Abstract
Abstract The HSC/MPP-like cell line EML has the capacity for multilineage (erythroid, myeloid, lymphoid) differentiation in vitro, and is functionally heterogeneous. Phenotypically, EML cell line consists of Sca-1+ c-kit+ Flk− CD34+ cells (80%) and Sca-1+ c-kit+ Flk− CD34− cells (20%), and resembles Lin− c-kit+ Sca-1+ Flk2− BM cells, which are also heterogeneous for CD34 expression. Functional analysis has revealed that purified CD34+ EML cells reproducibly contain the majority of progenitors capable of mutilineage differentiation in response to cytokines, whereas CD34− EML cells differentiated poorly or not at all. Since the CD34+ EML cell population can be divided into CD34low, CD34medium and CD34high subpopulations we investigated the correlation between the progenitor content and the level of CD34 expression. The CD34low, CD34med and CD34high EML cells were purified by FACS sorting, and their multilineage differentiation analyzed by CFC assays. Multiple CFC experiments have reproducibly shown that the progenitor frequency declines with a decrease in the level of CD34 expression on EML cells, with CD34high cells producing the highest and the CD34low cells producing the lowest number of colonies. These findings indicate that CD34+ EML cell population is functionally heterogeneous, and that the overall multilineage differentiation of EML cells strongly correlates with the CD34 expression levels. Since the majority of LTR-HSC reside within LKS Flk2− CD34− BM cells, and LKS Flk2− CD34+ BM cells are highly enriched for STR-HSC and MPPs, we tested whether CD34− EML cells are more primitive and give rise to CD34+ cells. The CD34− and CD34+ EML cells were purified (>99% pure) by sorting, cultured separately for 7 days, and cell aliquots from these cultures analyzed daily for CD34 expression. Four separate experiments have revealed that both CD34− and CD34+ EML cells can generate each other, suggesting that although the CD34− EML cells could be more primitive, a subpopulation of CD34+ cells retains the capacity to give rise to CD34− cells. Previous studies have shown that mouse LTR-HSC can alternate between CD34− “inactive” and CD34+ “activated” state, and that the heterogeneous population of LKS CD34+ BM cells contains activated LTR-HSC, STR-HSC and MPPs. Thus, we investigated how the level of CD34 expression correlates with the capacity of CD34+ EML cells to give rise to CD34− cells. Purified CD34low, CD34med and CD34high EML cells were cultured for 5 days, and the cells from each culture analyzed for CD34 expression. Remarkably, three separate experiments have revealed that CD34low EML cells have the highest capacity to give rise to CD34− EML cells, whereas CD34med and CD34high EML cells had very little capacity to generate CD34− EML cells. These findings suggest a model in which (a) CD34− EML cells are more primitive cells in an “inactive” (differentiate inhibited) state that generate CD34+ EML cells, and (b) the population of CD34+ EML cells is heterogeneous and consists of CD34med/high cells that can readily differentiate into multiple lineages, and CD34low cells that are in “activated” state and can revert back to the CD34− state. Cumulatively, these results support the notion that the maintenance and differentiation of EML cells is linked with the varying levels of CD34 expression. Therefore, EML cell line could represent a powerful in vitro model to study the linear differentiation of CD34− cells into CD34+ cells, and the mechanisms of reversible CD34 expression and activation of primitive hematopoietic cells.