Repeated Injections of IL-2 Break Renal Allograft Tolerance Induced via Mixed Hematopoietic Chimerism in Monkeys

Abstract

Tolerance of allografts achieved in mice via stable mixed hematopoietic chimerism relies essentially on continuous elimination of developing alloreactive T cells in the thymus (central deletion). Conversely, while only transient mixed chimerism is observed in nonhuman primates and patients, it is sufficient to ensure tolerance of kidney allografts. In this setting, it is likely that tolerance depends on peripheral regulatory mechanisms rather than thymic deletion. This implies that, in primates, upsetting the balance between inflammatory and regulatory alloimmunity could abolish tolerance and trigger the rejection of previously accepted renal allografts. In this study, six monkeys that were treated with a mixed chimerism protocol and had accepted a kidney allograft for periods of 1-10 years after withdrawal of immunosuppression received subcutaneous injections of IL-2 cytokine (0.6-3 × 10(6) IU/m(2) ). This resulted in rapid rejection of previously tolerated renal transplants and was associated with an expansion and reactivation of alloreactive pro-inflammatory memory T cells in the host's lymphoid organs and in the graft. This phenomenon was prevented by anti-CD8 antibody treatment. Finally, this process was reversible in that cessation of IL-2 administration aborted the rejection process and restored normal kidney graft function.

Keywords: animal models: nonhuman primate; basic (laboratory) research/science; bone marrow/hematopoietic stem cell transplantation; immunosuppression/immune modulation; kidney (allograft) function/dysfunction; kidney transplantation/nephrology; tolerance: chimerism; tolerance: mechanisms; translational research/science.

Figure 1. In vitro exposure to exogenous IL-2 restores alloresponses by T cells isolated from tolerant monkeys

Peripheral blood T cells from four tolerant monkeys were collected either pretransplantation or 1–10 years after acceptance of kidney allografts and cultured for 40 h in vitro with medium or irradiated stimulator cells derived from the recipient, the donor, or a third-party monkey. The posttransplant mixed lymphocyte reactions were conducted either in the absence of IL-2 or with serial concentrations of IL-2 (ranging from 2.5–25 000 IU/mL) as indicated on the X-axis. The frequencies of activated T cells producing γIFN were measured using ELISPOT. The results are expressed as numbers of γIFN spots per million T cells ± SD and are representative of four monkeys tested individually. p values were calculated by comparing the results obtained with and without IL-2. *Corresponds to p < 0.05 and **corresponds to p < 0.01.

Figure 2. IL-2 injections induce acute cellular rejection of kidney allografts in tolerant monkeys

Four tolerant monkeys, which had displayed stable Cr levels for over 2–10 years after withdrawal of immunosuppression, received multiple subcutaneous injections of IL-2, as indicated in (A). The horizontal black bars (above each graph) indicate the doses of IL-2 administered (ranging from 0.1 to 3 × 10⁶ IU/m²) and the duration of treatment for monkeys M6007, M2108, M8907, and M2800. Serum Cr levels (mg/dl) were recorded at different time points posttransplantation (before or after IL-2 injections). (B–E) Histological examination of the kidney allograft biopsies from monkey M2108 collected pre-IL-2 injection (B) revealed normal glomeruli, no endothelialitis in arteries and a few cell aggregates containing essentially CD3⁺CD4⁺ T cells, a few CD3⁺CD8⁺ T cells and virtually no CD20⁺ B cells (data not shown). Renal graft tissues examined after IL-2 injections (C–E) show acute cellular rejection (ACR, grade 2), with diffuse inflammation (C), renal artery endothelialitis (D, arrows), and vasculopathy (E). Histology results of kidney allografts from monkey M2800, M6007, and M8907 are shown in Figure S4. Tx, transplantation.

Figure 3. Frequencies of leukocyte subsets after IL-2 injections

Figure 3 shows the frequencies of different leukocyte subsets (A and B, number of cells/mm³ of blood) and their proliferation status (C, Ki67 expression) measured by FACS at different time points in the peripheral blood of M2108 prior to and during the course of IL-2 treatment. The horizontal black bars indicate the duration of the IL-2 treatment (1 × 10⁶ IU/m²). (A) The frequencies of CD4⁺CD25^highFOXP3⁻ T cells (effector T cells) and CD4⁺CD25^highFOXP3⁺ (regulatory T cells). (B) The frequencies of CD4⁺ and CD8⁺ T cells expressing either a naïve (CD95⁻) or memory (CD95⁺) phenotype, and the number of NK cells. (C) The percentages of Ki67-expressing cells among CD4⁺CD25^highFOXP3⁺ regulatory T cells, CD8⁺ T cells, and NK cells. The results are representative of six monkeys tested individually (see additional data in Figure S4). Tx, transplantation.

Figure 4. Further characterization of leukocyte subsets present in explanted kidney allografts and peripheral blood

Histological examination and FACS were used to assess the presence and phenotype of different leukocyte subsets isolated from kidney allografts and peripheral blood during IL-2 treatment. Figure 4 shows the presence and distribution of CD3⁺, CD4⁺, and CD8⁺ T cells as well as cells expressing FOXP3 within renal tissue allografts undergoing acute cellular rejection (ACR2) and displaying severe tubulitis, tubular necrosis, interstitial hemorrhage, and endothelialitis (not shown). (A) (PAS) shows interstitial inflammation; (B) shows rare FOXP3 staining cells; (C) shows interstitial CD8-positive T cells; (D) shows positive interstitial CD3 T cells; and (E) shows CD4-positive interstitial T cells. The first two panels of (F) (graft) and (G) (peripheral blood) show FACS profiles describing the presence of different T cell subsets, including CD3⁺ T cells, naïve (CD95⁻CD28⁺), and memory (central TMEM: CD95⁺CD28⁺ and effector TMEM: CD95⁺CD28⁻). The next three panels show the frequencies of regulatory T cells, CD4 gated, Tregs: CD25⁺FOXP3⁺, either resting (CD45RA⁺) or activated (CD45RA⁻) and their proliferative status examined through their coexpression of FOXP3 and Ki67 markers. These figures were obtained from the renal graft of M8907 (explanted after 14 days of IL-2 injections).

Figure 5. Restoration of alloresponses by T cells following IL-2 treatment

After a 14-day course of IL-2 treatment, T cells from the peripheral blood (white bars), spleen (shaded bars), and explanted kidney allografts (solid bars) were isolated from two monkeys and tested for their ability to mount a direct alloresponse. The frequencies of T cells producing γIFN when incubated for 48h with allogeneic donor or two distinct third party allostimulators (3rd party.1 and 3rd party.2) were investigated by ELISPOT. In addition, pro-inflammatory responses by T cells exposed to control self-APCs or medium were measured. The results are expressed as numbers of γIFN spots per million T cells ± SD and are representative of two monkeys (M6007 and M8907) tested individually.

Figure 6. Reversal of rejection after cessation of IL-2 treatment

IL-2 treatment was discontinued in three monkeys shortly after serum Cr levels had risen to 2–3 mg/dl (M4109, M2800, and M6007). (A) Serum Cr levels (mg/ml) for M4109 measured at different time points before, during and after IL-2 administration (10⁶ IU/m²). After recovery of a normal Cr level, this animal received another session of 14 day-course of IL-2 treatment, which showed the same pattern of serum Cr levels. The horizontal bars indicate the duration of the IL-2 treatment. (B) (day 525), (C) (day 553), and (D) (day 615) show histological features of renal graft biopsies collected from M4109. (B) Before IL-2 treatment without rejection. (C) During IL-2 treatment with acute cellular rejection (ACR grade 1) (infiltrate with an arrow). (D) 2 months after cessation of IL-2 treatment without rejection.

Figure 7. Effects of anti-CD8 mAb pretreatment on IL-2–mediated rejection

Two tolerant monkeys (A, M4109; B, M5710) were injected with an anti-CD8 depleting antibody (cM-T807) prior to IL-2 administration and then twice a week during the course of IL-2 treatment (10⁶ IU/m²). Of note, M4109 was studied in the previous experiment. The frequencies of CD4⁺ (solid triangles) and CD8⁺ (solid squares) memory T cells (TMEM) as well as the Cr levels (dotted blue lines) were measured at different time points after injection of anti-CD8 mAbs. The arrow heads indicate the times at which renal biopsies were obtained. (C–F) Histological features of renal graft biopsies. (C) (M4109) No acute cellular rejection. (D) No C4d staining of the peritubular capillaries. (E) (M5710) Acute rejection (ACR1) and acute humoral rejection (AHR) with acute glomerulitis (arrow). (F) Peritubular capillary C4d staining in M5710 (arrows). Tx, transplantation.

Figure 8. Possible mechanisms involved in IL-2–mediated breakdown of transplant tolerance

In tolerant monkeys, IL-2 reactivates undeleted pro-inflammatory alloreactive CD4⁺ (TH1) and CD8⁺ (CT1) memory T cells, which react in a direct fashion to allogeneic major histocompatibility complex (MHC) molecules displayed on graft donor cells (left and middle panels). In addition, IL-2 is likely to stimulate T cells (presumably CD4⁺ T cells) recognizing donor antigens through indirect allorecognition, which could provide help for the activation/differentiation of memory B cells into alloantibody- producing plasma cells (right panel).