E2 is the primary female sex hormone and plays a key role in various physiological events such as pregnancy, menstrual cycle, vasculogenesis, bone formation and lipid metabolism (11–14). E2 exerts its biological effects through the genomic (nuclear) and non-genomic (extranuclear) signaling pathways (15, 16). The genomic action of E2 is mediated by estrogen receptor α and β (ER-α and ER-β), which are dimeric nuclear receptors that bind to specific sequences of DNA known as E2 response elements. The DNA-ER complex then interacts with other co-factors and modulates activation or repression of target genes. E2 also can regulate gene expression independent of classical genomic pathway. The non-genomic actions of E2 are mediated by intracellular second messengers such as cavelolin-1, striatin, receptor tyrosine kinases and G protein-coupled receptor 30 (16). The non-genomic pathway confers the ability of the cell to rapidly respond to E2 stimulation. ER-α and/or ER-β are detected in different types of stem cells including HSCs, which suggests that E2 signaling in stem cells may be primarily mediated by ER-dependent genomic pathway (8). In this review, we first summarize the current knowledge of the effects of E2 on the proliferation of multipotent and pluripotent stem cells.

Accumulating evidence shows that E2 is involved in regulating the proliferation of multipotent mesenchymal stem cells (MSCs). This proliferative effect of E2 has been well defined in BM-MSCs. Hong et al. (17) demonstrated that E2 significantly enhances the proliferation of human BM-MSCs obtained from healthy male and female donors. Interestingly, this study showed that a wider range of E2 concentrations was observed to significantly promote proliferation in male BM-MSCs (1×10⁻¹² to 10⁻⁸ M) compared with that in female BM-MSCs (1×10⁻⁸ to 10⁻¹⁰ M). However, the molecular mechanisms of the gender difference of E2 regulation are still not elucidated. Other studies demonstrated that E2 enhances the proliferation and osteogenic differentiation of BM-MSC cultures obtained from healthy women with postmenopausal osteoporosis by activating Notch signaling pathway and upregulation of ERs transcripts (18, 19). These findings suggest that modulation of BM-MSCs by E2 can be a novel therapeutic strategy for treating oesteporosis. In addition, resveratrol, a polyphenolic phytoestrogen, induces an increase of [³H]-thymidine incorporation by human BM-MSCs derived from healthy donors, which is mediated via ER-dependent extracellular signal-regulated kinase 1/2 (ERK1/2) activation (20). The study indicates that synthetic or natural E2-like compounds including phytoestrogen may cause beneficial or adverse effects in stem cell behavior, which requires further investigations. Other studies indicate that the proliferative activity of E2 in MSCs is conserved across multiple tissues and species. Yun et al. (21) showed that E2 significantly increases [³H]-thymidine incorporation and the percentage of human umbilical cord-derived MSC population in the S phase via HIF-1α and VEGF expression through ER, protein kinase C, PI3K/Akt, and mitogen-activated kinase pathways. Recent studies found that E2 increases the number of cells in CFU-F colonies derived from rat BM-MSCs and upregulates the expression of anti-apoptotic Bcl-xL and Bcl-2, leading to maintenance of the cells (22). Zhou et al. (23) demonstrated that mouse BM-MSCs, treated with E2 (1×10⁻⁷ M), display a significant increase in ER-α mRNA and protein expression as well as proliferation rate. Together, these studies indicate that E2 is an important factor controlling the proliferation of BM-MSCs, which is mediated by various intercellular signaling pathways.

With defining the role of E2 in regulating proliferation of various adult and pluripotent stem cells, attempts were also made to define the role of E2 in regulating HSC functions during self-renewal and differentiation (Table 1). First evidence regarding the mechanism of E2 action on hematopoietic stem and progenitor cells (HSPCs) in the BM has been reported by Illing et al. (24). The study demonstrated that E2 increases HSPC number with reconstitutive potential in the vascular niche, but not in the endosteal niche. This increase is associated with the enhanced entry of HSPCs into S-phase independent of its ER-α-dependent anabolic bone effect. Similarly, the number and function of HSCs was significantly reduced in the ovariectomized (OVX) when compared with the Sham group (25). Furthermore, the significant reduction of hematopoietic growth factors including GM-CSF, SCF, and IL-3 was observed in the OVX group compared with the corresponding Sham group. These findings suggest a definite role of E2 in the proliferation of HSCs in the BM. HSCs are characterized by hierarchical organization in their self-renewal and differentiation, thus attempts were made to determine the effects of E2 on specific HSPC subpopulations in the BM. Indeed, recent study showed that ER-α and ER-β were differentially expressed in long-term (LT) repopulating HSC, multipotent hematopoietic progenitors (MPPs) and committed progenitors, suggesting differential role of E2 signaling in specific HSPC subpopulations (26). This study further demonstrated that while ER activation causes a rapid reduction in BM cellularity as well as the number of short-term (ST) HSCs and MPPs, the proliferation of quiescent LT-HSCs is induced by altered expression of c-Myc, associated with a loss of self-renewal activity and differentiation of HSCs. Notably, Nakada et al. (10) demonstrated that E2 promotes cell cycle activity of HSCs and MPPs in the BM, which is the tissue that does not show sex-specific morphological differences. In addition, HSCs in the female mice divide more frequently than in the male mice, which contributes to elevate the frequency of megakaryocyte-erythrocyte progenitor. This work paves the way towards a previously unexplored endocrine mechanism that controls HSC behavior and raises the question as to what effect does E2 have on LT repopulation of HSCs upon transplantation.

Human pluripotent stem cells (hPSCs) including ESCs, induced pluripotent stem cells (iPSCs), and ESCs derived somatic cell nuclear transfer are now being explored as a promising source for generating transplantable HSPCs and mature blood cells for treatment of various hematopoietic disorders (27). Although the role that E2 plays in hematopoietic development from hPSCs has not yet been defined, the expression of ER-α and ER-β has been reported in the early stage of human embryoid bodies (28). This finding suggests the influence of E2 on the proliferation and hematopoietic differentiation of human PSCs. Indeed, treatment of E2 (1×10⁻⁵ M) during spontaneous EB differentiation augments the expression level of brachyury mRNAs, associated with early hematopoietic specification. This contrasts with another study showing no alteration or reduction of brachyury expression in EBs treated with E2 (1×10⁻⁷ M and 1×10⁻⁹ M) (29). This discrepancy between these two studies may have been due to different doses and the period (14 days vs 30 days) of E2 treatment. Another set of experimentation has been performed to examine the influences of endocrine disruptors (EDs) including xenoestrogen on the proliferation and differentiation of PSCs (29–32). These works suggest the possibility that these EDs are able to mimic estrogen and may have potential risks in hematopoietic development, although further studies are needed to support this possibility.

Ontogeny of HSCs sequentially proceeds in discrete regions during embryogenesis where various cell components exert influence on the emergence, expansion, self-renewal and differentiation of HSCs (7). Despite the debate on the location of definitive hematopoiesis, HSCs are shown to first emerge from hemogenic endothelium in the aorta-gonad-mesonephros area in various model organisms and then migrate to the fetal liver for expansion (33). After birth, HSCs migrate to the BM from the fetal liver and predominantly reside in the BM in the adult. The function of HSCs is intimately regulated by the BM niche primarily consisting of nestin-expressing cells, CXCL12-abundant reticular cells and osteoblasts (34). Mounting studies have shown that cell components in the BM niche regulate the quiescence, self-renewal, mobilization and differentiation of HSCs either through secreting or presenting a variety of signaling molecules on the cell surface (34). On the other hand, humoral factors and hormones in the circulation, in addition to its direct regulation of HSCs, can also influence the activities of the BM niche cells to indirectly modulate the function of HSCs (Table 1).

Osteoblasts, derived from MSCs, are one of the most important cell components in the BM niche. Previous studies have shown that osteoblasts can produce signaling molecules such as IL-6, GM-CSF, and Notch ligand Jagged 1 to regulate the activities of HSCs (35, 36). Elegant studies from Scadden’s and Li’s groups have demonstrated that parathyroid hormone (PTH) and BMP signaling can directly regulate osteoblasts to control the number and function of HSCs, respectively (36, 37). These studies suggest that hormone in the circulation may indirectly modulate HSCs through controlling osteoblasts in the BM niche. In line with this, Qiu et al. (38) has shown that E2 induces the proliferation of HSCs, which is dependent on that E2 promotes the osteogenic differentiation of MSCs. The effect of E2 on osteogenic differentiation of MSCs is mediated through ER expressed in MSCs (20, 39). Recent studies have shed light on understanding the downstream mechanisms underlying the process. Gao et al. (40) has reported that E2 may promote osteogenic differentiation of MSCs through regulating G protein-coupled receptor 40 (GPR40). E2 has also been shown to augment osteogenic differentiation of MSCs by modulating both Wnt/β-catenin and Notch signaling pathways (40, 18). Interestingly, previous study has shown that bone cells express estrogen receptor, suggesting that E2 may play a role in the bone remodeling (41). Knockout of ER-β study demonstrated that ER-β represses bone growth during female adolescence (42). Moreover, E2 has been shown to maintain bone mass by inducing apoptosis of osteoclast (43, 44). Taken together, it suggests that E2 may be involved in the remodeling of the BM niche through controlling the activities of osteoclasts and bone cells. Thus, E2 may control the BM niche to regulate the biological functions of HSCs through diverse mechanisms where multiple signaling pathways are likely to be engaged.