Prostaglandin E₂ enhances long-term repopulation but does not permanently alter inherent stem cell competitiveness

Since the mid-1970s, a hematopoietic regulatory role for prostaglandin E₂ (PGE₂) has been described, demonstrating both inhibitory and stimulatory effects dependent on the cell type studied and exposure kinetics (reviewed elsewhere).^1-4  Using a zebrafish embryo chemical screen, a long-acting agonist of PGE₂, 16-16 dimethyl-PGE₂ (dmPGE₂), was shown to increase hematopoiesis, whereas inhibitors of PGE₂ biosynthesis decreased hematopoiesis.⁵  Using short-term ex vivo pulse exposure of bone marrow cells, similar to exposure strategies previously reported,^6,7  North and colleagues elegantly demonstrated that murine bone marrow transplantation was enhanced by dmPGE₂.⁵  We later demonstrated that this enhanced hematopoietic stem cell (HSC) engraftment resulted from an increase in CXCR4 on hematopoietic stem and progenitor cells and enhanced homing to the marrow; it also increased expression of Survivin, with reduced HSC apoptosis and increased HSC division.⁸  In addition, we showed that enhanced HSC engraftment was maintained in secondary transplantation.⁸  Enhanced HSC production and long-term repopulation by PGE₂ was also shown to be mediated through enhanced Wnt/β-catenin signaling.⁹ 

Based on these preclinical findings, a phase 1 clinical trial evaluating safety and efficacy of ex vivo dmPGE₂ pulse treatment of umbilical cord blood cells was initiated, the results of which are published in this edition of Blood.¹⁰  One prime question raised by treatment of HSCs with dmPGE₂, however, is whether its effects on HSCs are short lived or whether treatment alters long-term potential. Here, we report on extended analysis of dmPGE₂-treated hematopoietic grafts, following multilineage repopulation of hematopoiesis through 5 serial transplantations and comparison of the engraftment potential of vehicle- and dmPGE₂-treated HSCs on a cell-to-cell basis. We demonstrate that dmPGE₂ treatment does not alter long-term HSC competitiveness, lineage bias, or proliferative potential.

C57Bl/6 (CD45.2) mice were purchased from Jackson Laboratories (Bar Harbor, ME). B6.SJL-PtrcAPep3B/BoyJ (BOYJ) (CD45.1) and C57Bl/6 × BOYJ F1-hybrid mice (CD45.1/CD45.2) were bred in-house. All mice in transplant studies received doxycycline feed for 30 days posttransplant. The Animal Care and Use Committee of the Indiana University School of Medicine approved all protocols.

Competitive transplantation was performed as previously described,⁸  and the primary and secondary transplant data are reflective of these previously published results. For serial transplants, 2 × 10⁶ whole bone marrow (WBM) cells from previously transplanted CD45.1/CD45.2 F1-hybrid mice were injected into lethally irradiated CD45.1/CD45.2 F1-hybrid mice in noncompetitive fashion. Secondary, tertiary, quaternary, and quinary transplants were performed in a similar manner, with the tertiary transplant performed 24 weeks after the secondary transplant, and the quaternary and quinary transplants performed 12 weeks after the prior transplant. Chimerism and multilineage flow analysis was performed as previously described and representative flow plots of multilineage gating were shown.⁸ 

WBM cells from CD45.1 and CD45.2 donors were isolated and treated with vehicle or dmPGE₂ for 2 hours on ice. One cohort of lethally irradiated F1-hybrid CD45.1/CD45.2 mice were transplanted with 5 × 10⁵ vehicle-treated CD45.1 cells and 5 × 10⁵ dmPGE₂-treated CD45.2 cells. A separate cohort of irradiated F1-hybrid CD45.1/CD45.2 mice were transplanted in a similar fashion with strain and treatment groups reversed. After 12 weeks, peripheral blood chimerism was evaluated and bone marrow was acquired and stained for CD45.1/CD45.2 and SLAM SKL (lineage negative, c-kit⁺, Sca-1⁺, CD150⁺, CD48⁻) markers. CD45.1⁺ and CD45.2⁺ SLAM SKL cells were individually isolated by fluorescence-activated cell sorting, and a second group of irradiated F1-hybrid CD45.1/CD45.2 mice were transplanted with 2.5 × 10² CD45.1⁺ SLAM SKL cells, 2.5 × 10² CD45.2⁺ SLAM SKL cells, and 2.0 × 10⁵ CD45.1/CD45.2 F1-hybrid WBM cells as competitors. Contribution to chimerism of CD45.1 and CD45.2 SLAM SKL cells was evaluated 12 weeks later.

In a previous report, we used a competitive head-to-head transplant model that permitted comparison of engraftment and competitiveness of vehicle- and dmPGE₂-treated HSCs within the same animal.⁸  In the prior report, we validated enhanced engraftment after dmPGE₂ treatment and showed that this enhancement persisted through secondary transplantation without additional treatment of cells before retransplant (Figure 1A).⁸  We continued serial transplantations at 12-week intervals, monitoring chimerism and multilineage reconstitution through 5 serial transplantations. Grafts previously treated with dmPGE₂ maintained higher levels of chimerism through quaternary serial transplantation, with HSC exhaustion occurring by the quinary transplant, which is consistent with normal HSC exhaustion in a similar experimental model.¹¹  Bone marrow grafts from vehicle or dmPGE₂ treatment groups contributed to full multilineage reconstitution throughout the first 4 serial transplantations (Figure 1B). Although it was commonly assumed that a single HSC compartment gives rise to all hematopoietic lineages equivalently, transplant studies have demonstrated the presence of HSCs biased toward lymphoid or myeloid differentiation.^12-14  In secondary transplants, we observed a myeloid bias in mice transplanted with dmPGE₂-treated HSCs, suggesting a possible selective effect on myeloid-biased HSCs. However, a consistent lineage bias was not observed in subsequent serial transplants (Figure 1B).

PGE₂ signaling is a known mediator of cancer^15,16  and intersects with multiple signaling pathways downstream of 4 G protein–coupled receptors. Wnt/β-catenin signaling was recently shown to be required for the development of leukemia stem cells in acute myelogenous leukemia¹⁷ ; this pathway lies downstream of PGE₂ receptors. Throughout the quinary transplants, peripheral blood white blood cell differential counts remained within normal range in all mice at the 12-week time points. Moreover, histological observation indicated that there was no obvious alteration in the bone marrow cellular composition between quinary transplanted mice that had received grafts originally treated with dmPGE₂ before primary transplant vs bone marrow from control-graft transplanted mice (Figure 1C). These results suggest that dmPGE₂ pulse treatment does not lead to a preleukemic or myeloproliferative state.

Increased hematopoietic output by dmPGE₂-treated grafts through serial transplants could simply be the result of higher numbers of HSCs homed and engrafted in the primary graft, proportionally occupying available niche space and self-renewing. Alternatively, enhanced hematopoietic output could result from epigenetic changes in HSC elicited by dmPGE₂ treatment,^18,19  leading to long-term enhanced competitiveness, preferential dmPGE₂ stimulation of a subset of HSCs with enhanced inherent competitiveness,²⁰  or an unknown mechanism eliciting long-term competitive advantage. We reasoned that if serial transplant results are simply due to an increase in homed and engrafted HSCs in the primary graft, then “equalizing” the HSC content from dmPGE₂- and vehicle-treated grafts in a subsequent secondary transplant should result in equal repopulation. To test this hypothesis, dmPGE₂- and vehicle-treated congenic bone marrow cells were transplanted head to head into lethally irradiated F1-hybrid mice, and chimerism was evaluated 12 weeks later (Figure 2A). As seen in earlier transplants, dmPGE₂ treatment resulted in significantly increased chimerism (Figure 2B, left). Bone marrow from these primary recipient mice was subsequently harvested and fluorescence-activated cell sorted for congenic SLAM SKL cells. SLAM SKL cells derived from previously treated dmPGE₂ grafts were 2.1-fold more abundant than those from vehicle-treated grafts (0.0295% SLAM SKL cells in total marrow vs 0.0138%). We subsequently transplanted equal numbers of the sorted congenic SLAM SKL cells head to head into secondary recipients along with radioprotecting F1-hybrid bone marrow competitors without additional treatment. Equalizing the HSC content in secondary grafts resulted in no differential increase in repopulating ability of previously dmPGE₂-treated HSCs (Figure 2B, right). This result suggests that the enhanced engraftment of dmPGE₂-treated HSCs in primary and subsequently serially transplanted recipients is a function of increased numbers of homed HSCs and their subsequent survival/self-renewal, resulting in a transient increase in competitiveness, rather than a long-term increase in inherent HSC competitiveness. Thus, short-term treatment of hematopoietic grafts before transplant is a strategy to enhance HSC engraftment without concern for long-term alteration of normal HSC function.

These studies were supported by grants from the National Institutes of Health (NIH), National Heart, Lung and Blood Institute (HL096305); NIH training grants (DK07519, HL07910, and HL087735) (J.H.); and a Center of Excellence in Hematology (P01 DK090948) for additional core support. Flow cytometry was performed in the Flow Cytometry Resource Facility of the Indiana University Simon Cancer Center (National Cancer Institute P30 CA082709).

Contribution: J.H. designed and executed all described studies, analyzed data, and wrote the manuscript; K.S.M. performed pathologic analysis of bone marrow sections from serially transplanted mice; P.S. performed experiments and analyzed data; and L.M.P. designed and executed experiments, analyzed data, and wrote the manuscript.

Conflict-of-interest disclosure: J.H. and L.M.P. have received consulting fees from Fate Therapeutics. The remaining authors declare no competing financial interests.

Correspondence: Louis M. Pelus, Department of Microbiology and Immunology, Indiana University School of Medicine, 950 West Walnut St, Indianapolis, IN 46202; e-mail: lpelus@iupui.edu.