CTLA4Ig ALONE OR IN COMBINATION WITH LOW-DOSE CYCLOSPORINE FAILS TO REVERSE ACUTE REJECTION OF RENAL ALLOGRAFT IN THE RAT

Primary immunologic rejection of vascularized organs depends on the interaction between T cell receptors (TCR*) and histocompatibility alloantigens on the surface of antigen-presenting cells (1, 2). This engagement delivers specific intracellular signals that, through the expression and synthesis of the cytokine/growth factor interleukin-2 and its receptor, induces a resting T cell to exit G0 phase and proliferate (1). Evidence is available, however, that TCR stimulation alone is not enough to sustain full activation of T cell clones and that additional signals are required (3, 4). Thus, in vitro studies have documented that, in the absence of a second signal, TCR activation results in a state of anergy of T cells, which become unresponsive to stimulation for several weeks (5, 6). CD28, a member of the immunoglobulin supergene family identified on the T cell surface, modulates a TCR-independent signal transduction pathway and synergizes with the latter in promoting T cell proliferation (7-9). More recently, a molecule, CTLA4, that is only expressed on activated T cells and highly homologous to CD28 has been identified (9-11). Both molecules contribute to T cell activation upon binding to their specific ligands on antigen-presenting cells, i.e., B7-1 and B7-2 (9). A soluble chimeric protein (CTLA4Ig), consisting of the binding domain of CTLA4 fused to a human IgG tail, has been obtained (10) that binds with high efficiency to mouse and rat B7-1 and B7-2 ligands (12,13). Because the recombinant protein inhibits the interaction of CD28 with its ligands and thus impairs the function of those T lymphocytes activated by alloantigens, administration of CTLA4Ig to experimental animals has been attempted as a specific strategy to prevent graft rejection. Indeed a short course of CTLA4Ig treatment induced indefinite survival of human islets xenografted in mice (12) and prolonged cardiac or renal allograft survival in rats (13-15). These findings would indicate that costimulatory signals delivered through CD28 to antigen-specific T cell proliferation are instrumental to cell proliferation and subsequent graft rejection. Recent experiments also show that CTLA4Ig given in a single dose could be substantially more effective as a means to prevent allograft rejection if its administration was delayed until day 2 after transplantation, rather than if it was given at the time of surgery, as in the previous studies (14, 16). This finding was interpreted as an indication that T cells were more susceptible to costimulatory blockade when their TCR had been exposed to a specific antigen for a limited period of time. No one has studied so far whether the chimeric protein actually down-regulates T cell signals that help self-sustain an already active rejection process. We sought to address this issue by studying whether CTLA4Ig is capable of limiting the consequences of an already ongoing acute allograft rejection process. MHC-incompatible Brown-Norway (RT1n) rat kidneys were orthotopically transplanted into male Lewis (RT11) rat recipients by microsurgery (17). After transplantation, all animals were kept in individual metabolic cages and 24-hr urine samples were collected for measurement of urine output. Blood samples were collected from the tail vein every day for the first week after transplantation, at the time of development of oligoanuria, or at different follow-up times for measurement of serum creatinine concentration as an index of renal function. Survival time of the renal graft was defined as the time from transplantation to the development of anuria. Three groups of rats that had received transplants were administered CTLA4Ig (Bristol-Myers Squibb Pharmaceutical Research Institute, Seattle, WA) intraperitoneally at the daily dose of 0.2 mg (n=3), 0.5 mg (n=6), or 1 mg (n=3), beginning 4-5 days after transplantation, when urine output declined ≥25% as compared with day 1 or serum creatinine increased to ≥1.5 mg/dl. As control, additional rats received intraperitoneal injections of an irrelevant IgG (0.5 mg/day, n=5) starting as above. Both CTLA4Ig treatment and non-specific IgG treatment were continued until the animals became anuric. In another group of animals (n=3), CTLA4Ig (0.2 mg/day i.p.) was administered for 7 days, beginning 2 hr before kidney transplantation. Moreover, to determine the functional status of alloreactive T cells of transplanted animals on CTLA4Ig, peripheral blood lymphocytes (PBLs) from Lewis transplanted rats were tested for their ability to proliferate to Brown-Norway (BN) or third-party cells from Wistar-Furth (WF) rats in the standard one-way mixed lymphocyte reaction (MLR) assay. PBLs were isolated from peripheral blood of responder Lewis and stimulator BN or third-party WF rats by Ficoll-Hypaque density gradient centrifugation, and were suspended in RPMI 1640 medium (Gibco, Grand Island, NY) supplemented with L-glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 μg/ml), and 10% fetal calf serum. The MLRs were set up by culturing an equal number (2×105) of Lewis responder cells and irradiated (3000 rads) BN or third-party WF stimulator cells in 96-well U-bottom plates at 37°C with 5% CO2 for 4 days. Proliferation was assayed by [3H]-thymidine uptake and relative response calculation, as described previously (18). MLRs were performed in triplicate for each individual animal at the time they became anuric or later on (at day 30 after transplantation) in those who had long-term graft survival. Since it has been shown previously that combining treatment with CTLA4Ig and low-dose cyclosporine (CsA) from the day of transplantation results in a state of unresponsiveness to kidney allograft (15), we also evaluated whether the combined regimen of chimeric protein and CsA may synergize in the treatment of acute graft rejection. Lewis rats (n=5) were transplanted with BN kidneys and given CTLA4Ig (0.5 mg/day i.p.) in combination with CsA (2 mg/kg/day p.o.) starting 4-5 days after transplant. The results of the present study confirm that CTLA4Ig given for 7 days at the dose of 0.2 mg/day as the sole immunosuppressive therapy, beginning at the time of transplantation, prevented graft rejection and considerably prolonged rat renal allograft survival (>35 days, Table 1). As shown in Table 1, control Lewis recipients given a nonspecific immunoglobulin daily starting at day 4-5 after transplantation failed to reverse acute allograft rejection. Similarly daily intraperitoneal CTLA4Ig (0.2 or 0.5 mg/day) started at the same time point did not prolong graft survival, all of which failed within 7-12 days after transplantation. Daily CTLA4Ig failed to limit the consequence of acute rejection even at the highest dose used (1mg/day) and grafts inexorably lost their function within 10 days. Rats which started CTLA4Ig treatment on the day of surgery maintained a normal renal function, measured as serum creatinine concentration, for all 30 days after transplantation (0.61±0.06 mg/dl). By contrast, renal function was low at the beginning of treatment with irrelevant IgG (1.7±0.2 mg/dl) and deteriorated further thereafter. In the other transplanted, CTLA4Ig-treated rats, serum creatinine values increased from 1.7±0.3, 1.8±0.2, and 1.7±0.1 mg/dl at day 4-5 after transplantation to 3.21±16, 3.15±0.21, and 3.08±0.12 mg/dl, independently of whether animals were given 0.2, 0.5, or 1 mg/day, respectively, of the soluble chimeric protein. Proliferative response of PBLs from animals given CTLA4Ig starting at day 4-5 after transplantation to BN (CTLA4Ig 0.2 mg: 14,570±2,420 cpm; CTLA4Ig 0.5 mg: 15,219±1,585 cpm; CTLA4Ig 1 mg: 15,379±2,640 cpm) or third-party WF lymphocytes (CTLA4Ig 0.2 mg: 14,890±2,027 cpm; CTLA4Ig 0.5 mg: 15,700±1,850 cpm; CTLA4Ig 1mg: 14,850±2,300 cpm) was equivalent to that of control rats (16,020±2,540 cpm) (Fig. 1). By contrast, the MLR response of PBLs from longterm surviving, CTLA4Ig-treated rats to BN stimulator cells was 4165±530 cpm; the MLR response to third-party WF cells was 15,530±1,480 cpm. Combining treatment of CTLA4Ig and low-dose CsA also did not reverse acute allograft rejection in our model. In animals given CTLA4Ig plus CsA beginning on day 4-5, kidney grafts failed within 11 days after transplantation (Table 1). At that time, the serum creatinine concentration was 3.02±0.12 mg/dl. Why CTLA4Ig, when started 4-5 days after transplantation, did not reverse graft rejection remains speculative at this stage. It is possible that the costimulatory B7/CD28 pathway is not required to sustain the effector arm of an established allograft rejection response. Alternatively, humoral factors from T cells already infiltrating the graft may maintain subsequent cell proliferation that can no longer be influenced by blocking B7-1/CD28 interaction on the newly generated T lymphocytes. This would allow the rejection process to progress quite independently of the modulatory action of the costimulatory pathway. One cannot exclude, however, that, while CTLA4Ig is ineffective in reverting an advanced graft rejection process, the same soluble chimeric protein might be of value in the very early phases of this event. It has been reported that a single dose of CTLA4Ig after kidney transplantation allowed longterm graft survival in an orthotopic rat renal allograft model (19). This, however, requires further study of very early markers of acute graft rejection, still an elusive goal in transplant research. Acknowledgments. We thank Dr. Ingegerd Helstrom (Bristol-Myers Squibb Pharmaceutical Research Institute, Seattle, WA) and Dr. De Vita (Bristol-Myers Squibb, Italy) for kindly providing the CTLA4Ig molecule.Figure 1: Proliferative response of lymphocytes from CTLA4Ig-treated or nonspecific immunoglobulin-treated transplanted Lewis rats challenged with BN cells in the standard one-way MLR. Values are given as percent proliferative response as compared with that of MLR of lymphocytes from untreated control Lewis rats against BN cells (100%). Data are mean values of individual animals ± SD. * P<0.01 vs. all other groups. Background count of MLR in the absence of alloantigen is 207±25 cpm.