Ovariectomy expands murine short-term hemopoietic stem cell function through T cell expressed CD40L and Wnt10B

Abstract

Estrogen deficiency expands hemopoietic stem and progenitor cells (HSPCs) and mature blood lineages, but the involved mechanism and the affected HSPC populations are mostly unknown. Here we show that ovariectomy (ovx) expands short-term HSPCs (ST-HSPCs) and improves blood cell engraftment and host survival after bone marrow (BM) transplantation through a dual role of the T-cell costimulatory molecule CD40 ligand (CD40L). This surface receptor is required for ovx to stimulate T-cell production of Wnt10b, a Wnt ligand that activates Wnt signaling in HSPCs and stromal cells (SCs). Moreover, CD40L is required for ovx to increase SC production of the hemopoietic cytokines interleukin (IL)-6, IL-7, and granulocyte macrophage-colony-stimulating factor. Attesting to the relevance of CD40L and Wnt10b, ovx fails to expand ST-HSPCs in CD40L-null mice and in animals lacking global or T-cell expression of Wnt10b. In summary, T cells expressed CD40L, and the resulting increased production of Wnt10b and hemopoietic cytokines by T cells and SCs, respectively, plays a pivotal role in the mechanism by which ovx regulates hemopoiesis. The data suggest that antiestrogens may represent pharmacological targets to improve ST-HSPC function through activation of the microenvironment.

Figure 1

Effect of ovx on HSPC expansion in T cell–replete and T cell–deficient mice. (A-C) Effects of ovx on the relative frequency of LSK cells in WT mice, TCRβKO mice, and TCRβKO mice previously subjected to adoptive transfer of T cells. Lin⁻ cells were gated and analyzed for Sca-1 and c-Kit expression using isotype control settings. (Left) Representative flow cytometric dot plots from 1 mouse per group. The black box delineates c-Kit⁺ Sca-1⁺ cells. Parent population is Lin⁻. Data are expressed as percentage of total Lin⁻ cells. (Right) Mean ± standard error of the mean (SEM) for each group. Data are expressed as percentage of total BM mononucleated cells (BMMCs). (D) Effect of ovx on the number of BM B cells, monocytes, erythroid cells, and granulocytes. n = 10 mice per group. *P < .05 and **P < .01 compared with the corresponding sham-operated group.

Figure 2

Effect (mean + SEM) of ovx on the expansion of ST-HSPCs and LT-HSPCs in T cell–replete and T cell–deficient mice. (A-C) Effects of ovx on the relative frequency of CD150⁻CD48⁻ LSK cells (ST-HSPCs/MPPs) and CD150⁺CD48⁻ LSK cells (LT-HSPCs) in WT, TCRβKO, and TCRβKO mice previously reconstituted with T cells. (Left) Representative flow cytometric dot plots from 1 mouse per group using the signalling lymphocyte activation molecule receptors CD150 and CD48. Parent population is Lin⁻Sca1⁺c-Kit⁺. The upper boxes delineate LT-HSPCs. The lower boxes delineate ST-HSPCs + MPPS. (Right) Mean + SEM for each group. Data are expressed as percentage of BMMCs. n = 10 mice per group. (D-I) Effect of ovx on peripheral blood cell expansion after primary competitive repopulation. The panels show the percentage of CD45.2⁺ myeloid cells (CD11b⁺), granulocytic cells (GR-1⁺), and B lineage cells (B220⁺) in the peripheral blood of lethally irradiated WT recipient mice that received CD45.2⁺ BM donor cells mixed in a 1:2 ratio with CD45.1⁺ competitor BM cells. CD45.2⁺ BM cells were obtained from WT, TCRβKO, and reconstituted TCRβKO mice subjected to sham operation or ovx 2 weeks earlier. CD45.1⁺ BM cells were obtained from intact WT mice. *P < .05 compared with the corresponding sham-operated group.

Figure 3

Effects of ovx and of the antiestrogen ICI 182780 on the survival of lethally irradiated WT mice transplanted with a limiting number of BM cells derived from WT mice, TCRβ^−/− mice, and TCRβ^−/− mice previously subjected to adoptive transfer of T cells. (A-C) Donor mice were intact. Recipient mice were sham operated or ovx 2 weeks before BM transplantation. (D-F) Donor mice were sham operated or ovx 2 weeks before BM transplantation. Recipient mice were intact. (G) Donor mice were untreated. Recipient mice were treated with vehicle or ICI182780 (100 μg/mouse subcutaneously, twice a week for 4 weeks) starting at the time of transplantation. (H) Donor mice were treated with vehicle or ICI182780. Recipient mice were untreated. n = 10 per group.

Figure 4

Analysis of the effects (mean + SEM) of ovx in WT and CD40L^−/− mice and TCRβ^−/− mice previously reconstituted with CD40L^−/− T cells. (A) Effects of ovx on the relative frequency of BM LSK cells. (B-C) Effect of ovx on the relative frequency of ST-HSPCs/MPP and LT-HSPCs. (D) Effect of ovx on the number of BM B cells, monocytes, erythroid cells, and granulocytes. (E-G) Effect of ovx on peripheral blood cell expansion after primary competitive repopulation. In these experiments, CD45.2 CD40L^−/− and control WT mice were killed after 2 weeks of ovx or sham operation. BM was then mixed with BM from intact CD45.1⁺ WT mice at a ratio of 1:2 (donor/ competitor) and injected into lethally irradiated CD45.1⁺ recipient mice. The percentage of CD11b⁺, GR-1⁺, and B220⁺ cells in the peripheral blood of lethally irradiated WT recipient mice are shown. Recipient mice received CD45.2⁺ BM donor cells from WT and CD40L^−/− mice previously subjected to sham operation or ovx mixed in a 1:2 ratio with CD45.1⁺ competitor BM cells from intact WT mice. (H-K) Survival analysis of WT mice transplanted with limiting number of BM cells derived from WT and CD40L^−/− mice. (H-J) Donor mice were intact WT or CD40L^−/− mice. Recipient mice were WT mice subjected to sham operation or ovx 2 weeks before the BM transplantation. (I-K) Donor mice were sham operated or ovx 2 weeks before the BM transplantation. Recipient mice were intact WT mice. n = 10 per group. *P < .05 compared with the corresponding sham-operated group.

Figure 5

Role of CD40L on the production of hemopoietic cytokines by SCs. (A) The SC expression of IL-6, IL-7, and GM-CSF mRNAs was measured in SCs cultured alone or in the presence of T cells, anti-CD40L Ab, or rWnt10b. (B) Levels of IL-6, IL-7, and GM-SCF in the culture media of SCs from sham-operated and ovx WT mice, TCRβKO mice, and TCRβKO mice previously subjected to adoptive transfer of WT T cells. (C-E) Levels of IL-6, IL-7, and GM-SCF as measured by enzyme-linked immunosorbent assay in the 48-hour culture media of SCs purified from sham operated and ovx WT and CD40L^−/− mice. The first 4 bars to the left show data from WT and CD40L^−/− mice. The last 3 bars to the right show data from WT mice treated with irrelevant Ab or MR-1 Ab. In these experiments, BM was cultured for 1 week, SCs were then purified, and cytokine was measured by enzyme-linked immunosorbent assay in the 48-hour culture media. (F) Effect of ovx on the mRNA expression of Wnt10b in BM T cells. The first 4 bars to the left show WT and CD40L^−/− mice. The last 4 bars to the right show WT mice treated with irrelevant Ab or MR-1 Ab. *P < .05 compared with the corresponding sham-operated group.

Figure 6

Analysis of the effects (mean + SEM) of ovx in WT and Wnt10b^−/− mice. (A) Effects of ovx on the relative frequency of BM LSK cells. (B-C) Effect of ovx on the relative frequency of ST-HSPCs/MPP and LT-HSPCs. (D-F) Effect of ovx on peripheral blood cell expansion after primary competitive repopulation. The percentage of CD11b⁺, GR-1⁺, and B220⁺ cells in the peripheral blood of lethally irradiated WT recipient mice are shown. CD45.2⁺ WT mice and Wnt10b^−/− mice were killed 2 weeks after ovx or sham operation. Their BM was then mixed with BM from intact CD45.1⁺ WT mice at a ratio of 1:2 and injected into lethally irradiated CD45.1⁺ host mice. (G-J) Survival analysis of WT mice transplanted with limiting number of BM cells derived from WT and Wnt10b^−/− mice. (G-H) Donor mice were intact WT or Wnt10b^−/− mice. Recipients were WT mice subjected to sham operation or ovx 2 weeks before transplantation. (I-J) Donor mice were WT mice subjected to sham operation or ovx 2 weeks before transplantation. Recipient mice were intact WT mice. n = 10 in each group. *P < .05 compared with the corresponding sham-operated group.

Figure 7

Effect (mean + SEM) of ovx on the SC expression of mRNA of genes known to be up-regulated by Wnt signaling. BM harvested at death was cultured for 1 week. SCs were purified, and mRNA levels were determined by real-time reverse transcription-polymerase chain reaction. SCs were obtained from (A) WT mice, (B) TCRβKO mice, (C) TCRβKO mice previously subjected to adoptive transfer of WT T cells, (D) CD40L^−/− mice, (E) TCRβ^−/− mice previously reconstituted with CD40L^−/− T cells, (F) Wnt10b^−/− mice, and (G) TCRβKO mice previously subjected to adoptive transfer of Wnt10b^−/− T cells. The Wnt-dependent genes analyzed were aryl-hydrocarbon receptor (Ahr), axin2, cystein-rich protein 61 (Cyr61), naked cuticle 2 homolog (Nkd2), transgelin (tagln), transforming growth factor β 3 (TGFβ3), thrombospondin 1 (Thbs1), Twist gene homolog 1 (Twst1), and Wnt1 inducible signaling pathway protein 1 (Wisp1). n = 5 mice per group. *P < .05 compared with the corresponding sham-operated group.