In patients undergoing hematopoietic stem cell transplantation (HSCT) infectious complications are frequent causing substantial morbidity and mortality. Adoptive T cell therapy specific for single pathogens has previously shown to efficiently control viral and fungal infections but approaches targeting multiple pathogens are limited to T cells generated with EBV transformed B cells that are genetically modified expressing multiple viral antigens. Infections are often experienced by different viral and fungal pathogens such as cytomegalovirus (CMV), Epstein-Barr virus (EBV), adenovirus (AdV), Aspergillus fumigatus (AF) and Candida albicans (CA) that show a wide spectrum of memory T cell frequencies. Those with a low precursor frequency are not suitable for selection methods based on the secretion of cytokines such as IFN-g. As CMV seropositivity among HSCT donors may range only between 30–40% and immunity to the other pathogens can be detected simultaneously in more than 85% of HSCT donors we focused on the generation of a multi-specific T cell product for EBV, AdV, AF and CA for easy transfer under current regulatory requirements. We aimed to develop a simple protocol which (i) is able to enrich T cells specific for pathogens with low precursor frequency and (ii) allows simultaneous expansion of multiple pathogen-specific T cells in a single culture.
We determined if CD154, which is transiently expressed on antigen stimulated CD4+ but also to a lesser extend on CD8+ T cells, would be a potential candidate for selection of pathogen-specific T cells. For stimulation we used peptide pools for AdV hexon protein, EBV latent membrane protein 2 (LMP2) and CA mannose protein 65 (MP65) as well as one AF immune dominant epitope derived from the Crf1 protein. To select and expand antigen-specific T cells, we stimulated PBMC for 16 hours, separated them by CD154+ MicroBead Kit (Miltenyi) and co-cultured them with irradiated autologous PBMC with IL-2, IL-7 and IL-15 for 14 days. The isolated cells were on average 0.62% of the starting fraction and could be expanded 20- to 145-fold. The median frequency of AdV-specific T cells increased from day 1 to day 14 87-fold from 30 to 2620 spot forming counts (SFC)/2×105 cells, for EBV 229-fold from 15 to 3430 SFC/2×105 cells and for CA 960-fold from 3 to 2400 SFC/2×105 cells assessed by IFN-γ ELISPOT. AF-specific T cells that were undetectable in PBMC increased to a median of 2260 SFC/2×105 cells. Although isolation of CD154+ cells favors enrichment of CD4+ T cells, a low fraction of virus-specific CD8+ T cells were simultaneously expanded. Next, we tested the efficacy of the CD154-based enrichment for the generation of multi pathogen-specific T cell lines reactive to all 4 pathogens. Selection and expansion was comparable, there was however a notable shift in the frequencies of T cells specific for different antigens in multi pathogen-specific cultures compared to single lines. The median increase of AdV-and CA- specific T cell lines was comparable (2345 SFC/2×105 and 3205 SFC/2×105 cells) but the frequencies for EBV (575 SFC/2×105 cells) as well as for AF (465 SFC/2×105 cells) were diminished in multi-specific lines. Nevertheless, lysis of LCL pulsed with LMP2 or AdV peptide pools was efficient with 72% and 36% by single and 30% and 45% by multi-specific T cell lines (at an E:T ratio of 20:1) as assessed by 51Cr-release assay. The single and multi pathogen-specific T cell lines generated by peptides responded to endogenously processed antigens and were able to specifically proliferate upon antigen stimulation. In contrast, T cell-mediated allo-reactivity was almost abrogated when compared to the starting population.
In conclusion, we established a simple expansion protocol for selection, expansion and enrichment of allo-depleted single and multiple pathogen-specific CD4+ and CD8+ T cells specific for AdV, EBV, AF and CA that may further expand if the T cells are stimulated by their native antigen in vivo. This expansion protocol may form the basis for adoptive immunotherapy trials in HSCT recipients at risk for multiple infectious complications.
This study has been supported by a grant of BayImmunet.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.