B CELL PHENOTYPES IN BABOONS WITH PIG ARTERY PATCH GRAFTS RECEIVING CONVENTIONAL IMMUNOSUPPRESSIVE THERAPY

Abstract

Background

In the pig-to-baboon artery patch model with no immunosuppressive therapy, a graft from an α1, 3-galactosyltransferase gene-knockout (GTKO) pig elicits a significant anti-nonGal IgG response, indicating sensitization to the graft. A costimulation blockade-based regimen, e.g., anti-CD154mAb or anti-CD40mAb, prevents sensitization. However, neither of these agents is currently FDA-approved. The aim of the present study was to determine the efficacy of FDA-approved agents on the T and B cell responses.

Methods

Artery patch xenotransplantation in baboons was carried out using GTKO/CD46 pigs with (n=2) or without (n=1) the mutant transgene for CIITA-knockdown. Immunosuppressive therapy consisted of induction with ATG and anti-CD20mAb, and maintenance with different combinations of CTLA4-Ig, tacrolimus, and rapamycin. In addition, all 3 baboons received daily corticosteroids, the IL-6R blocker, tocilizumab, at regular intervals, and the TNF-α blocker, etanercept, for the first 2 weeks. Recipient blood was monitored for anti-nonGal antibody levels by flow cytometry (using GTKO/CD46 pig aortic endothelial cells), and mixed lymphocyte reaction (MLR). CD22⁺B cell profiles (naïve [IgD⁺/CD27⁻], non-switched memory [IgD⁺/CD27⁺], and switched memory [IgD⁻/CD27⁺] B cell subsets) were measured by flow cytometry. At 6 months, the baboons were euthanized and the grafts were examined histologically.

Results

No elicited anti-pig antibodies developed in any baboon. The frequency of naïve memory B cells increased significantly (from 34% to 90%, p=0.0015), but there was a significant decrease in switched memory B cells (from 17% to 0.5%, p=0.015). MLR showed no increase in the proliferative T cell response in those baboons that had received CTLA4-Ig (n=2). Histological examination showed few or no features of rejection in any graft.

Conclusions

The data suggest that immunosuppressive therapy with only FDA-approved agents may be adequate to prevent an adaptive immune response to a genetically-engineered pig graft, particularly if CTLA4-Ig is included in the regimen, in part because the development of donor-specific memory B cells is inhibited.

Introduction

In 2000, using wild-type pig hematopoietic cell grafts in baboons, it was determined that conventional immunosuppressive therapy (in the form of cyclosporine, mycophenolate mofetil [MMF], and corticosteroids), was ineffective in preventing a T cell-dependent elicited antibody response [1]. However, Buhler et al demonstrated that the adaptive immune response could be suppressed by therapy aimed at costimulation blockade, in the form of an anti-CD154 monoclonal antibody (mAb). Prolonged survival of pig hearts and kidneys was documented in nonhuman primates (NHPs) receiving agents that blocked the CD40/CD154 pathway, with or without B cell depletion [2–11]. Subsequently, Mohiuddin et al reported that an anti-CD40mAb could successfully replace an anti-CD154mAb. However, neither of these mAbs is currently approved by the US Food and Drug Administration (FDA). Furthermore, anti-CD154mAbs have been found to be thrombogenic [5,12,13].

Today, there is an increasing availability of pigs with genetic modifications that protect the pig organ from the primate innate immune response, and/or counteract molecular incompatibilities between pigs and primates [14,15]. As a result of these genetic manipulations (now including as many as 6 in a single pig), the transplantation of pig organs into NHPs is no longer limited by early antibody-mediated (hyperacute or delayed vascular) rejection. Furthermore, some of the genetic manipulations carried out in pigs to reduce the effects of primate natural antibody and complement and coagulation activation (the humoral response), as well as the innate immune cellular response, have been found to reduce the effects of the T cell response (the adaptive cellular immune response), thus possibly reducing the need for exogenous immunosuppressive therapy [9,14,16–18].

In the USA, calcineurin inhibitors, e.g., tacrolimus and cyclosporine, costimulation blockade agents (e.g., belatacept, abatacept [both CTLA4-Ig]), antimetabolites (e.g., mycophenolate mofetil [MMF], azathioprine), or mammalian target of rapamycin inhibitors (e.g., rapamycin), and corticosteroids are all FDA-approved agents and are used after allotransplantation [19]. Previous studies showed that a CTLA4-Ig+MMF regimen was insufficient to prevent anti-pig antibody deposition on artery patch grafts from α1,3-galactosyltransfearse gene-knockout (GTKO) pigs [5]. Moreover, in NHPs MMF requires i.v. administration because it is difficult to maintain stable drug concentrations if given orally [8,9]. Tacrolimus and rapamycin have the advantage that they can be administered i.m. [8–10].

An artery patch transplant in the GTKO pig-to-baboon model that exposed the baboon to nonGal antigens and induced an adaptive immune response was reported by our group previously [5,6,9]. This model allows assessment in NHPs of the cellular immune response to a xenograft and the effect of various immunosuppressive regimens. It is a simpler model than solid organ transplantation.

The aim of the present study was to use this model to investigate changes in T and B cell phenotypes, the anti-pig antibody response, and the T cell proliferative response when the immunosuppressive regimen consists only of FDA-approved agents, i.e., various combinations of tacrolimus, rapamycin, and CTLA4-Ig.

Materials and Methods

Pig-to-baboon artery patch xenotransplantation

Three baboons from a specific pathogen-free colony (Papio spp; Division of Animal Resources, Oklahoma University Health Sciences Center, Oklahoma City, OK), 3 years-old, weighing 7–9kg, were recipients of pig artery patch grafts. All three were selected on the basis of having low anti-pig antibody levels (with a relative mean fluorescence intensity of anti-nonGal IgM<5.0 and no anti-nonGal IgG), as previously described [10,20].

GTKO/CD46+/−CIITA-DN (mutant SLA class II – that reduces expression and activation of SLA class II [9] pigs of blood group O (nonA) (Revivicor, Blacksburg, VA) were used as donors. One baboon (B3715) received a graft from a GTKO/CD46 pig, and two (B1915 and B15003) received grafts from a GTKO/CD46/CIITA-DN pig. Although we did not test the cytomegalovirus status of the three baboons, they all received ganciclovir i.v. (while i.v. catheters were in place) or valganciclovir p.o. (after catheter removal) throughout the entire course of the experiment. Both pig donors were cytomegalovirus-negative [21].

All animal care was in accordance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH publication No. 86–23, revised 1985). Protocols were approved by the University of Alabama at Birmingham Institutional Animal Care and Use Committee.

The surgical technique of artery patch transplantation has been described previously [5,6,9,22]. Briefly, in donor pigs, a length of carotid artery was excised and stored in University of Wisconsin solution in ice (at approximately 4°C) for 2–4 hours until transplanted into the baboon. The wall of the carotid artery was divided longitudinally to form a patch of approximately 2×1 cm. In the recipient baboon, the abdominal aorta was exposed. After partial heparinization (100U/kg i.v.), the aorta was clamped immediately distal to the renal arteries and at the bifurcation, and opened longitudinally. The artery patch was sutured in place as an onlay graft using 7.0 polypropylene sutures (see Figure 1 in Ref [23]). The clamps were then released and, after determining there was good blood flow to both legs, the abdomen was closed.

Immunosuppressive, anti-inflammatory, and supportive therapy

Details are provided in Table 1. All baboons received thymoglobulin (ATG) and anti-CD20mAb (Rituximab) as induction therapy. Maintenance therapy consisted of:

Table 1:

Immunosuppressive, anti-inflammatory, and adjunctive therapy

Agent	Dose (duration)
Induction
Thymoglobulin (ATG) (Genzyme, Cambridge, MA)	10mg/kg i.v. (day −3) (to reduce the CD3+T cell count to <500/mm3)
Anti-CD20mAb (rituximab) (Genentech, South San Francisco, CA)	10mg/kg i.v. (day −2)
Maintenance
Belatacept (Nulojix; BMS, Princeton, NJ)	20 mg/kg i.v. days −1, 0, 4, 7, 14 and then every 14 days
Abatacept (Orencia; BMS)	25 mg/kg i.v every 7 days
Tacrolimus (Astellas, Deerfield, IL)	0.06–0.02 mg/kg i.m. ×2/d (target trough 8–12 ng/mL) (from day −3)
Rapamycin (LC Laboratories, Woburn, MA)	0.04–0.01 mg/kg i.m. ×2/d (target trough 8–12 ng/mL) (from day −3)
Methylprednisolone (Astellas)	5 mg/kg/d tapering to 0.125 mg/kg/d
Anti-inflammatory
Tocilizumab (IL-6R blockade) (Genentech)	10 mg/kg (days −1, 7, 14, and every 2 weeks)
Etanercept (TNF-α antagonist) (Amgen, Thousand Oaks, CA)	0.5 mg/kg (days 0, 3, 7, 10)
Adjunctive
Aspirin (Bayer, Deland, FL)	40 mg p.o. (alternate days)
Low molecular weight heparin (LMWH) (Eisai, Woodcliff Lake, NJ)	700 IU/d s.c
Famotidine (APP Pharmaceuticals, Schaumburg, IL)	0.25 mg/kg/d ×2 (days −5 to 14)
Erythropoietin (Amgen)	2000 U (every 2 weeks)
Ganciclovir (Genentech)	5 mg/kg/d i.v
Valganciclovir (Genentech)	15 mg/kg/d p.o
Sulfamethoxazole and trimethoprim (Teva, North Wales, PA)	10mg/kg i.v
Sulfamethoxazole and trimethoprim oral suspension (Akorn, Lake Forest, IL)	75 mg/m2 p.o (three times weekly)

B3715, belatacept for 3m (followed by abatacept for 3m) + tacrolimus + tapering corticosteroids for 6m. (The change from belatacept to abatacept was forced on us as belatacept became unavailable for research because of a shortage of supply.)

B1915, tacrolimus + rapamycin + tapering corticosteroids for 6m.

B15013, belatacept for one month (after which it was replaced by rapamycin) + tacrolimus + tapering corticosteroids for 6m.

In addition, all 3 baboons received the IL-6R blocker, tocilizumab, the TNF-α blocker, etanercept, and the other adjunctive therapy that we routinely administer to baboon recipients of pig kidney grafts (Table1). Both tocilizumab [24] and erythropoietin [25–27] have some immunosuppressive effects that may be beneficial.

Monitoring of recipient baboons

Blood cell counts, chemistry (kidney and liver function, etc.) and coagulation parameters were measured throughout the period of follow-up [9,10]. Blood cultures were performed whenever clinically indicated. Because we could not find a laboratory in Alabama that would measure immunosuppressive drug levels in baboon blood, the tacrolimus and rapamycin levels were measured at the University of Pittsburgh Medical Center, which necessitated our not receiving the results for 48h or longer. This resulted in an inability to modify the dose rapidly, resulting in erratic control of the levels.

Immunologic monitoring has been described in detail previously [6,9,10,28] (see Methods in Ref [23]). It included (i) flow cytometry to monitor T (CD3⁺T) and B (CD22⁺B) cell numbers in the blood (to determine the effect of ATG and anti-CD20mAb) (CD22 was used as a B cell marker because all baboons were administered anti-CD20mAb); (ii) flow cytometry to monitor binding of baboon xenoreactive anti-nonGal antibodies (IgM and IgG) to GTKO/CD46 pig aortic endothelial cells (pAECs), pig red blood cells (RBCs), and pig peripheral blood mononuclear cells (PBMCs). Binding of IgM and IgG was assessed using relative geometric mean (GM), which was calculated by dividing the geometric mean for each sample by the negative control [29–31].; (iii) carboxyfluorescein diacetate succinimidyl esterase (CFSE)-mixed lymphocyte reaction (MLR) pre-transplantation and at euthanasia using GTKO/CD46 pig PBMCs as stimulators and baboon PBMCs as responders, as previously described [8,22,32]. Before the MLR assay was set-up, the cells were washed thoroughly to remove any soluble immunosuppressive drugs that might influence the result (although intracellular drug remained). Percentage of T cell (CD3⁺CD4⁺, CD3⁺CD8⁺) proliferation was analyzed by Flowjo V10 (Tree Star, Ashland, OR). The stimulation index was calculated by the xenogeneic proliferative response divided by the spontaneous proliferative response [16] (iv) flow cytometry to measure CD22⁺B cell profiles (naïve [IgD⁺/CD27⁻], non-switched memory [IgD⁺/CD27⁺], and switched memory [IgD⁻/CD27⁺] B cell subsets) using baboon PBMCs, spleen mononuclear cells (SplMNCs) and lymph node mononuclear cells (LNMNCs) [33–36] (see Methods in Ref [23]); and (v) flow cytometry to measure the CD3⁺CD20⁻ (CD4⁺CD8⁻, CD4⁺CD8⁺, CD4⁻CD8⁺) T cell profiles (naïve [CD28⁺/CD95⁻], central memory [CD28⁺/CD95⁺] and effector memory [CD28⁻/CD95⁺] T cell subsets) using baboon PBMCs, SplMNCs, and LNMNCs. In NHPs, CD95 has been shown to be a marker of memory T cells [35] (see Methods in Ref [23]).

To isolate baboon SplMNCs and LNMNCs at euthanasia, the spleen and mesenteric lymph nodes were cut into small pieces, mashed, and diluted in PBS, after which they were placed onto Ficoll-Paque PLUS (Amersham Bioscience, Piscataway, NJ) followed by centrifugation (600g, 30min). The buffy coat was collected and washed with PBS. Isolated SplMSCs or LNMSCs were passed through 70μm and 40μm meshes, as previously described [37].

Histopathology of grafts

At euthanasia (6m after transplantation), the pig artery patch xenografts (with surrounding baboon aortic tissue) were excised. The central part of the graft, i.e., avoiding the areas around the suture lines, was used for histological examination. Tissues were fixed in 10% formalin and embedded in paraffin. Sections (4μm) were stained with hematoxylin and eosin (H&E) for light microscopy [9].

Statistical analysis

Continuous variables were expressed as mean+/−SD. Comparisons between groups were performed using an unpaired t test for continuous variables. All statistical analyses were performed using social sciences software GraphPad Prism 7 (GraphPad Software, San Diego, CA).

Results

Recipient baboon survival and clinical complications

Follow-up was planned for 6m, and all baboons were electively euthanized at that time. No baboon developed any complication.

Immunosuppressive drug concentrations

Trough levels of tacrolimus were generally well-controlled, but trough levels of rapamycin varied significantly and tended to be rather higher than the original target level (see Figure 2 in Ref [23]). Mean (+/−SD) tacrolimus trough levels were 12.24+/− 0.60ng/ml (B3715), 12.16+/−0.58ng/ml (B1915), and 10.55+/−0.40ng/ml (B15013). Mean (+/−SD) rapamycin trough levels were 13.76+/−0.85ng/ml (B1915) and 11.19+/−1.06ng/ml (B15013).

WBC, lymphocyte, T and B cell counts

Mean (+/−SD) WBC counts in B3715, B1915 and B15013 were 2,760+/−471.1/μl, 2,304+/−525.1/μl, and 2,074+/−133/μl, respectively. Lymphocyte counts in B3715 gradually increased from 1m after transplantation, and those in B1915 increased after 2m (Figure 1 and see Table 1 in Ref [23]). However, those of B15013 remained low throughout the 6m period of follow-up. CD3⁺T and CD22⁺B cell numbers followed similar trends. The data suggest that rapamycin (or the abnormally high peaks achieved at times) may have inhibited the recovery of lymphocytes, T and B cells.

Figure 1: WBC, lymphocyte, T and B cell counts.

In all 3 baboons, the WBC was largely maintained at >1000 /μl. Lymphocyte counts in B3715 gradually increased from 1m after transplantation, and those in B1915 increased after 2m. However, those of B15013 remained low throughout the 6m period of follow-up. CD3⁺T and CD22⁺B cell numbers followed similar trends. (Lymphocytes (red); CD3⁺T cells (blue); CD22⁺B cells (purple); TAC=tacrolimus; Rapa=rapamycin.)

B cell phenotype (based on IgD and CD27 expression) in blood and secondary lymphoid tissues, and dynamics of repopulating B cell phenotypes after transplantation

Analysis of B cell phenotype followed the schema illustrated in Figure 3 in Ref [23]. The B cell phenotype in the blood and secondary lymphoid tissue, i.e., SplMNCs and LNMNCs, 6m after transplantation was mostly naïve B cells (Figure 2B), indicating inhibition of the development of memory B cells (Figure 2A). The frequency of naïve CD22⁺ B cells increased significantly in all baboons (from 33.60+/−6.48% at day −5 pretransplantation to 88.97+/−2.99% at 6m post-transplantation, p=0.0015). In contrast, there was a significant decrease in switched memory B cells (from 17.07+/−4.03% on day −5 to 0.53+/−0.28% at 6m, p=0.015) (see Figure 4 in Ref [23]). Between days 10–30, the frequency of double-negative B cells increased, but subsequently decreased below baseline (Figure 2A, see Figure 4 in Ref [23]). These data suggested that all 3 immunosuppressive regimen inhibited memory B cell switch.

Figure 2: B cell responses in pig artery patch recipients.

(A) Dynamics of repopulating B cell phenotypes after transplantation.

(Left) Y axis represents cell numbers. (Right) Y axis represents percentage of positive cells. These data include B cell numbers before immunosuppressive therapy was initiated (control). In B3715 and B1915, the number of CD22⁺B cells recovered 2–3m after transplantation, but the number remained few in B15013. The frequency of naïve CD22⁺B cells increased in all baboons. In contrast, there was a decrease in switched memory B cells.

(B) Analysis of B cell phenotype on the basis of IgD and CD27 expression in blood and secondary lymphoid tissues (at euthanasia 6m after transplantation).

(LNMNCs=lymph node mononuclear cells; SplMNCs=spleen mononuclear cells.) The B cell phenotype in all 3 baboons in the blood and secondary lymphoid tissue (SplMNCs and LNMNCs) 6m after transplantation was mostly naïve B cells (IgD⁺/CD27⁻/CD22⁺ cells). (N.B. In B15013, the number of naive B cells in the blood appears to be low, but the actual number is reported in Table 1 in Ref [23])

Elicited IgM and IgG antibody responses

There was no increase in IgM or IgG to nonGal antigens expressed on pAECs (Figure 3), RBCs, or PBMCs (data not shown), in any baboon, indicating no B cell sensitization to nonGal antigens.

Figure 3: Anti-pig IgM and IgG antibody levels in baboon recipients.

IgM (left) and IgG (right) antibodies were measured by flow cytometry using GTKO/CD46 pig aortic endothelial cells. Graphs represent relative geometric mean (GM) values. IgM or IgG antibodies to nonGal antigens expressed on pAECs after transplantation did not increase in any baboon compared to pre-transplantation.

Proliferative cellular response on MLR

On MLR, B1915 developed an increase in PBMC proliferation in response to GTKO/CD46 pig PBMC, indicating T cell sensitization (Figure 4). However, the two recipients that received CTLA4-Ig (B3715, B15013) showed reduced PBMC proliferation after transplantation. Although based on only one experiment, these tentative data suggest that therapy with tacrolimus and rapamycin alone may be inadequate.

Figure 4: Cellular response to pig artery patch grafts.

Baboon PBMCs were collected before (white) and 6m after (black; at euthanasia) artery patch xenotransplantation, and were tested for proliferation in response to irradiated GTKO/CD46 pig PBMCs in MLR. Proliferation is presented as stimulation index. Stimulation index was calculated by the xenogeneic proliferative response divided by the spontaneous proliferative response. B1915 developed an increase in PBMC proliferation, but B3715 and B15013 showed reduced PBMC proliferation after transplantation.

T cell phenotype (based on CD28 and CD95 expression) in blood and secondary lymphoid tissues, and dynamics of repopulating T cell phenotypes after transplantation

Analysis of T cell phenotype followed the schema illustrated in Figure 5A in Ref [23].

CD4⁺T cells

B1915 and B15013 showed gradually increasing central memory phenotypes, especially in blood (see Figure 5B in Ref [23]). The CD4⁺T cells in the LNMNCs were mostly of the naïve phenotype (see Figure 5C in Ref [23]). In B1915, central memory T cells began to increase immediately after transplantation, whereas in B15013 the increase did not begin until 50 days after transplantation (which approximated to the discontinuation of CTLA4-Ig). In contrast, in B3715, a naïve phenotype persisted in both the blood and secondary lymphoid tissue (SplMNCs and LNMNCs) (see Figure 5B,5C in Ref [23]). Although based on only one experiment, these tentative data suggest that discontinuing CTLA4-Ig may have allowed the development of central memory CD4⁺T cells.

CD8⁺T cells

One month after transplantation, B3715 showed an increasing CD8⁺ effector memory phenotype, followed by a gradual recovery to a naïve phenotype. In contrast, B1915 and B15013 showed an immediate increase in CD8⁺ effector memory phenotypes, especially with regard to the percentage of positive cells (see Figure 5D in Ref [23]). Six months after transplantation, in B3715 CD8⁺T cells in the blood and secondary lymphoid tissue were mostly of the naïve phenotype. The CD8⁺T cells in B15013 in SplMNCs were also mostly of the naïve phenotype. However, 6m after transplantation the CD8⁺T cells in B1915 and B15013 in the blood and LNMSCs were mostly of the central and effector memory phenotypes (see Figure 5E in Ref [23]). These tentative data suggest that continuing CTLA4-Ig may have inhibited the switching of CD8⁺ central and effector memory T cells.

Macroscopic and microscopic findings in pig xenografts at euthanasia

No graft showed any macroscopic abnormality, (e.g., necrosis or thrombosis) (Figure 5, left). On microscopy, no platelet adhesion/aggregation or thrombus formation on the endothelial surface was seen in any graft. B3715 showed a mild-to-moderate increase in subintimal thickness with a mild, patchy inflammatory cellular infiltrate in the outer adventitia and periadventitia. B1915 showed moderate-to-marked subintimal and medial thickening with eosinophilic inflammatory infiltrates. The graft in B15013 showed no features of inflammation or rejection (Figure 5, right).

Figure 5: Macroscopic and microscopic findings in pig artery patch xenografts at euthanasia.

(Left) No graft showed any macroscopic abnormality. (Right) The graft in B3715 showed a mild-to-moderate increase in thickness of the subintima with a mild, patchy inflammatory cellular infiltrate in the outer adventitia and periadventitia. The graft in B1915 showed moderate-to-marked subintimal and medial thickening with an eosinophilic inflammatory infiltrate. In B15013, the graft showed no cellular infiltration.

In summary, histopathological abnormalities suggestive of rejection were minimal or absent. This is in contrast to previous studies [not shown] in which a graft in a nonimmunosuppressed baboon showed a massive full-thickness eosinophilic, lymphocytic, and monocytic infiltrate, particularly at the adventitial-medial junction [9].

Discussion

Genetically-modified pigs could be an alternative source to human organs for clinical transplantation. Current preclinical results indicate increasingly extended xenograft survival in NHPs [4,7,10,11]. However, in all cases the immunosuppressive protocol included anti-CD40mAb or anti-CD154mAb costimulation blockade, but neither of these agents is likely to be employed in initial clinical trials of xenotransplantation.

Conventional agents have been administered to NHPs by several groups in the past, but blood levels tended to be significantly higher than ideal for clinical application and were associated with significant infectious complications [38–41]. To the best of our knowledge, our present study (although based on a very limited number of experiments), is the first to suggest that clinically-acceptable dosages of immunosuppressive therapy using only FDA-approved agents, especially if CTLA4-Ig is included, may be adequate to prevent an adaptive immune response to a genetically-engineered pig graft.

Our current tentative conclusions are based on the results of pig artery patch transplants in baboons. We have considerable experience with this technically-simple model, and our previous results suggest that they correlate well with those obtained after pig organ transplantation [5,8,10,22]. However, our previous studies have related almost entirely to immunosuppressive regimens that were directed towards costimulation pathway blockade, and we have not previously administered conventional immunosuppressive agents to baboons with either patch or organ grafts.

Blockade of both the CD40/CD154 and CD28/B7 pathways of costimulation is known to successfully prevent de novo antibody production in NHPs after allotransplantation [42] and xenotransplantation [2–6,8,10,11]. It was originally reported that in vitro blockade of the CD28/B7 costimulation pathway is more effective than blockade of the CD40/CD154 pathway in the inhibition of the human anti-pig T cell response [43], but our previous in vivo studies showed that CD28/B7 pathway blockade was not as efficient as CD40/CD154 pathway blockade in preventing de novo anti-pig antibody formation in xenograft recipients [5,6].

In our early studies, we did not combine CTLA4-Ig with either tacrolimus or anti-CD20mAb therapy. Recently, in clinical allotransplantation, the combination of CTLA4-Ig and an extended period (11m) of tacrolimus therapy has been reported to be associated with a significantly lower acute rejection rate than the combination of CTLA4-Ig and a short-period (5m) of tacrolimus therapy [44].

In the present study, the increase in T cell memory phenotypes seen in two baboons suggests potential T cell sensitization, but few T cells remained (see Figure 5B,5D in Ref [23]), and this may have been sufficient to prevent T cell-dependent antibody production, even in the presence of a positive cellular response on MLR in one case.

In secondary lymphoid tissues, the development of memory B cells was inhibited in all baboons, while naïve B cells gradually recovered 3 months after transplantation (Figure 2A, 2B, see Figure 4 in Ref [23]). This might be related to anti-CD20mAb and/or IL-6R blockade (tocilizumab) therapy (see below). Some previous kidney allotransplant [34] or autoimmune disease [45,46] studies also reported that anti-CD20mAb therapy resulted in repopulation of mostly naïve B cells during follow-up. McGregor et al. [47], Mohiuddin et al. [3], and our group [8] have provided data to suggest that a peri-transplant course of anti-CD20mAb improves outcome after pig organ transplantation in NHPs. In an islet allotransplantation model, Liu et al. reported that B cell reconstitution after anti-CD20mAb therapy was characterized by a preponderance of immature and transitional cells (corresponding to our naïve phenotype) [48]. Of importance, however, is that the course of anti-CD20mAb we administered (a single dose of 10mg/kg) was significantly less than that administered by McGregor and Mohiuddin and their respective colleagues (17–19mg/kg x4 on days −7, 0, 7, and 14).

In the present study, in two of the baboons the grafts were taken from a pig expressing a mutant CIITA transgene, which we have demonstrated in vitro to suppress SLA expression and inhibit SLA activation [49]. This may have played a role in the results we report. However, the graft in B3715 did not express this mutant transgene, but the outcome was good, and the immunosuppressive regimen appeared to be more important in regard to T and B cell recovery.

We have previously reported evidence for a prolonged systemic inflammatory response to a pig xenograft [22,50,51]. As inflammation can enhance the immune response [52], suppressing the inflammatory response may well facilitate the role of the immunosuppressive therapy and reduce the risk of both rejection and coagulation dysfunction [51–53]. IL-6 is a key cytokine that regulates inflammation and the development, maturation, and activation of T cells, B cells, and plasma cells [54]. Blockade of these interactions with an anti-IL-6R mAb results in (i) significant reduction in antibody production by splenic and bone marrow plasma cells, (ii) direct inhibition of plasma cell anti-HLA antibody production, and (iii) induction of T regulatory cells with inhibition of T follicular helper cells [24]. Thus, IL-6 shapes T cell immunity and is a powerful stimulant for IgG production [24,55,56]. We suggest that prolonged therapy with an IL-6R inhibitor, such as tocilizumab, may contribute to suppression of the adaptive immune response to a xenograft.

Although this is a limited study involving only three experiments, it is clear that the immunosuppressive regimens prevented B cell sensitization for up to 6m. The specific targeting of deleterious B cell subsets, especially switched memory B cells, could be of clinical benefit in xenotransplantation. In conclusion, FDA-approved immunosuppressive drug regimens (at clinically-acceptable dosages) might be able to inhibit an adaptive immune response to a genetically-engineered pig organ graft in NHPs and humans. However, it is uncertain whether these results can be confidently extrapolated to highly-vascularized solid organ xenografts. Modifications of these regimens will be tested in the pig-to-baboon kidney transplantation model.

Highlights.

• Immunomonitoring in the pig-to-baboon artery patch xenotransplantation for up to 6m

• FDA-approved immunosuppressive regimens prevented B cell sensitization.

• CTLA4-Ig including immunosuppressive regimens prevented T cell sensitization.

• Histological examination showed few or no features of rejection in xenografts.

Abbreviations

GTKO

α1,3-galactosyltransfearse gene-knockout

LNMNCs

lymph node mononuclear cells

mAb

monoclonal antibody

MLR

mixed lymphocyte reaction

MMF

mycophenolate mofetil

NHPs

nonhuman primates

pAECs

pig aortic endothelial cells

PBMCs

peripheral blood mononuclear cells

RBCs

red blood cells

SplMNCs

spleen mononuclear cells