Umbilical cord blood–naive T cells but not adult blood–naive T cells require HLA class II on antigen-presenting cells for allo-immune activation

Graft predominance after double umbilical cord blood transplantation: a review

Several parameters are involved in graft predominance after double umbilical cord blood transplantation (dUCBT), of which T-cell alloreactivity between the grafts is now considered to be the major denominator. We recently showed that alloreactive CD4+ T-cells originating from the predominant CBU recognize HLA-class II allele mismatches and can readily be detected in the majority of patients. In addition, it was shown that HLA-class II allele-specific CD4+ T-cells were able to recognize primary leukemic cells when the mismatched HLA-class II allele was shared between the rejected CBU and the patient. These results further underscored the role of alloreactive T-cells, notably class II specific CD4+ T-cells, in graft-versus-graft reactions and in graft-versus-leukemia after dUCBT.

CD4+ T-cell alloreactivity toward mismatched HLA class II alleles early after double umbilical cord blood transplantation

Although double umbilical cord blood transplantation (dUCBT) in adult patients may be associated with less graft failure compared with single UCBT, hematopoietic recovery generally originates from a single cord blood unit (CBU). CBU predominance is still incompletely understood. We recently showed that blood CD4⁺ T-cell numbers rapidly increase after dUCBT, and early CD4⁺ T-cell chimerism predicts for graft predominance. Given the frequent HLA class II allele mismatches between CBUs in dUCBT, we hypothesized that alloreactive HLA class II-specific CD4⁺ T cells from the "winning" CBU may contribute to rejection of the "loser" CBU. We evaluated whether CD4⁺ T cells originating from the predominant (PD)-CBU would recognize HLA class II allele mismatches, expressed by the nonengrafting (NE)-CBU. Alloreactive effector CD4⁺ T cells toward 1 or more mismatched HLA class II alleles of the NE-CBU were detected in 11 of 11 patients, with reactivity toward 29 of 33 (88%) tested mismatches, and the strongest reactivity toward DR and DQ alleles early after dUCBT. Mismatched HLA class II allele-specific CD4⁺ T cells recognized primary leukemic cells when the mismatched HLA class II allele was shared between NE-CBU and patient. Our results suggest that cytotoxicity exerted by CD4⁺ T cells from the PD-CBU drives the rapid rejection of the NE-CBU, whose alloreactive effect might also contribute to graft-versus-leukemia.

Cord blood nucleated cells induce delayed T cell alloreactivity

Cord blood (CB) mononuclear cells (MNCs) can be transplanted in HLA mismatched recipients with limited graft rejection or graft-versus-host disease (GVHD). Previous studies have shown that naive T cells and hyporesponsive dendritic cells are largely represented in CB. Data presented here demonstrate that CB MNCs are unable to stimulate allogeneic T cell proliferative or cytotoxic responses in standard in vitro assays. However, a suppressive effect of CB MNCs was ruled out because purified CD34(+) cells or CD14(+) monocytes stimulated T cell responses that were not inhibited by add-back of CB MNCs. The lack of antigen-presenting cell (APC) activity of CB MNCs in primary mixed lymphocyte culture (MLC) did not induce allogeneic T cell anergy. In fact, rechallenge of T cells with CB CD34(+) cells, or immature monocyte-derived dendritic cells (iMo-DCs) in secondary MLC induced potent T cell proliferative responses. A delayed APC activity of CB MNCs was observed after stimulation with irradiated allogeneic T cells for 6 days, likely because of the upregulation of CD86 and HLA-DR on CB cells. Cytotoxic lymphocytes (CTL) were generated after stimulation of blood T cells with CB MNCs for 4 weeks or CB-derived iMo-DCs for 1 week. Concomitant stimulation of T cells with CB iMo-DC obtained from 2 CB units resulted in the generation of CTLs specific for each CB, independently of the CB:CB cell ratio. These data suggest that the APC activity of CB cells possibly increases posttransplant, and may contribute to delayed graft rejection and/or acute and chronic GVHD.

Umbilical cord blood transfusions for prevention of progressive brain injury and induction of neural recovery: an immunological perspective

One of the most promising treatments for neurodegenerative diseases appears to be human umbilical cord blood cell transplantation. A variety of studies demonstrate some benefit of this method of treatment in a number of different animal models and case studies. However, before the methodologies and results of these animal studies and case studies can be translated into successful widespread treatments, aspects relating to the immunological properties of the transplanted cells must be considered. In this perspective, we discuss the benefit of the cellular immaturity of these cells with respect to the immune response, and compare cord blood transplantation to blood transfusions, as well as discussing what future studies should entail.

Early detection and rapid isolation of leukemia-reactive donor T cells for adoptive transfer using the IFN-gamma secretion assay

PURPOSE

The poor immunogenicity of most leukemias and the lack of specificity of the donor T cells limit the in vivo effectiveness of conventional donor lymphocyte infusions in many patients suffering from persistent or recurrent leukemia after allogeneic stem cell transplantation. These limitations may be overcome by the adoptive transfer of in vitro generated leukemia-reactive T cells. Although the potential clinical efficacy of this approach has been shown previously, lack of reproducibility of the procedure and the inability to show persistence and survival of the transferred T cells hampered further clinical application. The purpose of this study was to develop a new, broadly applicable strategy for the efficient generation and isolation of leukemia-reactive T cells with a better probability to survive and expand in vivo.

EXPERIMENTAL DESIGN

Myeloid and B-cell leukemias were modified into professional immunogenic antigen-presenting cells, and used to stimulate HLA-matched donor T cells. After two stimulations, responding donor T cells were isolated based on their secretion of IFN-gamma and tested for their capacity to recognize and kill the primary leukemia.

RESULTS

Using one universal stimulation and isolation protocol for various forms of leukemia, T-cell populations containing high frequencies of leukemia-reactive T cells could reproducibly be generated and early isolated under mild stimulatory conditions. Isolated T cells still had high proliferative potential and their reactivity seemed to be restricted to cells of the patient's hematopoiesis.

CONCLUSION

We here show a new robust procedure for the generation and isolation of leukemia-reactive T cells for adoptive transfer.

Similar potential to become activated and proliferate but differential kinetics and profiles of cytokine production of umbilical cord blood T cells and adult blood naive and memory T cells

Low alloreactivity of umbilical cord blood (UCB) T-cells may explain diminished graft-versus-host-disease after UCB transplantation. We investigated whether UCB T-cells have an intrinsic lower capacity to become activated. T-cells from UCB or adult blood (AB) were stimulated with anti-CD3 and anti-CD28 antibodies. On days 1-3 after stimulation, T-cell activation was determined by CD25 expression, proliferation was measured, and kinetics of cell division were analyzed by staining with CFSE. UCB and AB T cells exhibited similar numbers of activated and proliferating cells, but the extent of activation was lower in UCB T-cells. Enzyme-linked immunospot analysis showed lower levels and slower kinetics of IL-2, IL-4, IL-10, and IFN-gamma secreting cells for UCB T-cells. Comparison of UCB T-cells with CD45RA+ naive or CD45RO+ memory T cells purified from AB showed relatively low numbers of IL-4 and IL-10 secreting T cells in CD45RA+ AB T-cells and UCB T-cells as compared with CD45RO+ AB T cells. Numbers of IL-2 or IFN-gamma secreting cells in adult CD45RA+ T-cells were lower than in CD45RO+ T-cells but higher than in UCB T-cells. Thus diminished reactivity of UCB T-cells was not caused by a lower capacity to become activated and proliferate but may be explained by a lower extent of activation in UCB T cells, the absence of memory T cells in UCB, and differences between naive T cells from UCB and AB.