Sex-dependent niche responses modulate steady-state and regenerative hematopoiesis

Abstract

Hematopoietic stem cells (HSCs) adapt to organismal blood production needs by balancing self-renewal and differentiation, adjusting to physiological demands and external stimuli. Although sex differences have been implicated in differential hematopoietic function in males versus females, the mediators responsible for these effects require further study. Here, we characterized hematopoiesis at a steady state and during regeneration following hematopoietic stem cell transplantation (HST). RNA sequencing of lineage(–) bone marrow cells from C57/Bl6 mice revealed a broad transcriptional similarity between the sexes. However, we identified distinct sex differences in key biological pathways, with female cells showing reduced expression of signatures involved in inflammation and enrichment of genes related to glycolysis, hypoxia, and cell cycle regulation, suggesting a more quiescent and less inflammatory profile compared with male cells. To determine the functional impacts of the observed transcriptomic differences, we performed sex-matched and mismatched transplantation studies of lineage(–) donor cells. During short-term 56-day HST recovery, we found a male donor cell proliferative advantage, coinciding with elevated serum TNF-α, and a male recipient engraftment advantage, coinciding with increased serum CXCL12. Together, we show that sex-specific cell responses, marked by differing expression of pathways regulating metabolism, hypoxia, and inflammation, shape normal and regenerative hematopoiesis, with implications for the clinical understanding of hematopoietic function.

Hematopoietic stem cells (HSCs) are responsible for producing blood and bone marrow (BM) cells throughout life, maintaining a balance between differentiation and self-renewal influenced by their own gene regulation and the BM environment.^1,2 In healthy adults, most HSCs remain quiescent under hypoxic conditions, consequent to the chemical signaling and cell-cell interaction with non-hematopoietic cell types, ensuring HSC genomic stability while limiting replicative stress and exhaustion.^1–3 Conversely, during stress triggered by infection, bleeding, illness, injury, aging, or other exogenous factors,^4–6 hematopoietic regeneration occurs wherein HSCs receive signals to replenish depleted populations and restore homeostasis.

Sex differences are defined as biological variations in gene expressions on X and Y chromosomes, sex hormones, and reproductive organ functions between males and females.⁷ A growing literature suggests a crucial role for sex differences in the mechanisms of hematopoietic regulation, a hypothesis partially spurred by the male predisposition to and worse prognosis for most hematological malignancies.^8–10 Sex differences have been attributed to variances in circulating progenitor cell frequencies¹¹ and hormone influences on HSC maintenance,^12–15 with sex-mismatched hematopoietic stem cell transplantation (HST) clinical outcomes further underscoring sex-specific responses to hematopoietic stress.^16,17 Here we characterize murine hematopoiesis and employ in vivo transplantation assays to examine the functional consequences of steady-state sex differences in the context of transplant-induced stress conditions.

METHODS

Animals

Approximately 8- to 12-week-old male and female C57/Bl6 mice (Jackson Laboratories) were housed in the AAALAC-accredited facilities of the Case Western Reserve University School of Medicine under IACUC protocol 2019-0065.

Complete Blood Counts

Peripheral blood was collected into Microtainer EDTA (ethylenediaminetetraacetic acid) tubes (Becton–Dickinson) by submandibular cheek puncture. CBCs (complete blood counts) were analyzed using a Hemavet 950 FS hematology analyzer.

Quantification of Hematopoietic Stem and Progenitor, and Myeloid and Lymphoid Cells

Total BM cells were obtained by flushing hindlimb bones into FACS (fluorescence-activated cell sorting; PBS (phosphate-buffered saline) + 2% FBS (fetal bovine serum)) buffer. Cells were stained and then fixed in 2% PFA (paraformaldehyde) before being resuspended in 200 μL of FACS buffer. Antibodies are detailed in Supplementary Methods. Data were acquired on an LSRII flow cytometer (BD Biosciences) using FSC/SSC gating to differentiate live cells. Analysis was performed on FlowJo software (TreeStar).

Cell Separations, RNA Extraction, and qPCR

Lineage(–) cells were collected using the mouse lineage cell depletion kit (Miltenyi Biotec). RNA was extracted using the RNeasy Plus Mini Kit (QIAGEN) with on-column gDNA exclusion. Isolated RNA concentration and integrity were evaluated with a TapeStation Instrument (Agilent). Details on qPCR (quantitative polymerase chain reaction) are found in Supplementary Methods.

RNA Sequencing

Samples were shipped on dry ice to Azenta for library preparation and PE150 sequencing on Illumina. Data were analyzed in collaboration with colleagues at the Cleveland Clinic Center for Immunotherapy and Precision Immuno-Oncology’s Platform Immunology Group. The raw sequencing data from this study are available in the NCBI GEO database, under accession number [GEO: GSE253024]. Detailed methodology for RNAseq analysis can be found in Supplementary Methods.

Bone Marrow Transplantation

Recipient mice were exposed to 10 Gy total body irradiation from a cesium source. After 16–24 hours, mice received 1e5 lineage(–) BM cells by retroorbital injection.

Serum Cytokine Quantification

Serum samples were collected in BD Microtainer tubes (Becton–Dickinson) by submandibular cheek puncture. Cytokines were measured in the Translational Research Core of the Case Comprehensive Cancer Center using the Ella Automated Multiplex ELISA system (ProteinSimple). Additional details are found in Supplementary Methods.

Statistical Analysis

Values were graphically presented, using GraphPad Prism for analysis. Group comparisons were made with unpaired two-tailed Student’s t-tests, except for peripheral blood and BM recovery kinetics, which employed two-way analysis of variance.

RESULTS AND DISCUSSION

Female BM Lineage(–) Cells are Transcriptionally More Quiescent and Less Inflammatory

To identify potential sex-specific differences in the cellular composition of the three major hematopoietic tissues of young C57/Bl6 mice, we characterized the peripheral blood, splenic, and BM compartments. We observed no significant differences in peripheral blood outside of a moderate but insignificant increase in male neutrophil (NE) counts (Supplementary Figure E1A). We found insignificant differences in splenic cellularity and LSK cell frequencies, and a trend toward increased BM LSK cell frequencies in males (Supplementary Figure E1B, C).

We analyzed CD45(–) nonhematopoietic BM gene expression and observed a significant decrease in male Ptn (fourfold decrease), suggesting a greater capacity for HSC mobilization coupled with decreased self-renewal.¹⁸ Significant yet minor reductions in Tgfb1, Il6, Vcam1, and Angpt1 further highlight a unique marrow microenvironment between sexes^1,19,20 (Supplementary Figure E2).

To dissect intrinsic sex differences in HSPCs, we performed RNA sequencing on purified lineage(–) BM cells (Figure 1A). Principal component analysis and volcano plots indicated no obvious stratification between male and female samples (Supplementary Figure E3A, B). However, the top 50 differentially expressed genes (Supplementary Figure E3C) and gene set enrichment analysis revealed significant variations in key biological pathways. Notably, female lineage(–) cells displayed decreased expression of pathways involved in inflammatory responses, such as interferon-gamma response and TNF-α signaling via NF-κB, as well as pathways linked to metabolic activity, including oxidative phosphorylation and reactive oxygen species pathways. Additionally, G2M checkpoint and unfolded protein response pathways were enriched in female cells, suggesting tighter cell cycle regulation and enhanced ability to resolve protein-induced HSC stress.^21,22 Female cells also exhibited an upregulation of signatures regulating hypoxia and glycolysis, aligning with the energy production preference of quiescent HSCs in low-oxygen environments³ (Figure 1B, C). Thus, although the broad transcriptional landscape remains similar between sexes, we identify distinct traits of female lineage(–) cells: they tend toward a more quiescent program and exhibit a reduced inflammatory profile compared with their male counterparts.

Figure 1.

Female BM lineage(–) cells are transcriptionally more quiescent and less inflammatory. (A) Schematic depicts lineage(–) cell isolation and subsequent RNA sequencing. (B) Gene set enrichment analysis (GSEA) from a list of DEGs to determine statistically significant, concordant differences between two biological states. Fifty well-defined biological pathways are shown. GSEA shown is enriched or depleted in females relative to males. Red color represents statistically significant enrichment; blue color represents enrichment that is trending toward significance. (C) Heatmaps of genes involved in inflammatory response, unfolded protein response, hypoxia, and glycolysis. N = 5 mice/arm. Schematic made using BioRender.com. BM = Bone marrow; DEG = Differentially-expressed gene.

Male Mice Display Accelerated Rates of Peripheral Recovery Following Sex -Matched HST

We next investigated if the observed sex-specific gene expression in lineage(–) cells correlated with functional behavior by tracking hematopoietic recovery posttransplant in sex-matched groups (F→F and M→M), sacrificing mice at days 14, 28, 42, and 56 after transplanting 1e5 lineage(–) cells into lethally irradiated recipients (Figure 2A). We observed significant differences in the recovery of the peripheral blood, with M→M mice showing accelerated recovery of WBC, LYMPH, NE, and RBC counts compared with the F→F mice (Figure 2B). In the BM we observed significantly accelerated recovery in M→M cellularity, coupled with trends toward increased male LSK populations through D42, LT-HSCs through D28, and minor differences in the recovery of lymphoid B220+, CD3+, and myeloid CD11b+ cells (Figure 2C, Supplementary Figure E4A). Multiplex ELISA on serum from mice at D14 and D42 (representing early and intermediate recovery stages) showed differential expression of CXCL12 and TNF-α between M→M and F→F cohorts (Figure 2D; additional factors found in Supplementary Figures E4B and E5B). CXCL12, recognized for its role in HSC homing and HSC engraftment,²³ was significantly elevated at D14 and D42 in the M→M cohort. TNF-α, which has been demonstrated to exert pro-survival effects on HSCs by inhibiting necroptosis and predisposing cells toward myeloid differentiation,²⁴ was significantly increased at D14 in M→M versus F→F mice. These data demonstrate notable advantages in peripheral recovery for M→M mice while suggesting correlative roles for CXCL12 and TNF-α toward that recovery.

Figure 2.

Male mice display accelerated rates of peripheral recovery following sex-matched HST. (A) Schematic depicting transplant conditions. 1e5 lin(–) BM cells from male donors were transplanted into lethally irradiated (10 Gy) male recipients, and the equivalent was performed using female donors and recipients. (B) Complete blood count analysis of circulating white blood cells (WBC) lymphocytes (LYMPH), neutrophils (NE), red blood cells (RBC), and platelets (PLT) in male (blue) and female (purple) mice at indicated time points post–bone marrow transplantation. (C) Cytometric analysis of total BM cellularity, BM LSKs, and BM LT-HSCs at indicated time points post–bone marrow transplantation. N = 10 mice/arm over two replicates. Error bars represent SEM. Repeated measure ANOVA performed on all analyses. (D) Serum concentrations (pg/mL) of CXCL12 and tumor necrosis factor alfa (TNF-α) in male and female mice at days 14 and 42. N = 10 mice/arm. Error bars represent SEM. *p < 0.05. **p < 0.01. ***p < 0.001. Student’s t-test performed at each time point. Schematic made using BioRender.com. ANOVA = Analysis of variance; BM = bone marrow; HST = hematopoietic stem cell transplantation; SEM = standard error of mean.

Male Donor Cells Accelerate Peripheral Hematopoietic Recovery During HST Independent of Recipient Sex

To understand the impact of donor cells versus the recipient microenvironment on recovery, we established reciprocal transplantations of male donors into female recipients (M→F) and female donors into male recipients (F→M) (Figure 3A). CBC analyses demonstrated significantly accelerated rates of peripheral blood WBC and LYMPH in M→F mice, with a transient increase in NE counts at D42 and a trend toward accelerated RBC recovery (Figure 3B). BM populations surprisingly showed an inverse rate of recovery, as the F→M mice cohort trended toward increased BM cellularity and LSK recovery (Figure 3C), akin to the M→M mice from Figure 3. Lymphoid B220 + recovery was slightly faster in M→F and CD3+ in F→M cohorts (Supplementary Figure E5A).

Figure 3.

Male donor cells accelerate peripheral hematopoietic recovery in HST independent of recipient sex. (A) Schematic depicting transplant conditions. 1e5 lin(–) BM cells from male donors were transplanted into lethally irradiated (10 Gy) female recipients, and the equivalent was performed using female donors and male recipients. (B) Complete blood count analysis of circulating white blood cells (WBC), lymphocytes (LYMPH), neutrophils (NE), red blood cells (RBC), and platelets (PLT) in male (red) and female (blue) mice at indicated time points post–bone marrow transplantation. (C) Cytometric analysis of total BM cellularity, BM LSKs, and BM LT-HSCs at indicated time points post–bone marrow transplantation. N = 10 mice/arm over two replicates. Error bars represent SEM. Repeated measure ANOVA performed on all analyses. (D) Serum concentrations (pg/mL) of CXCL12 and tumor necrosis factor alfa (TNF-α) in male and female mice at days 14 and 42. N = 10 mice/arm. Error bars represent SEM. *p < 0.05. **p < 0.01. ***p < 0.001. ****p < 0.0001. Student’s t-test performed at each timepoint. Schematic made using BioRender. ANOVA = Analysis of variance; BM = bone marrow; HSC = hematopoietic stem cells; SEM = standard error of mean.

Multiplex ELISA data on D14 and D42 serum samples revealed higher CXCL12 levels in F→M at D14, trending higher at D42, whereas the M→F cohort contained higher TNF-α at both time points (Figure 3D). The elevated CXCL12 in the F→M mice, in context with the accelerated BM recovery in this cohort, suggests that the male microenvironment may achieve increased CXCL12-mediated homing to the BM niche. The elevated TNF-α levels in the M→F group, like the levels seen in the M→M cohort from Figure 3, suggest that male lineage(–) cells in the context of HST are associated with increased TNF-α-driven peripheral recovery following HST. The data in Figure 3B–D, complemented with that of Figure 2B–D, indicates the male microenvironment enhances BM engraftment and male lineage(–) cells have a higher proliferative capacity in short-term recovery.

Together, this work identifies sex-specific cellular and molecular responses, including distinct transcriptomic profiles and regeneration capabilities in hematopoiesis of C57/Bl6 mice. We do acknowledge several limitations, including the need to investigate cause–effect relationships for the associations uncovered here, whether these can be extrapolated to other strains and species, and possible confounding factors derived from the analysis of heterogenous cell populations or from the effects of conditioning.

However, we posit that these findings may inform clinical observations demonstrating that male HSCs show an earlier functional decline and higher cancer susceptibility than females,^7–9,25,26 and strengthen the idea that targeting inflammatory signaling pathways could be useful in the context of sex-specific treatment.^4,5

HIGHLIGHTS.

• Sex dimorphism is present both in steady-state and regenerative hematopoiesis.

• Female lin(–) cells show a more quiescent and less inflammatory bone marrow gene signature.

• Male donor cells promote accelerated short-term transplant recovery.

• Male recipient microenvironment promotes increased short-term engraftment.