Lenard M. B. Vaessen

A broad and strong humoral immune response to donor HLA after implantation of cryopreserved human heart valve allografts

Cryopreserved human heart valves are used for valve replacement in patients with congenital or acquired heart disease. Although no blood group or human leukocyte antigens (HLA) matching is performed and no immunosuppression is administered, the clinical results are relatively good. After valve replacement, the majority of the patients develop HLA antibodies, whereas a smaller group of patients shows valve-related events at the long term after right ventricular outflow tract reconstruction. Therefore, we hypothesized that not the mere presence, but rather the titers of antidonor HLA antibodies may be related to valve allograft failure. The presence and specificity of HLA class I antibodies were determined by complement-dependent microlymphocytotoxicity (CDC) test in longitudinally taken peripheral blood samples of 35 valve allograft recipients. In eight patients with an antibody response specific against donor-HLA class I, the titers were measured by this CDC method after stepwise dilution of the plasma. Panel reactive antibodies of more than 10% were found in 31 of 35 (89%) valve allograft recipients. From these 31 patients, 24 (77%) developed donor-specific HLA class I antibodies. All eight selected patients had detectable donor-specific antibody titers, ranging from 1:2 to 1:8,000. Two donor valve recipients before retransplantation had (donor-specific) HLA antibodies and showed high titers of 1:256 and 1:8,000 shortly after the second allograft valve replacement, which was associated with an early graft failure in the latter patient. We conclude that transplantation of cryopreserved human heart valve allografts leads to a broad and strong humoral response, which is probably the result of a lack of immunosuppressive therapy after valve transplantation. Patients receiving a second or following valve allograft appeared to be sensitized and developed early and high allo-antibody titers after second valve allograft implantation. Valve failure was diagnosed in a patient with extremely high titers. These findings suggest that preoperative cross-matching may identify patients with high donor-specific HLA antibody titers and may reduce the risk for early recurrent graft failure.

Differential avidity and cyclosporine sensitivity of committed donor-specific graft-infiltrating cytotoxic T cells and their precursors : relevance for clinical cardiac graft rejection

We have used limiting dilution analysis to study the qualitative and quantitative differences between graft-infiltrating cytotoxic T cell populations propagated from endomyocardial biopsies of heart transplant patients who experienced one or more acute rejection episodes and patients who never showed signs of rejection. Limiting dilution cultures were stimulated with autologous or donor cells both in the absence or in presence of cyclosporine and of CD8 in the cytotoxic phase. Almost all antigen-primed, committed cytotoxic T cells (cCTL) present in the graft of patients with rejections were CsA resistant. In contrast, in most patients of the nonrejector group, a substantial part of the cCTL could be inhibited by CsA. The CTL precursors (pCTL) in both groups were predominantly CsA sensitive. Addition of CD8 mAb during the cytotoxicity phase of the limiting dilution analysis was used to differentiate between CTL populations with high avidity for donor antigens and populations with low avidity. The predominant subpopulation in the graft of rejectors was a CsA-resistant cCTL with high avidity, while in the graft of most nonrejectors, cCTL with low avidity dominated. In most rejectors, CD8 mAb had only a minor influence on the pCTL frequency estimates, and thus on T cells with high avidity. CsA-sensitive pCTL with high avidity might represent an intermediate stage between the naive pCTL and mature, functional, CsA-insensitive cCTL with high avidity for donor antigens.

Donor-specific cytokine production by graft-infiltrating lymphocytes induces and maintains graft vascular disease in human cardiac allografts.

BACKGROUND The development of graft vascular disease (GVD) in the allograft is a major impediment for long-term survival of heart transplant recipients. GVD may be mediated by cellular processes, in response to the transplanted heart, and regulated by cytokines. METHODS We studied donor-specific cytokine production patterns in graft-infiltrating lymphocyte cultures propagated from endomyocardial biopsies. The biopsies were derived from patients with and without signs of GVD, as diagnosed by angiography at 1 year after heart transplantation. RESULTS In the first year after transplantation, significantly more T-helper (Th) 1 cytokines (interleukin [IL]-2: P=0.04; interferon-gamma: P=0.01), but not Th2 (IL-4 and IL-6) cytokines, were produced by cultures of patients with GVD compared with patients without GVD. Thereafter, the Th1 cytokine levels in patients with GVD normalized to the levels of patients without GVD. Detectable levels of IL-6 were produced significantly more often (P=0.009) by cultures obtained more than 1 year after transplantation from patients with GVD. CONCLUSIONS The results suggest that high levels of Th1 cytokines produced by graft-infiltrating lymphocytes early after transplantation may be responsible for activation of vascular endothelium, leading to the migration and proliferation of smooth muscle cells that is characteristic of GVD. IL-6, produced later after transplantation, continues this process by promoting smooth muscle cell proliferation.

Calcineurin Inhibitor Withdrawal in Stable Kidney Transplant Patients Decreases the Donor-specific Cytotoxic T Lymphocyte Precursor Frequency

Background. In a prospective study, calcineurin inhibitors (CNI) were withdrawn in patients two years after kidney transplantation. We questioned whether stopping CNI had an effect on the donor-specific reactivity, as CNI might hinder immune responses leading to graft acceptance. Methods. We measured the donor-specific cytotoxic T lymphocyte (CTL) precursor frequency (CTLpf) in 54 patients before and after withdrawal of CNI. In addition, the T-cell reactivity of PBMC to donor and third-party antigens was tested in MLR, and in IFNγ-Elispot. Reactivity to tetanus toxoid (TET) was studied as well. Results. Donor-specific CTLpf significantly decreased after CNI withdrawal (P=0.0001). In contrast, no difference was observed in third-party reactive CTLpf, donor and third-party reactive MLR and IFNγ-Elispot. Proliferative responses and the number of IFNγ-producing cells to TET also decreased after CNI withdrawal. The decrease in CTLpf correlated with the time between the two blood samples (before and after stopping CNI, P=0.05). This decrease was caused by stopping CNI, because there was no correlation between CTLpf and the duration of the CNI treatment after transplantation. Moreover, the percentage of regulatory T cells in the peripheral blood increased after CNI withdrawal. Conclusions. We report here that after withdrawal of CNI the donor-specific CTLpf decreases. We hypothesize that CNI suppress regulatory mechanisms that have the potential to down-regulate donor-specific CTL responses and reactivity to TET.

The frequency of interferon-gproducing cells reflects alloreactivity against minor histocompatibility antigens

Background. In human leukocyte antigen (HLA)-identical living-related renal transplant recipients, no donor-specific alloreactivity can be detected in regular tests (mixed lymphocyte culture, helper T-lymphocyte precursor frequencies, cytotoxic T-lymphocyte precursor frequencies) to identify patients responding to minor histocompatibility antigens (mHag). We questioned whether a more sensitive method like the Elispot-assay could be more appropriate. Methods and Results. Peripheral blood mononuclear cells (PBMC) from 16 HLA-identical living-related kidney transplant patients 3 months (range, 2 weeks to 5 months) after transplantation were tested for the frequency of interferon (IFN)-&ggr; producing cells by the Elispot-assay. PBMC from the recipient were stimulated with irradiated donor PBMC and corrected for backward stimulation. Donor-specific IFN-&ggr; producing cells (pc) in the range of 5 to 115 per million PBMC (median, 30 per million PBMC) were found. To evaluate the frequency of spot forming cells in time, PBMC from six patients who received an HLA-identical renal transplant were stimulated with irradiated donor PBMC before, approximately 3 months after, and 12 months after transplantation. Four patients who received an HLA-mismatched living-related kidney served as positive control. In the HLA-identical group, frequencies in the range of 0 to 10 IFN-&ggr; pc per million PBMC were found before transplantation, 0 to 30 per million PBMC 3 months after transplantation, and 0 to 45 per million PBMC 12 months after transplantation. In the HLA-mismatched group, significantly higher numbers were found: 10 to 480 IFN-&ggr; pc per million PBMC before, 20 to 360 per million PBMC at 3 months, and 30 to 590 per million PBMC 12 months after transplantation. Conclusion. Under immunosuppressive therapy, IFN-&ggr; pc specific for donor mHag can be found after HLA-identical living-related renal transplantation.

Peripheral monitoring of direct and indirect alloantigen presentation pathways in clinical heart transplant recipients

It has been reported that the response to alloantigens presented by the direct and indirect pathway may be of differential relevance after human kidney transplantation. Accordingly, we monitored these routes in peripheral blood mononuclear cells (PBMC) of heart transplant patients from before transplantation and up to 2 years thereafter in an attempt to find a correlation with the clinical status of the patients. Both before and after transplantation, comparable proportions of PBMC samples reacted in mixed lymphocyte culture to nondepleted donor spleen cells (direct route), but never to donor cells depleted for antigen-presenting cells (indirect route). In contrast, the latter route could easily be activated by a nominal antigen and persisted after transplantation, although the proportion of PBMC samples responding was significantly suppressed, irrespective of the occurrence of rejection. Consequently, complete removal of antigen-presenting cells from the stimulator population in a mixed lymphocyte culture with PBMC as responder is not a suitable tool for measuring indirect presentation of alloantigens, and therefore not relevant for monitoring the immunological status of heart transplant recipients.

The Application of Human Monoclonal Antibodies for Monitoring Donor Derived Soluble HLA Class I Molecules in the Serum of Heart Transplant Recipients

Increased levels of both donor and recipient derived HLA molecules can be found in serum and plasma of transplanted patients during rejection. Recent data suggest that levels of donor specific soluble HLA Class I (sHLA-1) correlate better with graft rejection than total sHLA Class I [1, 2]. Therefore, quantification of donor specific soluble counterparts of HLA Class I in the serum of the recipient may be a new way for non-invasive monitoring of rejection after organ transplantation. Up to now, only a limited number of mouse monoclonal antibodies (alpha HLA-A2, and alpha HLA-B7) has been used in enzyme linked immunosorbent assays (ELISAs) to detect donor specific HLA molecules in the plasma of transplant recipients. To monitor other donor-recipient combinations, we tested some of our HLA Class I specific human monoclonal antibodies, routinely used in complement dependent cytotoxicity, for their suitability in ELISA based assays. In the present model system, we used alpha HLA-A9 (BvK5C4) or alpha HLA-A3 (OK2F3) hybridoma-supernatant to set up a sHLA-A9 and sHLA-A3 specific ELISA. In a pilot study we show that these assays were sensitive enough to detect an increase of donor specific sHLA-I during rejection in the plasma of two heart transplant recipients. Use of a large set of human hybridomas will enable monitoring most recipient/donor combinations in the near future.

Down-regulated donor-specific T-cell reactivity during successful tapering of immunosuppression after kidney transplantation

Stable cadaveric renal transplant patients were routinely converted from cyclosporin A (CsA) to either azathioprine (AZA) or mycophenolate mofetil (MMF) 1 year after transplantation to reduce the side effects of long‐term immunosuppressive therapy. Thereafter, the AZA and MMF dose was gradually tapered to 50% at 2 years after transplantation. We questioned whether a reduction of immunosuppressive treatment results in a rise of donor‐specific T‐cell reactivity. Before transplantation (no immunosuppression), 1 year (high dose immunosuppression) and 2 years (low dose immunosuppression) after transplantation, the T‐cell reactivity of peripheral blood mononuclear cells (PBMC) against donor and third‐party spleen cells was tested in mixed lymphocyte cultures (MLC) and against tetanus toxoid (TET) to test the general immune response. We also measured the frequency of donor and third‐party reactive helper (HTLpf) and cytotoxic (CTLpf) T‐lymphocyte precursors in a limiting dilution assay. Donor‐specific responses, calculated by relative responses (RR = donor/third‐party reactivity), were determined. Comparing responses after transplantation during high dose immunosuppression with responses before transplantation (no immmunosuppression), the donor‐specific MLC‐RR (P = 0·04), HTLp‐RR (P = 0·04) and CTLp‐RR (P = 0·09) decreased, while the TET‐reactivity did not change. Comparing the responses during low dose with high dose immunosuppression, no donor‐ specific differences were found in the MLC‐RR, HTLp‐RR and CTLp‐RR, although TET‐reactivity increased considerably (P = 0·0005). We observed a reduction in donor‐specific T‐cell reactivity in stable patients after renal transplantation during in vivo high dose immunosuppression. Tapering of the immunosuppressive load had no rebound effect on the donor‐specific reactivity, while it allowed recovery of the response to nominal antigens.

A longitudinal study of TGF‐β1 protein levels in renal allograft recipients converted from CsA to MMF or AZA

Cyclosporine (CsA) is thought to enhance transforming growth factor (TGF)‐β1 production in vitro and in vivo and this may have a negative effect on long‐term graft survival. Therefore, we studied TGF‐β1 plasma levels in 30 patients before kidney transplantation, after transplantation during CsA treatment and after conversion from CsA to azathioprine (AZA) or mycophenolate mofetil (MMF). We questioned whether TGF‐β1 plasma levels would decrease after the discontinuation of CsA and whether the TGF‐β1 plasma levels did correlate with CsA trough levels and kidney function, measured by serum creatinine levels. TGF‐β1 plasma levels measured 1 yr after transplantation were lower compared to levels measured before transplantation, however not significantly (p=0.08). After conversion from CsA to MMF or AZA, a slight increase was observed in some patients, but in the total group TGF‐β1 levels remained unaffected. No correlation was found between the TGF‐β1 levels and CsA trough levels nor with creatinine levels. In conclusion, we did not observe higher TGF‐β1 plasma levels in plasma levels of patients receiving CsA treatment compared to blood from the same patients while on AZA or MMF.

Phenotypic analysis of lymphocytes infiltrating human cardiac allografts during acute rejection and the development of graft vascular disease

Abstract  Acute rejection (AR) and graft vascular disease (GDV) are processes mediated, at least in part, by cellular processes. Therefore, we cultured graft‐infiltrating lymphocytes (GIL) from endomyocardial biopsies (EMB) taken during the first year after transplantation, determined their phenotypic composition, and correlated it to AR and GVD. We observed more often GIL growth from EMB with AR than from non‐rejection EMB (P = 0.02), but no difference was found between patients with and without GVD 1 year after transplantation. CD4 + cells were always more numerous than CD8 + cells, and no difference in phenotypic composition was found between AR and non‐rejection EMB nor between EMB derived from patients with or without signs of GVD. In conclusion, AR is correlated with cell growth of EMB, but the development of GVD is not associated with AR, GIL growth from EMB, or their phenotypic composition.