Costimulation blockade and regulatory T-cells in a non- human primate model of kidney allograft transplantation Haanstra, K.G.

(1)

human primate model of kidney allograft transplantation

Haanstra, K.G.

Citation

Haanstra, K. G. (2008, March 13). Costimulation blockade and regulatory T- cells in a non-human primate model of kidney allograft transplantation.

Retrieved from https://hdl.handle.net/1887/12636

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden

Downloaded from: https://hdl.handle.net/1887/12636

Note: To cite this publication please use the final published version (if applicable).

(2)

No synergy between ATG induction and costimulation blockade induced kidney allograft survival in rhesus monkeys

Krista G. Haanstra¹, Ella A. Sick¹, Jan Ringers², Jacqueline A.M. Wubben¹, Eva- Maria Kuhn³, Bert A. ’t Hart^1,4, Louis Boon⁵, and Margreet Jonker¹

Transplantation 2006; 82(9): 1194-1201

1Department of Immunobiology, Biomedical Primate Research Centre, Rijswijk, The Netherlands.

2Department of Surgery, Leiden University Medical Centre, Leiden, The Netherlands.

3Intervet International BV, Boxmeer, The Netherlands.

4Department Immunology, Erasmus Medical Centre Rotterdam, Rotterdam, The Netherlands.

5PanGenetics BV, Utrecht, The Netherlands.

(3)

Abstract

BackgroundCostimulation blockade with antibodies directed against human CD40 and CD86 leads to prolonged kidney allograft survival in rhesus monkeys, but fails to induce permanent graft acceptance. We have tested whether costimulation blockade is more effective after peripheral T-cell ablation with antithymocyte globulin (ATG), with the aim to remove already primed autoreactive cells present in the nor- mal repertoire.

MethodsRhesus monkeys were transplanted with a mismatched kidney allograft.

ATG was given around the time of transplantation (day -1 and 0). Costimulation blockade with anti-CD40+anti-CD86 was given at tapering dosages from day -1 to 56. Cyclosporin A (CsA) was given from day 42 onwards and ﬁrst rejections occurring after day 42 were treated with Prednisone.

ResultsWe observed accelerated rejection in ATG-treated monkeys, compared to animals receiving only costimulation blockade. The accelerated rejection of the kidney allograft occurred despite the application of rejection therapy with steroids and CsA.

Three of the five ATG-treated animals were found seropositive for donor-specific alloantibodies. Early biopsies (day 21) from animals treated with ATG and anti- CD40+anti-CD86 show substantially reduced expression of cytotoxic T lymphocyte associated antigen-4 (CTLA-4) and forkhead box P3 (FOXP3) in focal infiltrates as compared to animals treated with only costimulation blockade. Furthermore, we observed the rapid reappearance of CD8⁺T-cells with a memory phenotype (disap- pearance of naive CD95^low/CD11a^lowT-cells) in peripheral blood.

ConclusionWe conclude that (subtotal) T-cell depletion using ATG does not add to costimulation blockade induced kidney allograft survival.

(4)

Introduction

The immature immune system of naive mice can be easily diverted to a state of tolerance, but it is far more difficult to achieve this in the presence of acquired immunological memory [1]. Conceptually, immunological memory towards alloantigens can be the result of priming of T-cells by the specific alloantigens or by pathogens that evoke a cross-reactive immune reaction towards the alloantigens (heterologous immunity) [2]. As primed T-cells are less dependent on costimulatory signals for their reactivation, costimulation blockade may be less effective in the induction of allospe- cific tolerance in non-SPF animals [2, 3]. We hypothesised therefore that ablation of peripheral T-cells with antithymocyte globulin (ATG) may have a dual beneficial effect on graft acceptance in a rhesus monkey model of kidney transplantation. We reasoned that ATG not only removes the allocrossreactive T-cells that seem to ob- struct graft acceptance [2, 3], but also may render the immune system more sensitive to tolerization, as was shown in mice [4].

In the current study, we have treated rhesus monkeys in a kidney allograft model with antibodies against human CD40 and CD86 to achieve costimulation blockade.

Treatment with these antibodies was found to prevent graft rejection during the treatment period, although long-term drug-free allograft survival was not achieved [5]. Prolonged graft survival was associated with expression of CTLA-4 and FOXP3 in early graft biopsies.

A reduction in the number of circulating peripheral T-cells in rhesus monkeys was achieved with the human T-cell depleting agent ATG. This reagent has been effectively tested as an induction therapy of immunosuppression for kidney transplantation in humans [6]. Our results show that monkeys receiving ATG at the start of the treatment with anti-CD40 plus anti-CD86 antibodies rejected their kidney allografts signiﬁcantly earlier than monkeys receiving only costimulation blocking antibodies. Rejection occurred in spite of high dose CsA after day 42 in both groups.

Pretreatment with ATG appeared associated with an increase of the incidence of al- loantibody production.

The observation that ATG treatment at the start of the installation of costimulation blockade reduces the beneﬁcial effect of the latter treatment on allograft survival in a valid preclinical model of kidney transplantation is important for the clinical set- ting.

Materials and methods

Animals and renal transplantation

Naive, captive bred 3.5-8 kg rhesus monkeys (Macaca mulatta) (n=5) were typed for Mamu-A, and -B antigens by serology [7]. Disparity for serologically deﬁned DR locus antigens was conﬁrmed by Mamu-DRB typing at the molecular level [8]. Re- cipients were mismatched for one or two Mamu-DR antigens, and had at least one Mamu-A and -B mismatch with the donor. In addition, selected monkeys had a stimulation index of >3 in a one-way mixed lymphocyte reaction (MLR) of the re-

(5)

cipient cells directed against the donor antigens. Heterotopic kidney allotransplan- tation with bilateral nephrectomy was performed on postoperative day (POD) zero, as previously described [5, 9]. All procedures were performed in accordance with guidelines of the Institutional Animal Care and Use Committee installed by Dutch law.

Eighteen-gauge needle biopsies were collected from transplanted kidneys at regular time intervals. For histological examination of biopsy and necropsy tissue samples, these were formalin-fixed and paraplast-embedded. Then 4 μm thick sections were stained with H&E, periodic acid Schiff, and a silver impregnation staining (Jones). Histopathological evaluation of allograft rejection was performed according to the Banff classification [10]. Anti-donor antibodies were determined by incu- bation of donor cells with recipient serum+anti-immunoglobulin (Ig) G or anti-IgM antibodies and subsequent fluorescence-activated cell sorting (FACS) analysis as described previously [5].

Immunosuppressive treatment

Immunosuppressive treatment with antagonistic chimeric monoclonal antibodies (mAbs) speciﬁc for human CD40 (ch5D12) and human CD86 (chFun-1) (Pan- Genetics BV, Utrecht, The Netherlands) was given intravenously (i.v.) as described previously [5] to both groups (Table 4.1). Brieﬂy, 20 mg/kg dosages of both antibodies were given on PODs -1 and zero, 10 mg/kg dosages were given on PODs 4, 7, 11, and 14 and 5 mg/kg dosages were given twice weekly from POD 18 until 56. In addition to the costimulation blockade, CsA (Sandimmune, Novartis) was given once daily by intramuscular (i.m.) injections to the animals in the costimulation blockade (COS)/CsA group and in the ATG group from POD 42 onwards (Table 4.1). Tapering dosages were given to the COS/CsA group to achieve trough levels of 300 ng/ml from POD 42-70; 200 ng/ml from POD 70-98 and 100 ng/ml from POD 98-126, as described previously [11]. CsA trough levels in animals in the ATG group were targeted to be 300-400 ng/ml until POD 98, where after CsA was discontinued. Twenty-four-hour CsA trough levels were monitored twice a week.

CsA blood levels were measured using a radioimmunoassay after ethanol extraction (INCSTAR Corp., Stillwater, MN) of ethylenediamine-tetraacetic acid (EDTA) blood samples and dosing was adjusted based on these levels to reach indicated target trough levels. The animals in the ATG group (Table 4.1) received in addition to the costimulation blocking antibodies and the CsA also Thymoglobulin (ATG) (Imtix- Sangstat, France) at a dose of 20 mg/kg; i.v. on POD -1 and subcutaneously on POD zero. To minimise side effects of i.v. ATG administration, methylprednisolone (Solu- Medrol, Pharmacia & Upjohn, Woerden, The Netherlands) was administered i.v. at a dose of 10 mg/kg at POD -1 prior to ATG infusion.

For animals in the ATG group only, ﬁrst rejections (a 50% rise in serum creatinine levels within two subsequent twice weekly measurements) if occurring between PODs 42 and 56 were treated with three i.m. doses of 10 mg/kg methylprednisolone followed by 1 mg/kg i.m. prednisone (Di-adreson-F, Nourypharma, The Nether- lands), which was tapered after POD 90. Animals rejecting before POD 42 were not

(6)

Group Treatment 1st rejection episode Graft survival

(days) (days)

ATG ATG 38, 18, >49^a, 56, 46 38, 54, >56^a, 97,

+αCD40/αCD86 + CsA 109

+ steroid rescue therapy

COS/CsA αCD40/αCD86 + CsA 140, 228, >1750, 140, 231, >1750,

>1780 >1780

Table 4.1: Description of experimental groups. Costimulation blockade (COS) was achieved with antibodies against human CD40 and human CD86, which were administered from POD -1 to 56. Dosages of 20 mg/kg were given on PODs -1 and 0;

10 mg/kg was given twice weekly for the next two weeks, and 5 mg/kg was given twice weekly for the remaining six weeks. The COS/CsA group received in addition to the costimulation blockade with antibodies also CsA from POD 42 to 126, starting with 300 ng/ml trough levels, which was tapered every four weeks with 100 ng/ml.

The ATG group received in addition to the costimulation blockade and CsA treatment an induction therapy with ATG on PODs -1 and zero. Methylprednisolone was given on POD -1 to alleviate the side effects of the ATG. CsA was given from POD 42 to 98 targeting trough levels between 300-400 ng/ml. Rejections, as deﬁned by a 50%

rise in serum creatinine levels within two subsequent twice weekly measurements, occurring after POD 42, were treated with three dosages of 10 mg/kg methylprednisolone and subsequently 1 mg/kg i.m. prednisone, which was tapered after POD 90. Rejections occurring before POD 42 were not treated. In one animal this led to final rejection on day 38 and in another animal (day 54), the first rejection was on POD 18, but this resolved partially and the following rejection after POD 42 was treated with steroids. The first rejection episode (50% creatinine rise within two subsequent twice weekly measurements) is depicted in the same order of animals relative to the graft survival. First rejections occurred significantly earlier in the ATG group (p = 0.0147) and despite anti-rejection therapy, graft survival was also significantly earlier in the ATG group (p= 0.0101). âUreter obstruction, no rejection

treated and when serum creatinine showed a signiﬁcant rise or when the clinical con- dition began to deteriorate after animals had already received a rejection treatment, the animals were euthanized and necropsy was performed.

FACS analysis

Subset analyses were performed at regular time points using whole EDTA blood.

Aliquots of whole blood (50 μl) were washed to remove circulating free antibody present in the serum. Cells were stained with fluorochrome-conjugated antibodies directed against human CD4, CD8 and CD20 (respectively clones SK3, SK1 and L27, Becton Dickinson, San Jose, CA). After lysis of the red blood cells, the remaining cells were washed and fixated using formaldehyde. Staining patterns were visualised by flow cytometry (FACSort, BD) and data analysis was performed using CellQuest

(7)

software (BD). In addition, the presence or absence of naive cells was analysed by staining of stored frozen cells. The percent naive cells can be reliably determined in rhesus monkeys by staining of CD4⁺ and CD8⁺ cells with CD95 and CD11a, whereby the naive cells are determined as CD95^low/CD11a^low[12] and memory cells are determined as 100% - percentage CD95^low/CD11a^lowcells. Cells were stained for CD3, CD4, CD8, CD95 (clones SP34, SK3, SK1 and DX2 respectively from BD/BD Pharmingen), and CD11a (clone B-B15, Diaclone). To enable ﬁve- and six-colour analysis, staining patterns were visualized using a FACSAria. Analysis was performed using BD FACSDiva software. Absolute numbers of T-cells with a naive or memory phenotype were calculated from the number of lymphocytes in fresh EDTA blood and the proportion of naive and memory lymphocytes as determined from frozen cells.

Immunohistochemistry

Five-micrometre thick cryosections were immunolabelled with antibodies speciﬁc for CD2 (clone T11, Beckman Coulter, CA), CD4 (home-made mix of OKT4, OKT4a, RIV6, RIV7 and MT-310), CD8, CD20, CD68 (clones DK25, L26 and KP1, DAKO, Denmark), CD83 (HB15A; Immunotech, Czech Republic), CD152 (CTLA-4; clone BNI3, BD Pharmingen, CA), and FOXP3 (hFOXY, eBioscience, CA) or with a poly- clonal antibody speciﬁc for FOXP3 (ab2484, Abcam, United Kingdom). Slides were incubated for 20 hr at 4^◦C. After washing, slides were incubated with a biotinylated rabbit anti-mouse mAb or a biotinylated donkey anti-goat mAb (ab2484), at room temperature. Slides were incubated with Strep ABComplex/HRP (CD2, CD4, CD8, CD20, CD68 and CD83) or Strep ABComplex/AP (CD152 and FOXP3) for 30 min.

Substrate (DAB or Fuchsine) was added for 25 min. Slides were counter stained with haematoxylin. Stained cells were counted and expressed as percent positive cells of the total number of infiltrating cells. Focal infiltrates and diffuse infiltrates were evaluated separately. Several microscopic fields for each infiltration type were scored blindly. Results of both FOXP3 stainings were comparable, with respect to intra-animal variation as well as with respect to group means. The mean numbers of FOXP3⁺cells have been calculated from both stainings.

Statistic analysis

Data are represented as means±SEM, unless otherwise indicated. Animal survival was analysed by the Kaplan-Meier method and the log-rank test was used to determine statistical signiﬁcance between the groups [13]. For non-parametric data, the Mann-Whitney U test was used. A p-value of≤0.05 was considered signiﬁcant.

(8)

Results

Accelerated graft rejection in monkeys receiving costimulation blockade with ATG

Without immunosuppressive treatment, rhesus monkeys reject an MHC mismatched kidney allograft within 6±0.7 days [9]. We previously showed that the rejection of a kidney allograft is substantially delayed to 80±9 days in rhesus monkey recipients receiving costimulation blocking antibodies speciﬁc for human CD40 and CD86 [5].

Combination of costimulation blockade with delayed CsA treatment even induced long-term survival of the kidney allograft in 50% of these treated monkeys (Table 4.1;

COS/CsA group) [11].

In the current study we have investigated whether T-cell ablation with ATG in combination with costimulation blockade and delayed CsA treatment further pro- longs the survival of the kidney allograft. The ATG preparation used depletes T-cells from peripheral blood and lymph nodes [14]. In the current study, ATG was given one day before and at the day of transplantation, i.e. PODs -1 and zero. An eight- week treatment with anti-CD40 and anti-CD86 was started at POD -1, identical to the previously published COS/CsA group [5, 11]. Starting POD 42, CsA treatment was given to both the ATG and COS/CsA groups. Tapering dosages of CsA were given to animals in the COS/CsA group to investigate compatibility of COS treatment with conventional immunosuppression. CsA treatment was started POD 42 in the ATG group, identical to the COS/CsA group. However, we observed increased numbers of clinical rejections, in the ATG group. Rejections are deﬁned by a 50%

rise in serum creatinine within two subsequent twice-weekly measurements. CsA treatment was not tapered after four weeks, as in the COS/CsA group, because of the high rejection rate, but maintained at high dose for a total of eight weeks. In addition, ﬁrst rejection episodes occurring after POD 42 were treated with steroids.

The Kaplan-Meier survival analysis (Fig. 4.1) shows that despite this increased immunosuppression in the ATG group, ﬁrst rejections occurred signiﬁcantly earlier in the ATG group as compared to the COS/CsA control group (p = 0.0147, log-rank test). High-dose immunosupppression with COS blockade, CsA, and prednisone could not prevent kidney rejection (mean survival time 71±14 days, Table 4.1).

Increased numbers of early rejections (a 50% rise in serum creatinin within two subsequent twice-weekly measurements, before POD 42) were observed in the ATG group (2 of 5, PODs 21 and 38) as compared to the four animals of the COS/CsA group [11] and the previously published ﬁve animals treated with costimulation blockade only [5] (0 of 9; log-rank test, p = 0.0448). Note that until POD 42, all 14 animals received identical immunosuppression, except for the ATG and methyl- prednisolon (POD -1) in the ATG group. The observed rejections during costimulation blockade treatment, occurring in spite of high-dose CsA and steroids in the ATG group, prompted us to investigate the early (until POD 42) effects of the ATG treatment. This allows us to compare animals of the ATG group with the nine animals of the, for this early period, combined COS-COS/CsA group.

(9)

Figure 4.1: Kidney allograft survival. Kaplan-Meier survival curve of the graft survival time until the ﬁrst rejection episode (creatinine rise of≥50% within 3-4 days).

Animals in the ATG group (solid line) reject their allograft signiﬁcantly earlier than animals in the COS/CsA group (dotted) (p= 0.0147). One animal in the ATG group had a creatinine rise of≥50% on POD 49, but this was due to a blocked ureter, as became apparent on POD 56, at euthanasia.

Disturbed memory versus naive T-cell balance after ATG treatment

Early effects of the ATG treatment on peripheral blood T-cell subsets were studied.

In conformity with literature data [15, 16], CD4⁺as well as CD8⁺cells are partially depleted from the peripheral blood after ATG infusion (Fig. 4.2A). The same ﬁgure shows that the reduction of CD4⁺cell levels (below 25% of initial value) is sustained for prolonged time (>50 days), while already at POD 21 a signiﬁcant number (50%) of CD8⁺T-cells returned into the circulation, leading to a reversal of the CD4/CD8 ratio. The animals of the COS-COS/CsA group also demonstrate reduced peripheral T- and B-cells on POD zero and seven. This is caused by a reduction in the total number of lymphocytes in the circulation, but the relative portions of CD4⁺, CD8⁺ and CD20⁺ cells remain unaltered, and the CD4/CD8 ratio is not reversed. The effect seems to be caused by treatment with anti-CD40+anti-CD86 and not by the transplantation itself, because the effects were already observed on POD zero, before transplantation.

We subsequently analysed blood samples collected at PODs -7, 21, 42 and at necropsy by flow cytometry to determine whether reappearing cells were naive or memory T-cells. CD3⁺CD4⁺ and CD3⁺CD8⁺ T-cells were analysed for expression of CD95 and CD11a. As demonstrated by Pitcher et al. [12], naive cells are CD95^low/ CD11a^low(Fig. 4.2B, gated cells) while memory cells are the remainder of the CD4⁺ and CD8⁺ T-cells. ATG-treated animals all showed a significant reduction in the percentage of circulating naive CD4⁺ and CD8⁺cells already on POD 21, the first time point measured after POD zero, while this was not observed in the COS group

(10)

ATG group COS group p-value (t test)

CD4

Naive (10⁹cells/l) 0.03±0.01 0.68±0.31 0.03 Memory (10⁹cells/l) 0.11±0.04 0.42±0.18 0.20 Ratio naive/memory 0.31±0.11 1.51±0.37 0.03 CD8

Naive (10⁹cells/l) 0.03±0.02 0.31±0.14 0.06 Memory (10⁹cells/l) 0.32±0.19 0.27±0.13 0.89 Ratio naive/memory 0.14±^0.06 ^1.24±^0.31 ^0.03

Table 4.2: Early effects of ATG on naive and memory T-cell subsets were studied in cells taken on POD 21. CD4⁺ and CD8⁺ T-cells were analysed for naive (CD95^low/CD11a^low) and memory (non-naive) subsets. Data are given as mean±SEM. Absolute numbers of memory cells do not differ between groups, but absolute numbers of naive CD4⁺ and CD8⁺ cells are lower in the ATG group, although only signiﬁcant for naive CD4⁺cells. The ratio of naive and memory T-cells reveals a signiﬁcant difference between both groups for both CD4⁺and CD8⁺cells.

In the COS group, naive cells are present in access over memory T-cells. In contrast, animals treated with ATG have less naive T-cells than memory T-cells.

(see Fig. 4.2B for a representative example). Absolute numbers of naive and memory T-cells present on POD 21 were determined, to assess early effects of the ATG.

While absolute numbers of memory CD4⁺ and CD8⁺ cells were not different between groups, absolute numbers of naive CD4⁺ cells were signiﬁcantly lower in ATG-treated animals (Table 4.2). A disturbed balance between memory and naive cells is also apparent from the ratio between naive and memory cells. In animals from the COS group CD95^low/CD11a^low naive cell numbers exceed memory cell numbers for both CD4 and CD8 cells (ratio 1.51±0.31 and 1.24±0.31, respectively), while these ratios are reversed in ATG-treated animals (0.31±0.11 and 0.14±0.06, respectively; Table 4.2). The naive/memory CD4 and CD8 T-cell ratio’s differ signif- icantly between both groups (both p = 0.0286).

Naive CD3⁺CD4⁺ and CD3⁺CD8⁺cells were also studied using the true naive CD45RA⁺CCR7⁺subset [17]. Although Pitcher et al. [12] report that CD45RA is a less informative marker of naive T cells in rhesus monkeys, we found comparable results using CD45RA⁺CCR7⁺ T-cell subsets as markers of naive cells (CD45RA⁺CCR7⁺) as described above for the CD95^low/CD11a^lowsubsets (data not shown).

Immunohistochemical analysis of graft-inﬁltrating cells

Early effects of ATG on the expression of regulatory T-cell (Treg) markers by graft inﬁltrating cells were studied. Kidney biopsies from POD 21 were analysed for expression of the markers CD2, CD4, CD8, CD20, CD68, CD83, CTLA-4 (CD152) and FOXP3. Despite (slightly) elevated serum creatinine levels in animals of the ATG group, no differences between animals treated with and without ATG were found

(11)

!

Figure 4.2: Effect of ATG on whole blood levels of CD4⁺, CD8⁺ and CD20⁺ lymphocytes and on naive CD4⁺ and CD8⁺ T-cell subsets in PBMC. (A) Whole blood stainings with CD4, CD8 and CD20 show different mean absolute number of CD4⁺ cells+SEM, of CD8⁺cells−SEM, and of CD20⁺cells+SEM of all animals of the ATG group and the COS group. The CD4/CD8 T-cell ratio is reversed in the ATG group after T-cell depletion as compared to pretreatment values. The temporary lowering of the number of lymphocytes in the COS-COS/CsA group is due to a decrease in the number of circulating lymphocytes, and does not lead to a reversal of the CD4/CD8 T-cell ratio. (B) Isolated and stored PBMC were used for analysis of naive T-cell subsets. Gates were set on CD3⁺/CD4⁺ and CD3⁺/CD8⁺ cells. Subsequently, gates were set around naive cells (CD11a^low/CD95^low). Shown are representative examples of an animal of the ATG group (+ATG; survival 109 days) and of an animal of the COS group (-ATG; survival 75 days). Stainings of cells isolated at POD -7 and 21 are shown. Numbers in the graphs indicate the percent of cells in the gate for naive cells for CD4 (upper panels) and CD8 T-cells (lower panels). Both the naive CD4 and the CD8 T-cell subsets are reduced with >35% after T-cell depletion on POD 21.

(12)

ATG group COS-COS/CsA group p-value (Mann-Whitney)

CD4 45±15 (4) 73±1.7 (6) 0.07

CD25 0.6±0.4 (5) 3.6±2.0 (8) 0.17

FOXP3 0.1±0.1 (5) 0.9±0.4 (8) 0.13 CTLA-4 3.5±1.0 (5) 9.6±1.5 (8) 0.02

CD2 84±^{2.4 (4)} ⁸⁶±^{2.0 (6)} ^0.61

CD8 24±3.1 (4) 24±6.6 (6) 0.61

CD20 6.8±2.4 (4) 5.7±0.8 (7) 0.41

CD68 16±2.4 (4) 15±2.7 (7) 0.79

CD83 13±4.6 (4) 13±4.5 (7) 0.93

Table 4.3: Immunohistological analyses of graft-infiltrates in POD 21 biopsies.Data are given as mean±SEM percent positive cells of the total number of cells in the focal infiltrates with the number of animals analysed. A trend is observed that T- cell markers associated with T-cell regulation, CD4, CD25, CTLA-4 and FOXP3 are lower in the ATG group, which is significant for CTLA-4 (p= 0.02, bolded value).

Composition of the inﬁltrates does not differ signiﬁcantly between the ATG group and the combined COS-COS/CsA group, for CD2, CD8, CD20, CD68 and CD83.

with respect to the number of inﬁltrating cells, as determined by interstitial inﬁltra- tion (i), according to Banff (data not shown), or CD2 staining (Fig. 4.3, A and B).

POD 21 kidney biopsies from animals of the COS-COS/CsA group were char- acterised by the expression of CTLA-4 (CD152) and FOXP3, mainly within focal perivascular inﬁltrates and less in diffuse interstitial inﬁltrates (Fig. 4.3, C and E).

Both CTLA-4 and FOXP3 are expressed on Tregs. Although CTLA-4 is also induced upon activation of cells, the expression of CTLA-4 and FOXP3 indicate that within focal infiltrates active downregulation of the immune response may take place. In POD 21 biopsies from animals in the ATG group, reduced numbers of CTLA-4⁺ or FOXP3⁺ cells were detectable (Fig. 4.3, D and F). The percent CTLA-4⁺ cells was significantly lower in the ATG group (3.5±1.0%) as compared to the combined COS- COS/CsA group (9.6±1.5%, Fig. 4.3G and Table 4.3; p = 0.0186, Mann-Whitney U test). Notably, lower percentages of other Treg markers (FOXP3, CD25 and CD4), were found in biopsies of ATG-treated animals, although not significant (Fig. 4.3H and Table 4.3). Percentages of CD2⁺, CD8⁺, CD20⁺, CD68⁺ or CD83⁺ cells were similar in both groups (Table 4.3). At euthanasia, all animals had about 23-30%

CTLA-4⁺ cells in the focal inﬁltrates in the kidney (data not shown). The number of FOXP3⁺ cells at euthanasia still reﬂected the differences in group means seen at POD 21, although the percent positive cells was higher at euthanasia compared to POD 21 (data not shown).

(13)

"#$%

%#&'()*+ ",-",- ".%'()*+

*

+)./0/123244.

5,67

%#&'()*+ ",-",-".%'()*+

+

+)./0/123244.

Figure 4.3:Expression of Treg markers CTLA-4 and FOXP3 in kidney biopsies. Kid- ney biopsies from POD 21 were stained for CD2, CTLA-4 (CD152) and FOXP3. Stain- ing was mainly detected in focal infiltrates. (A, B) Similar numbers of CD2⁺infiltrating cells were found in both groups. Percentage CTLA4⁺cells and percent FOXP3⁺ cells were scored blindly. Shown are representative examples of CTLA-4 staining (C and D) and of FOXP3 staining (E and F) of an animal from the COS group (survival 61 days; A, C, and E) and of an animal from the ATG group (survival 56 days; B, D, and F). Percentage positive cells are plotted for animals in the ATG group and for animals of the combined COS-COS/CsA group. When no focal infiltrates were detected, the biopsy was excluded from evaluation. (G, H) The percent CD152 expressing cells was significantly different between the two groups (p= 0.0025), but the percent FOXP3 expressing cells was not statistically significant between groups (p= 0.1274).

(14)

ATG pretreatment inﬂuences the antibody responses against the donor

Increased amounts of anti-donor IgG antibodies were found in animals of the ATG group, which are absent in all animals from the COS and COS/CsA groups (n=9) and did not develop donor-specific IgG alloantibodies. Three out of the five ATG-treated animals (survival 38, 56 and 97 days) developed a significantly higher level of IgG alloantibodies, ranging from 12% to 38% donor cells staining positive for recipient serum. One animal (survival 109 days) had only anti-donor antibodies of the IgM isotype. These results indicate that after ATG treatment, costimulation blockade and CsA cannot prevent formation of anti-donor antibodies.

Discussion

Transplantation of an organ from an MHC-mismatched donor evokes a strong allore- action in the recipient, in rodents as well as in primates. Without immunosuppression transplantation leads to acute tubular necrosis and beginning signs of acute cel- lular allograft rejection in non-human primates within six to seven days [9]. Current immunosuppressive drugs prevent acute rejection almost completely, but chronic rejection still poses a threat to the long-term function of the graft. The ultimate goal in transplantation research is therefore to tolerize a recipient to the alloantigens of the donor.

A long list of tolerance induction strategies have been tested in rodents, but most of these fail when they are applied in primates [18]. One reason may be that the rodent and human immune systems differ fundamentally [19]. Another reason is the different degree of immunological maturity between the rodent and primate immune systems due to the different exposure to pathogens that shape the immune repertoire during early life [2]. The mature human memory repertoire contains alloreactive T-cells [20] that form the major hurdle in the establishment of allospeciﬁc tolerance by costimulation blockade [2]. Conceptually, the removal of these cells with ATG should have a positive effect on costimulation blockade induced tolerance, but the current study shows that ATG treatment does not synergise with costimulation blockade.

We have demonstrated previously that antibody-mediated blockade of costimulatory molecules on antigen presenting cells (CD40 and CD86) suppresses the rejection of a kidney allograft in rhesus monkeys [5]. Combination with late CsA treatment even created long-term survivors who accept their kidney allograft for more than four years without further immunosuppression [11]. In this study, we show that this encouraging effect is not enhanced further when recipient monkeys are treated with ATG at the start of costimulation blockade treatment. In addition to ATG, methylprednisolone was given on POD -1, as integral part of the ﬁrst ATG administration, to alleviate the side effects. It may be possible that methylprednisolone is implicated in the observed negative effects on the graft in the ATG group. How- ever, methylprednisolone has a reported biological half-life of only 12-36 hr in humans and no T-cell depleting effects have been reported. It is therefore unlikely that

(15)

methylprednisolone is responsible for the observed effects and no treatment group was included to control for this possible effect. The CsA treatment was also different between the ATG group and the COS/CsA group. CsA treatment was maintained at high-dose after POD 70 in the ATG group, because all animals had already expe- rienced one or more rejection episodes. Only one animal in the ATG group was still alive, with ongoing rejection, when CsA was stopped after POD 98. It is therefore unlikely that the increased survival in the COS/CsA group is caused only by the prolonged low-dose CsA treatment until POD 126 in this group.

The current study shows that in conformity with literature data, ATG causes rapid partial depletion of CD4⁺ and CD8⁺ T-cells from the circulation [14]. The sequence of reappearing cells is ﬁrst CD8⁺ T-cells with a memory phenotype, and then followed by the much slower reappearance of CD4⁺ memory T-cells. Naive cells are hardly present, after ATG treatment. A possible explanation for the pre- dominance of memory cells could be homeostatic repopulation of the free space.

Mostly memory T-cells reappear, while naive cells would appear slower, as they are less capable of self-renewal [21–24]. It has been confirmed in mice that cells undergo- ing homeostatic proliferation cannot be rendered tolerant by costimulation blockade [25]. However, mechanisms such as increased proliferation of alloreactive cells after ATG treatment may also result in increased numbers of memory cells. Memory CD4⁺T-cells can provide efficient help to anti-donor CD8⁺ cells in the presence of costimulation blockade [26], leading to graft destruction. CD4⁺memory T-cells also can provide help for antibody isotype switching. This could explain the increased amounts of anti-donor IgG antibodies that we found in the ATG group (3 of 5 treated with ATG versus 0 of 9 treated without ATG). The presence of donor-specific alloantibodies correlates with graft failure [27]. The induction of donor-specific alloantibodies in ATG-treated animals is therefore unfavourable as compared to treatment with costimulation alone.

Our observation that reappearing T cells after ATG treatment are predominantly CD8⁺ contrasts with observations by Pearl et al. [28]. However, others have reported that, similar to our ﬁndings, CD8⁺cells are the dominating T-cell subset after peripheral T-cell depletion [15, 16].

The current study shows a substantially reduced expression of CTLA-4 (CD152) and FOXP3 in graft biopsies from ATG-treated animals at POD 21 (Fig. 4.3). Graft biopsies from animals treated only with anti-CD40 antibodies, which survive be- yond the treatment period [5], have high percentages of CTLA-4⁺infiltrating cells in the graft (22.5±4.6% of infiltrating cells) and two animals also have higher percentages of FOXP3⁺ cells in the graft than any of the ATG, COS or COS/CsA animals (data not shown). Reduced numbers of the Treg-specific marker FOXP3 and the down-regulatory molecule CTLA-4 in early biopsies in ATG-treated animals may be a possible explanation for the observed early rejections in this group.

We conclude that partial T-cell depletion with ATG does not have an additive effect to the graft survival observed with costimulation blockade with or without CsA.

Rather, we ﬁnd a disturbed balance of memory over naive CD4⁺ and CD8⁺ T-cell ratios and reduced numbers of CTLA-4⁺and FOXP3⁺cells in early graft biopsies.

(16)

Acknowledgements

The authors thank the veterinary staff and caretakers of the BPRC for their care for the animals. The authors also thank H. van Westbroek for help with preparing the graphical work, E. Remarque for help with the statistical analyses. We express our gratitude to Dr. R. Bontrop and Prof.Dr. F. Claas for critical review of the manuscript.

References

[1] Anderson CC, Matzinger P. Immunity or tolerance: opposite outcomes of mi- crochimerism from skin grafts. Nat Med. 2001;7(1):80–7.

[2] Adams AB, Williams MA, Jones TR, Shirasugi N, Durham MM, Kaech SM, et al.

Heterologous immunity provides a potent barrier to transplantation tolerance.

J Clin Invest. 2003;111(12):1887–95.

[3] Valujskikh A, Pantenburg B, Heeger PS. Primed allospeciﬁc T cells prevent the effects of costimulatory blockade on prolonged cardiac allograft survival in mice. Am J Transplant. 2002;2(6):501–9.

[4] Zhai Y, Meng L, Gao F, Busuttil RW, Kupiec-Weglinski JW. Allograft rejection by primed/memory CD8+ T cells is CD154 blockade resistant: therapeutic im- plications for sensitized transplant recipients. J Immunol. 2002;169(8):4667–73.

[5] Haanstra KG, Ringers J, Sick EA, Ramdien-Murli S, Kuhn EM, Boon L, et al.

Prevention of kidney allograft rejection using anti-CD40 and anti-CD86 in primates. Transplantation. 2003;75(5):637–43.

[6] Swanson SJ, Hale DA, Mannon RB, Kleiner DE, Cendales LC, Chamberlain CE, et al. Kidney transplantation with rabbit antithymocyte globulin induction and sirolimus monotherapy. Lancet. 2002;360(9346):1662–4.

[7] Bontrop RE, Otting N, Slierendregt BL, Lanchbury JS. Evolution of major his- tocompatibility complex polymorphisms and T-cell receptor diversity in primates. Immunol Rev. 1995;143:33–62.

[8] Doxiadis GG, Otting N, de Groot NG, Noort R, Bontrop RE. Unprecedented polymorphism of Mhc-DRB region conﬁgurations in rhesus macaques. J Im- munol. 2000;164(6):3193–9.

[9] Ossevoort MA, Ringers J, Kuhn EM, Boon L, Lorre K, van den Hout Y, et al. Pre- vention of renal allograft rejection in primates by blocking the B7/CD28 path- way. Transplantation. 1999;68(7):1010–8.

[10] Racusen LC, Solez K, Colvin RB, Bonsib SM, Castro MC, Cavallo T, et al.

The Banff 97 working classiﬁcation of renal allograft pathology. Kidney Int.

1999;55(2):713–23.

[11] Haanstra KG, Sick EA, Ringers J, Wubben JA, Kuhn EM, Boon L, et al. Costim- ulation Blockade followed by a 12-Week Period of Cyclosporine A Facilitates Prolonged Drug-Free Survival of Rhesus Monkey Kidney Allografts. Trans- plantation. 2005;79(11):1623–6.

[12] Pitcher CJ, Hagen SI, Walker JM, Lum R, Mitchell BL, Maino VC, et al. De-

(17)

velopment and homeostasis of T cell memory in rhesus macaque. J Immunol.

2002;168(1):29–43.

[13] R-Development-Core-Team. R: a language and environment for statistical com- puting; 2004.

[14] Preville X, Flacher M, LeMauff B, Beauchard S, Davelu P, Tiollier J, et al. Mech- anisms involved in antithymocyte globulin immunosuppressive activity in a nonhuman primate model. Transplantation. 2001;71(3):460–8.

[15] Heitger A, Winklehner P, Obexer P, Eder J, Zelle-Rieser C, Kropshofer G, et al.

Defective T-helper cell function after T-cell-depleting therapy affecting naive and memory populations. Blood. 2002;99(11):4053–62.

[16] Novitzky N, Davison GM, Hale G, Waldmann H. Immune reconstitution at 6 months following T-cell depleted hematopoietic stem cell transplantation is predictive for treatment outcome. Transplantation. 2002;74(11):1551–9.

[17] Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature.

1999;401(6754):708–12.

[18] Sachs DH. Tolerance: of mice and men. J Clin Invest. 2003;111(12):1819–21.

[19] Mestas J, Hughes CC. Of mice and not men: differences between mouse and human immunology. J Immunol. 2004;172(5):2731–8.

[20] Heeger PS, Greenspan NS, Kuhlenschmidt S, Dejelo C, Hricik DE, Schulak JA, et al. Pretransplant frequency of donor-speciﬁc, IFN-gamma-producing lymphocytes is a manifestation of immunologic memory and correlates with the risk of posttransplant rejection episodes. J Immunol. 1999;163(4):2267–75.

[21] Almeida AR, Borghans JA, Freitas AA. T cell homeostasis: thymus regeneration and peripheral T cell restoration in mice with a reduced fraction of competent precursors. J Exp Med. 2001;194(5):591–9.

[22] Ge Q, Hu H, Eisen HN, Chen J. Different contributions of thymopoiesis and homeostasis-driven proliferation to the reconstitution of naive and memory T cell compartments. Proc Natl Acad Sci U S A. 2002;99(5):2989–94.

[23] Neujahr DC, Chen C, Huang X, Markmann JF, Cobbold S, Waldmann H, et al.

Accelerated memory cell homeostasis during T cell depletion and approaches to overcome it. J Immunol. 2006;176(8):4632–9.

[24] Tanchot C, Le Campion A, Martin B, Leaument S, Dautigny N, Lucas B. Con- version of naive T cells to a memory-like phenotype in lymphopenic hosts is not related to a homeostatic mechanism that ﬁlls the peripheral naive T cell pool. J Immunol. 2002;168(10):5042–6.

[25] Wu Z, Bensinger SJ, Zhang J, Chen C, Yuan X, Huang X, et al. Homeostatic proliferation is a barrier to transplantation tolerance. Nat Med. 2004;10(1):87–

92.

[26] Chen Y, Heeger PS, Valujskikh A. In vivo helper functions of alloreactive memory CD4+ T cells remain intact despite donor-speciﬁc transfusion and anti-CD40 ligand therapy. J Immunol. 2004;172(9):5456–66.

[27] Piazza A, Poggi E, Borrelli L, Servetti S, Monaco PI, Buonomo O, et al. Impact of donor-speciﬁc antibodies on chronic rejection occurrence and graft loss in renal transplantation: posttransplant analysis using ﬂow cytometric techniques.

(18)

Transplantation. 2001;71(8):1106–12.

[28] Pearl JP, Parris J, Hale DA, Hoffmann SC, Bernstein WB, McCoy KL, et al.

Immunocompetent T-Cells with a Memory-Like Phenotype are the Dominant Cell Type Following Antibody-Mediated T-Cell Depletion. Am J Transplant.

2005;5(3):465–74.

(19)