Abstract
Despite extensive studies over the last decades, little is known about the mechanisms governing human hematopoietic stem cell (HSC) fate decisions. In particular, it has been challenging to define culture conditions in which HSCs can be expanded for clinical benefit. Application of small molecule screening to modulate stem cells has emerged as a useful tool for identification of new compounds with ability to expand hematopoietic stem and progenitor cells (HSPCs). Such screens have mainly relied on the expression of CD34 as predictor of stem cell activity in cultured cells. However, CD34 defines a broad repertoire of progenitor cells and does not define stem cell function. We found that the long-term repopulation potential of cultured human HSPCs is exclusively contained within a discrete cell population co-expressing CD34 and CD90, while the vast majority of progenitor cells are found in the CD34+CD90- population. Tracking the CD34+ CD90+ population is therefore a sensitive and specific tool to predict stem cell activity in cultured hematopoietic cells and provides a good basis for a screen aimed at discovering modifiers of stem cell expansion. To search broadly for novel and potential modifiers of ex vivo HSCs expansion we next developed and optimized a small molecule screen in human cord blood (CB) derived CD34+ cells. We screened >500 small molecules from 8 different annotated chemical libraries for the phenotypic expansion of CD34+ CD90+ cells following a 6-day culture in serum-free medium supplemented with stem cell factor (SCF), thrombopoietin (TPO) and fms-like tyrosine kinase 3 ligand (FL). The numbers of CD34+ CD90+ cells for each molecule, tested at two different concentrations, was compared to DMSO treated controls. Following the initial screen, several candidate hits were selected and subjected to a dose response validation experiment from which we selected four top candidate molecules. Two of these molecules were histone deacetylase (HDAC) inhibitors, which recently have been reported to facilitate expansion of CB derived HSCs. One of the top candidates, Ciclopirox ethanolamine (CE), had previously not been implicated in HSC expansion. Ciclopirox ethanolamine is known as an antifungal agent and iron chelator. It has further been shown to suppress cancer cell survival through inhibition of Wnt/beta catenin signaling. We found that CB cells cultured with CE had a 4-fold increase in CD34+90+ cell number compared to DMSO treated controls following 6 days of culture. Interestingly, the total cell count was not different, suggesting a specific increase in CD34+ CD90+ cell number rather than an overall higher proliferation rate. When plated in methylcellulose, CE cultured cells generated increased numbers of myeloid colonies. Moreover, CE treated cells gave rise to multilineage colonies (CFU-GEMM) that could not be detected from the control cultures. To further test the functional capacity of cells cultured with CE, we transplanted cultured equivalents of 30,000 CB CD34+ cells (cultured with or without CE) into sub lethally irradiated NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice. Human hematopoietic reconstitution in peripheral blood was determined 16 weeks later. Mice transplanted with CE cultured cells showed higher human CD45 engraftment 16 weeks post transplant compared to control cells (33.2±6.7% vs 14.6±5% p=0.04). The engrafted cells contributed to both myeloid and lymphoid lineages. This shows that Ciclopirox ethanolamine enhances the long-term engraftment capacity of ex vivo cultured HSCs and suggests that it should be considered in stem cell expansion protocols, either alone or in combination with other molecules. We are currently addressing the basis for the increased stem cell activity mediated by Ciclopirox ethanolamine using parameters for differentiation, cell cycling and apoptosis. In addition, we are comparing Ciclopirox ethanolamine with other recently defined modifiers of HSC expansion.
Disclosures
No relevant conflicts of interest to declare.
Sox11 regulates mammary tumour-initiating and metastatic capacity in Brca1-deficient mouse mammary tumour cells