Cord Blood Allograft Rejection Mediated By Coordinated Donor-Specific Cellular and Humoral Immune Processes

Recent statistical analyses of the clinical outcome have revealed that the presence of donor-specific anti-HLA antibodies (DSAs) in pre-transplant recipient is correlated with increased risk for graft rejection after allogeneic hematopoietic stem cell (HSC) transplantation. However, while cytotoxic T lymphocytes (CTLs) recognizing mismatched HLA molecule of the donor have been shown to be involved in graft rejection, there has been no biological evidence in humans that graft rejection is mediated by DSAs. In the present study, we demonstrate a case of cord blood allograft rejection in which DSAs acted to mediate graft rejection. Interestingly, CTL clones specific for donor HLA molecule were also isolated, suggesting that humoral and cellular immune responses were together responsible for allograft rejection.

A patient with graft rejection after HLA-mismatched cord blood transplantation was studied. The white blood cell count increased transiently, subsequently decreased to undetectable level, and finally graft rejection was diagnosed on day 34. Screening for pre-transplant anti-HLA antibodies was not routinely performed at that time. We initially presumed that DSA would be responsible for graft rejection. The patient serum on day 36 were screened for HLA antibodies and further evaluated to determine HLA specificities using a LABScreen Single Antigen Kit. An antibody against HLA-B* 54:01, which was expressed in donor cells but not in patient cells, was detected. To test the effect of the DSA against HLA-B* 54:01 on HSC engraftment, antibody-dependent cellular cytotoxicity (ADCC) activity was assessed using unrelated HLA-B* 54:01-positive bone marrow mononuclear cells (BMMNCs) pre-cultured with patient serum and pre-transplant NK cells. The patient serum clearly showed an inhibitory effect on colony formation. Complement-dependent cytotoxicity activity of the DSA could not be detected. These data suggest that the DSA impaired the cord blood engraftment through ADCC activities. We next determined if cellular immunity was also responsible for allograft rejection. Two independent CTL clones, CD4^-CD8⁺ and CD4⁺CD8^-, were isolated from the patient blood at the time of graft rejection by limiting dilution. Both CTL clones were of patient origin by using short tandem repeat analysis and showed cytotoxicity against donor cells but not patient cells in Cr release assay. A CTL stimulation assay using COS cells transfected with various mismatched donor HLA cDNA constructs demonstrated that CD8⁺ and CD4⁺CTL clones recognized HLA-B* 54:01 and -DRB1* 15:02 molecules, respectively, both of which were expressed in donor cells but not in patient cells. To test the effect of the CTL clones on engraftment, a colony forming assay using either HLA-B* 54:01 or -DRB1* 15:02-positive unrelated BMMNCs pre-cultured with the corresponding CTL clones was performed. Each of these CTL clones inhibited colony formation from unrelated BMMNCs sharing target HLA in a dose-dependent manner. These data suggest that the CTLs also impaired the cord blood engraftment. Furthermore, anti-HLA-B* 54:01 antibody was detected in cryopreserved pre-transplant patient serum, and in nested PCR assays specific for the HLA-B* 54:01-specific CTL clone’s uniquely rearranged T cell receptor V beta chain, PCR product was detected by amplification of cDNA from pre-transplant patient peripheral blood mononuclear cells. Thus, both HLA-B* 54:01-specific antibody and CTL clone developed in the patient prior to transplantation.

The present data provide the first direct evidence showing that humoral and cellular immune responses were involved in the graft rejection. Of note is the recognition of the same mismatched HLA-B* 54:01 molecule by the DSA and CTL, suggesting that humoral and cellular immune processes acted together to mediate allograft rejection. Given the difficulty in detecting HLA-specific CTLs in pre-transplant patient blood in contrast to the easiness in screening for DSAs, the presence of DSA not only means a direct deleterious effect on donor cells but it may also reflect the presence of CTLs that causes allograft rejection. Further studies are warranted to clarify whether the present observation can be duplicated in other patients with DSA. We are conducting the study to clarify whether patients with anti-HLA antibodies have CTLs that recognize the same HLA molecules before transplantation.