyS inhibition has been extensively studied as a method to reduce autoantibody and disease severity. However, this study remains novel due to the finding of decreased autoantibody with APRIL/BLyS blockade in a kidney transplant model. Additionally, the commonly used Bm12 to B6 chronic heart transplant model has not previously been described in kidney transplant literature. Despite these changes, cellular rejection and kidney function were not improved, which may be due to the absence of any T cell centric immunosuppression. Regardless, these findings suggest that APRIL/BLyS blockade may play a role in decreasing antibody formation long-term in kidney transplantation.