Early hematopoietic reconstitution after clinical stem cell transplantation: evidence for stochastic stem cell behavior and limited acceleration in telomere loss

Our inability to purify hematopoietic stem cells (HSCs) precludes direct study of many aspects of their behavior in the clinical hematopoietic stem cell transplantation (HSCT) setting. We indirectly assessed stem/progenitor cell behavior in the first year after HSCT by examining changes in neutrophil telomere length, X-inactivation ratios, and cycling of marrow progenitors in 25 fully engrafted allogeneic HSCT recipients. Donors were sampled once and recipients at engraftment and 2 to 6 months and 12 months after HSCT. Telomere length was measured by an in-gel hybridization technique, X-inactivation ratios were measured by the human androgen receptor assay, and cell cycle status was determined by flow cytometric analysis of pyronin Y- and Hoechst 33342-stained CD34(+)CD90(+) and CD34(+)CD90(-) marrow cells. Compared with their donors, recipients' telomeres were shortened at engraftment (-424 base pairs bp]; P <.0001), 6 months (-495 bp; P =.0001) after HSCT, and 12 months after HSCT (-565 bp; P <.0001). There was no consistent pattern of change in telomere length from 1 to 12 months after HSCT; marked, seemingly random, fluctuations were common. In 11 of 11 informative recipients, donor X-inactivation ratios were faithfully reproduced and maintained. The proportion of CD34(+)CD90(+) progenitors in S/G(2)/M was 4.3% in donors, 15.7% at 2 to 6 months (P <.0001) after HSCT, and 11.5% at 12 months after HSCT (P <.0001, versus donors; P =.04, versus 2-6 months). Cycling of CD34(+) CD90(-) progenitors was largely unchanged. We infer that (1) HSCT-induced accelerated telomere loss is temporary and unlikely to promote graft failure or clonal hematopoietic disorders and (2) the striking fluctuations in telomere length and variation in pattern of telomere loss reflect stochastic determination of HSC fate after HSCT.