CMV infection is still a negative prognostic factor in hematopoietic stem cell transplantation, largely due to adverse effects of antiviral chemotherapy. Advances have been made in the development of adoptive cellular therapy for CMV with cell products derived from CMV seropositive donors. However no such cell products from CMV negative donors are currently available, although CMV negative recipients of CMV positive grafts bear the greatest risk of CMV induced morbidity and mortality. Mature DC are capable of inducing in vivo primary T cell responses. We used an in vitro culture system employing MoDC generated with GM-CSF and IL-4, pulsed with CMV lysate and matured with CD40L to induce primary immune responses to CMV. CMV-specific T cell proliferation measured by 3H-Thymidine uptake could be induced in some 6 to 9-day cultures, however, the success rate was very low, and T cells could usually not be kept alive after becoming activated. The addition of IL-15 to cultures after 7 days resulted in some improvement, although some non-CMV specific T cell proliferation in response to control lysate or in the absence of antigen also occurred in some cultures. IL-12 added to cultures from day 0 resulted in a short-term increase in T cell proliferation that was followed by increased cell death and was also not entirely CMV-specific. The real benefit of IL-15, alone or in combination with IL-12, was seen in 2-week cultures: In 6/7 cultures from different CMV seronegative donors counts of viable T cells (by trypan blue exclusion) increased up to 10-fold. Two cultures that were additionally set up with control lysate-pulsed DC to detect proliferation in response to non-CMV components of the antigen, confirmed CMV-specificity.

CMV-specificity was also shown by T cell receptor (TCR) complement determining region (CDR) 3 spectratyping. TCR-BV(variable region β)-size class expression patterns across 23 BV families were very similar in pre-culture unstimulated T cells and in T cells stimulated with control lysate-pulsed DC for 2 weeks, showing the typical near-normal distribution of PCR product amongst the different size classes indicative of un-stimulated T cells. T cells co-cultured with CMV lysate-pulsed DC for 2 weeks produced very skewed spectratypes, indicative of oligoclonal T cell expansion in response to the antigen. The effect of IL-15 on T cell spectratypes from CMV antigen-driven cultures was tested using cells from another donor. With or without IL-15, post-culture spectratypes showed oligoclonal T cell expansions in the same BV families and size classes. However, despite their similar shapes, spectratypes from IL-5 containing cultures were skewed to a greater extent, possibly indicating a greater effect of IL-15 on already activated T cells.

No CD8+ T cells specific for single immunodominant epitopes could be detected by staining with up to 5 different HLA-CMVpeptide-tetramers in culture output from 4/4 donors. In 1 of 2 cultures analyzed by cytokine secretion assay, a significant sub-population of T cells (1%), in CD4 positive and negative fractions, secreted IFN-γ in response to re-stimulation with CMV antigen. No IFN-γ secreting CMV-specific pre-cursor T cells were detected in fresh PBMC, as expected. Whilst further work is required to make generation of CMV specific T cells from CMV seronegative donors more reproducible and to ensure antigen-specificity, these preliminary data are encouraging for the future generation of CMV-specific T cells from CMV seronegative donors for adoptive cellular therapy.

Author notes

Corresponding author