Abstract
Abstract 4016
The suboptimal number of hematopoietic stem cells (HSC) and relatively high incidence of graft failure following umbilical cord blood (CB) transplantation in adult patients remain as major obstacles to the widespread use of CB grafts. Using chromatin modifying agents (CMAs), including 5aza-2-deoxycytidine (5azaD) and trichostatin A (TSA), we have developed a culture strategy that permits 10 to 12 fold expansion of CB CD34+CD90+ cells containing in vivo repopulating cells (Araki et al. Blood 2007, Exp Hematol 2009). It is important that potential ex vivo expansion strategies not increase the capacity of the graft to mount an immune response to an allogeneic host. We have therefore assessed the allo-stimulatory capacity of expanded grafts following 9 days of culture examining immunophenotypic markers of dendritic cell (DC) commitment and also co-stimulatory molecules expressed on CMA-expanded CB cells. 5azaD/TSA-expanded cells co-expressing antigen presenting cell (APC) markers such as HLA-DR/CD86 comprise 17 ± 7% of the graft in contrast to 38 ± 9% in CB cells expanded in control culture and 2 ± 1% of primary CD34+ cells (n=3). Expression of HLA-DR/CD11c was only 3% in 5azaD/TSA-expanded cells in contrast to 37% in CB controls and 60% in primary CD34+ cells. Expression of the co-stimulatory molecule CD40 was not altered. We next performed mixed lymphocyte cultures (MLC), co-culturing irradiated primary or expanded CB cells with allogeneic peripheral blood T cells. The immune stimulatory capacity was measured by thymidine uptake, with CPM values of 3,213 ± 2205, 26,461 ± 670, and 2,859 ± 761 in primary CB cells, CB cells grown in control culture and 5azaD/TSA-expanded cells respectively (control vs. 5azaD/TSA p=0.001, 5azaD/TSA vs D0 p=0.893, n=3). The MLC results suggest that 5azaD/TSA-expanded CB grafts possess 9-fold less allo-stimulatory capacity than control cultures but are similar to primary CB cells. We hypothesized that this reduced allo-stimulatory capacity of 5azaD/TSA-expanded grafts could be due to three possible reasons:1) CMA-induced inhibition of expression of co-stimulatory molecules; 2) inhibition of the differentiation of CD34+ cells to APC; and/or 3) apoptosis of APC due to exposure to CMA. It has been previously shown that activation of the transcription factor STAT3 is indispensable for differentiation of DC from HSC. To study the molecular mechanisms involved in the reduced allo-stimulatory capacity of CMA expanded grafts, transcript levels of several genes, including STAT3 were determined by real time quantitative RT PCR. Interestingly the transcript levels of STAT3 were 2.5-fold higher in 5azaD/TSA-expanded CD34+ cells compared to control cultures. Furthermore the expression of two known inhibitors of STAT3, P21 and GATA1, were upregulated 10.5- and 6.8-fold respectively in 5azaD/TSA-expanded CD34+ cells. Therefore although STAT3 gene expression is increased in 5azaD/TSA-expanded cells it is likely inhibited by the increased levels of the inhibitors P21 and GATA1. In support of this we observed a reduced capacity for DC generation in cultures from 5azaD/TSA-expanded cells compared to control cultures. As expected, a LINE-1 assay, a surrogate marker of global methylation, shows that CpG methylation in 5azaD/TSA-expanded CD34+ cells is reduced to 66% compared to 78% in control cultures. Notably however the decrease in methylation specifically affects CpGs near the promoter site of GATA1 (control 82%, 5azaD/TSA 66%), but not STAT3 or P21. Furthermore, SHP1, a tyrosine phosphatase known to inhibit STAT3 activation, is also upregulated 2-fold in 5azaD/TSA-expanded CD34+ cells compared to control cultures. SHP1 transcription is regulated both by promoter methylation and by interactions with c-MYC, however no appreciable difference in c-MYC expression was detectable between 5azaD/TSA-expanded and control cultures suggesting that SHP1 upregulation in 5azaD/TSA-expanded CD34+ cells also involves promoter hypomethylation. Taken together, our studies indicate that the allo-stimulatory capacity of CB cells expanded in 5azaD/TSA- is reduced rather than increased, likely due to inhibition of an early stage of DC differentiation from CD34+ cells, which is mediated by inhibition of STAT3 activation. Reduced allo-stimulatory capacity of CMA expanded CB grafts may provide benefits by potentially allowing transplantation with less stringent HLA matching.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.