Search

SummaryAdult hematopoietic stem cells (HSCs) with serially transplantable activity comprise two subtypes. One shows a balanced output of mature lymphoid and myeloid cells; the other appears selectively lymphoid deficient. We now show that both of these HSC subtypes are present in the fetal liver (at a 1:10 ratio) with the rarer, lymphoid-deficient HSCs immediately gaining an increased representation in the fetal bone marrow, suggesting that the marrow niche plays a key role in regulating their ensuing preferential amplification. Clonal analysis of HSC expansion posttransplant showed that both subtypes display an extensive but variable self-renewal activity with occasional interconversion. Clonal analysis of their differentiation programs demonstrated functional and molecular as well as quantitative HSC subtype-specific differences in the lymphoid progenitors they generate but an indistinguishable production of multipotent and myeloid-restricted progenitors. These findings establish a level of heterogeneity in HSC differentiation and expansion control that may have relevance to stem cell populations in other hierarchically organized tissues.

Mouse haematopoietic stem cells (HSCs) undergo a postnatal transition in several properties, including a marked reduction in their self-renewal activity. We now show that the developmentally timed change in this key function of HSCs is associated with their decreased expression of Lin28b and an accompanying increase in their let-7 microRNA levels. Lentivirus-mediated overexpression of Lin28 in adult HSCs elevates their self-renewal activity in transplanted irradiated hosts, as does overexpression of Hmga2, a well-established let-7 target that is upregulated in fetal HSCs. Conversely, HSCs from fetal Hmga2−/− mice do not exhibit the heightened self-renewal activity that is characteristic of wild-type fetal HSCs. Interestingly, overexpression of Hmga2 in adult HSCs does not mimic the ability of elevated Lin28 to activate a fetal lymphoid differentiation program. Thus, Lin28b may act as a master regulator of developmentally timed changes in HSC programs with Hmga2 serving as its specific downstream modulator of HSC self-renewal potential. Fetal haematopoietic stem cells (HSCs) display higher self-renewal potential than their adult counterparts. Eaves and colleagues show that adult HSCs express lower levels of Lin28b and higher levels of let-7 miRNA. They demonstrate that Lin28 overexpression, as well as that of Hmga2 (a target of let-7), induces fetal properties in adult HSCs. Conversely, HMGA2 loss in fetal HSCs results in premature induction of adult HSC self-renewal properties.

Abstract 45 Fetal hematopoietic stem cells (HSCs) in mice differ from their adult counterparts in a number of key properties. These include a higher cycling activity, an ability to more rapidly reconstitute the HSC compartment of irradiated recipient mice, a higher output of myeloid as compared to lymphoid progeny, and a greater sensitivity to the self-renewal promoting activity of Steel factor. We have previously shown that most of these features of fetal HSCs are sustained until 3 weeks after birth at which time they are rapidly (within 1 week), completely and permanently replaced with the corresponding properties of adult HSCs. A candidate regulator of this transition, Hmga2, was identified based on its greater expression in highly purified fetal versus adult HSCs (CD45+EPCR+CD48−CD150+; E-SLAM cells) with persistence of this difference in the matching lineage-negative (lin−) compartments. Experiments in which Hmga2 was overexpressed by lentiviral transduction of purified adult HSCs which were then transplanted into irradiated mice provided evidence that this chromatin remodeling factor can activate a fetal-like HSC program in these cells; i.e., more rapidly reconstitute the HSC compartment (increased self-renewal response) and produce clones with a higher proportion of myeloid cells. Based on the known ability of the let-7 family of microRNAs (miRNAs) to target Hmga2 transcripts resulting in their degradation and/or translational repression, we next hypothesized that let-7 miRNAs might be involved in controlling HSC developmental programs. A comparison of the levels of expression of 6 members of the let-7 family in purified fetal and adult HSCs, as well as in lin− hematopoietic cells, showed that transcripts for all of these are higher in the adult subsets, although this difference was significant only for let-7b (p Disclosures: No relevant conflicts of interest to declare.

Abstract 576 Hematopoietic stem cells (HSCs) represent a rare subset of bone marrow (BM) cells in adult mice that are ultimately responsible for the replenishment of all different mature blood cell types through a hierarchically organized cascade of differentiation steps. Although HSCs were previously considered a functionally and phenotypically homogeneous population, recent studies tracking the 4 to 6 month progeny of a large number of highly purified, individually transplanted, single mouse HSCs and their serially generated derivatives in sublethally irradiated hosts clearly reveal a functional heterogeneity of HSCs. The results have revealed 2 distinct and stably propagated clonal subtypes of HSCs with unrestricted self renewal activity that we have termed α- and β-HSCs. β-HSCs are those that show a relatively balanced output of mature myeloid and lymphoid cells. α-HSCs have an equivalent self-renewal and myelopoietic activity to β-HSCs but, in contrast, are characterized by a variable and often extreme failure to produce mature lymphoid cells. Since it is well established that the reduced activity of the immune system with aging corresponds with a decrease in the frequency and activity of both B- and T lymphocytes and their respective progenitors, it is of interest to determine whether this simply represents a generalized aging of the hematopoietic system or rather an age-related change in the composition of the HSC compartment. The most rigorous approach to determine the time of appearance and kinetic changes in the distribution of different types of HSCs during development and aging is to study single-cell transplants and their clonal progeny using an ontogeny-independent HSC purification scheme. We recently showed that the E-SLAM (CD45+ Endothelial protein receptor EPCR+ CD150+ CD48−) phenotype could be used to achieve high HSC purity across development (E14.5 fetal liver, 3-week BM, 4-week BM, 10–12 week BM and aged BM) despite known ontological differences in HSC proliferative activity between these populations. These highly-purified populations of HSCs can thus be utilized to prospectively isolate and singly transplant these HSCs to enable study of their clonally progeny. Clonal analysis of the E-SLAM HSC compartment of aged mice showed an increase in frequency of α-HSCs. To investigate the mechanism underlying the lymphopoietic deficiency characteristic of α-HSCs, we undertook a set of experiments to compare the in vivo generated progeny of α- vs β-HSCs. We particularly focused on a quantitative and qualitative analysis of the common lymphoid progenitor (CLP) compartment (Lin− CD127+ CD117low Sca1low). In contrast to previous reports of a decrease in CLP numbers in old mice, we simply found a broader distribution of CLPs in old mice. We did however find a decreased number of CLPs generated per HSC measured as the CLP/LSK ratio (Lin− CD127− CD117+ Sca1+) with age suggesting that – if the aged BM is dominated by α-HSCs – that α-HSCs possess a quantitative defect in CLP production. Intriguingly, the number of CLPs derived from mice reconstituted with an α stem cell were significantly reduced in comparison to CLPs derived from β-HSCs. To measure the B-cell production activity of CLPs produced from α- vs β-HSCs we utilized the OP9 co-culture system and plated limiting numbers of CLPs derived from reconstituted mice which were then compared to the CLP productivity of young and old CLPs from non-reconstitued mice. Functional analysis of CLPs from individual young mice showed a consistent B-cell frequency of 1/3.6 (n=13) in the OP9 co-culture system whereas CLPs from old mice showed either good B-cell potential comparable to young CLPs or very reduced B–cell potential. CLPs derived from β-HSCs show a consistent B-cell potential (1/11.3, n=9) even after reaching a physiological age of >100wks whereas when CLPs could be detected from α-HSCs they showed a reduced B-cell activity in vitro. The results of our study demonstrate that the decrease in lymphoid activity during aging is likely caused by the relative increase in α-HSCs which have both a quantitative and qualitative CLP defect. Disclosures: No relevant conflicts of interest to declare.

Abstract 2634 Fetal and early neonatal hematopoietic stem cells (HSCs) are distinct from their adult counterparts by their rapid turnover and expansion rates in vivo. However, the mechanisms underlying the regulation of these properties are poorly understood. In previous studies using serial limiting-dilution competitive repopulating transplant assays, our lab has shown that the rapid expansion phenotype of fetal HSCs is at least partially intrinsically determined since significantly more daughter HSCs are produced from fetal as compared to adult HSCs when similar numbers are transplanted into the same type of irradiated adult host. Additionally, we have observed a conversion of fetal HSCs to the adult regeneration phenotype that occurs within six weeks of transplantation in the primary host. To facilitate a comparison of highly-purified subsets of fetal and adult HSCs identified by an identical phenotype, we adopted the use of the CD45+EPCR+CD150+CD48− (E-SLAM) phenotype which we found gave HSC purities of 20–50% for hematopoietic tissues from early fetal to aged adulthood. We then used comparative gene expression analysis to identify candidate regulators of the fetal HSC high self-renewal program. This gave 20 candidate genes whose transcript levels were measured by quantitative real time PCR in E-SLAM cells isolated from E14.5 fetal liver (FL) and adult bone marrow (ABM). Of these genes only Hmga2 and Smarcc1 showed significant differences (p Disclosures: No relevant conflicts of interest to declare.