LPS stimulation reveals male and female dimorphism in human iPSC-derived macrophages

Summary

Clinical disparities in the inflammatory responses between males and females have been reported for decades, revealing sex-biased disease severity and patient outcomes. To better understand the intrinsic biological sex differences contributing to these disparities, we characterized the functional and transcriptomic responses of induced pluripotent stem cell (iPSC)-derived macrophages from male and female donors under control and inflammatory conditions using the bacterial antigen lipopolysaccharide (LPS). Bulk RNA sequencing identified 67 genes differentially expressed between male and female macrophages in the absence of inflammation, the majority of which mapped to the sex chromosomes. This sexual dimorphism was amplified by LPS stimulation, with genes predominantly located on autosomes, leading to divergent patterns in secreted protein production that were consistent with clinical findings following LPS exposure. Although future validation with a larger patient cohort is needed, these data collectively suggest that iPSC-derived macrophages can serve as a meaningful tool for understanding and improving treatment of sex-biased pathophysiological conditions.

Graphical abstract

Highlights

• •RNA sequencing reveals sex-specificity in male and female iPSC-derived macrophages
• •Differentially expressed genes at baseline map predominately to sex chromosomes
• •LPS stimulation causes sex-specific protein production responses by iMacs

Immunology; Cell biology

Introduction

Studies of sexual dimorphism are becoming increasingly important amid a growing body of evidence of significant biological sex differences in the context of disease and injury.^1,2,3 In particular, sexual dimorphism in the immune system has been recorded in both the innate and adaptive immune response as detailed in two excellent reviews.^4,5 In adults, women generally demonstrate stronger immune responses than males, resulting in benefits like faster pathogen clearance and improved vaccine efficacy, providing greater protection against infectious diseases.⁴ However, women experience higher rates of autoimmune disease, with 80% of all autoimmune diseases occurring in women.⁶

Male and female dimorphism has been documented in several immune cell populations, though emerging evidence in mice suggests that macrophages exhibit significant transcriptional sexual dimorphism compared to other immune cells (granulocytes, T cells, and natural killer cells).⁷ Macrophages serve as key contributors to early and late-stage innate immune defense, tissue repair, and homeostasis,⁸ and are well-known drivers of cell-mediated immunity through T cell antigen presentation and inflammatory cytokine release. Consequently, macrophages play a critical role in the development and progression of many pathological conditions, including infectious diseases, diabetes, cancers, and chronic inflammatory diseases.^8,9 Importantly, male and female macrophages derived from human peripheral blood have been shown to display functional differences in disease contexts and responses to stimuli, including migratory capacity, phenotypic polarization, and chemokine signaling,^10,11 which may impact tissue health and repair. Although this pattern of clinical dichotomy is becoming increasingly acknowledged, the mechanisms at play are not well understood due to various biological (e.g., chromosomal complement, hormone status, and age) and environmental factors (e.g., nutrition, exercise, and air pollution, etc.) at play.^{12,13,14,15,16} Investigating the influence of each factor contributing to biological sex-based differences could lead to better mechanistic understanding of diseases, increased efficacy and safety of drug development, and inform targeted sex-specific treatment design.

While pre-clinical animal models and in vitro assays using primary human cells shed light on biological effects of sex on the immune system in a controlled environment, these models are subject to a number of limitations. Both small and large animal models are limited by species-specific differences.^17,18 Also, the utility of primary human macrophages obtained from whole blood is hindered by limited cell numbers per donor, significant donor variability, and finite in vitro culture times. Primary cells have been conditioned within the body of the donor, and thereby exposed to confounding factors that influence specificity, including sex hormones and systemic inflammatory signals. Alternatively, human induced pluripotent stem cell (iPSC)-derived macrophages (iMacs) offer an accessible, reproducible, and scalable source to study human biological sex differences in macrophages.¹⁹ iMacs are cultured entirely in vitro, isolated from confounding effects present clinically; as a result, biological sex-specific effects of the chromosomes can be isolated, allowing further insight into molecular-level differences among macrophages between sexes removed from lifelong environmental and hormonal stimuli.

Here, we investigate biological sex-specific differences in human macrophages under control and inflammatory conditions using iMacs from male and female donors. To simulate an inflammatory environment, we conditioned macrophages with lipopolysaccharide (LPS), a potent macrophage activator that initiates an acute inflammatory response through toll-like receptor 4 (TLR4) signaling.^20,21,22 Through this analysis, we identified intrinsic biological sex-specific differences in human iMacs and uncovered sexually dimorphic cell states that arise during inflammation.

Results

To investigate sexual dimorphism in human macrophages, iPSC lines were obtained from three healthy male donors (WTC11, AA1, and FA10), ages 27–36, and three healthy female donors (BS2, CC1, and FA11), ages 20–49 years old (Figures 1 and S1). iPSCs were differentiated into macrophages (iMacs) using previously published protocols,^23,24 as depicted in Figure 1. To capture sex-specific traits exhibited by iMacs solely due to differences in cellular genetics, a serum-free medium was used for all culture conditions to limit exposure to hormones. iMacs were cultured in the presence of 100 ng/mL LPS to elicit an inflammatory response, and macrophage activation was compared to resting iMacs in terms of bulk RNA sequencing, functionality, and secreted protein production. The results were stratified by donor biological sex to discern sex-specific disparities.

Figure 1.

Overview of iPSC-macrophage generation and experimental design

iPSCs were derived from apparently healthy 3 male and 3 female donors between the ages of 20 and 49 years old. iPSCs were induced to mesoderm, then differentiated into hematopoietic stem and progenitor cell (HSPC)-producing hemogenic endothelium. Differentiation of monocytes from hemogenic endothelial cultures was induced by stimulation with IL-3 and macrophage colony-stimulating factor (M-CSF). Monocytic-specific lineage cells were then collected over multiple weeks in the suspension fraction (once a week) and further cultured with M-CSF for at least 5 days to achieve macrophage differentiation. The resulting iPSC-macrophages of each line were subsequently cultured in the absence (control) or presence of LPS and analyzed for differences between the sexes in terms of gene expression, protein production, and functionality.

Characterization of iMacs from male and female donors

Prior to LPS stimulation, the identity and functionality of all iMacs were characterized. Surface marker analysis indicated expression of CD11b, CD14, HLA-DR, and SIRP-⍺ for all male and female derived iMacs (Figure 2A), confirming iPSC differentiation into a myelomonocyte lineage. Macrophage differentiation was further identified by the expression of pan-macrophage marker CD68 via immunohistochemistry (Figure 2B). iMac functionality was subsequently assessed by quantifying phagocytosis, a key macrophage contribution to homeostatic maintenance and injury response. Flow cytometry revealed co-expression of CD45⁺ macrophages with fluorescently tagged latex beads (Figure 2C), indicating successful phagocytosis of latex beads as a proxy for dead cells. There were no statistical differences in bead uptake between male and female iMacs, which was consistent with our findings for primary human monocyte-derived macrophages, supporting the use of iMacs for studying biological sex-driven differences.

Figure 2.

Characterization of male and female iPSC-macrophages

(A) Surface marker validation confirms male and female iMacs showing CD11b, CD14, HLA-DR, and SIRP-⍺ expression.

(B) Immunofluorescence staining of iPSC-macrophages showing expression of DAPI (blue), Phalloidin (green), and CD68 (magenta). Representative images shown for n = 3 replicates per cell line; scale bars, 50 μm.

(C) Uptake of latex beads, confirming phagocytic functionality of iMacs (left) in comparison to primary human peripheral blood monocyte-derived macrophages. Quantification of uptake shows no differences between male and female macrophages, both in iMacs and primary monocyte-derived macrophages. Data represent mean ± SD and were assessed via two-tailed t test.

iMacs exhibit sexually dimorphic transcriptomic profiles

Bulk RNA sequencing analysis was performed to comprehensively delineate the differences between resting male and female macrophages. Principal component analysis (PCA) revealed tight clustering of male iMacs across all donors and technical replicates, and large variability among iMacs derived from female donors (Figure 3A). Although there was no clear separation between resting male and female macrophages, PC1 and PC2 accounted for only 29.9% of the total variation between samples, suggesting other sources of variance within the data, including inter-donor variability. In particular, we observed large variability across iMacs from female lines, which may be attributed to hormonal contributions, as the menopausal status of these patients was unknown. Hierarchical clustering of the top 1,000 genes similarly showed tight clustering between technical replicates, with separation between male and female iMacs (Figure 3B). These data suggest that the differences are driven to a greater extent by biological sex rather than donor.

Figure 3.

Functional and bulk transcriptomic analysis of male versus female iPSC-macrophages at baseline

(A) PCA plot of resting male and female iMacs from bulk RNA-seq data.

(B) Hierarchical clustering of samples.

(C) Top impacted KEGG pathways for genes differentially expressed in males compared to female iMacs.

(D) Chord graph mapping differentially expressed genes to their chromosome location, showing distribution across the chromosomes.

(E) Quantitative analysis of cytokine production for resting male and female iMacs.

(F) Chemotaxis assay shows chemotaxis of male and female iMacs in response to CXCL-12 chemoattractant. (E and F) Data represent mean ± SD and were assessed via two-tailed t test; ∗p < 0.05.

To identify the biological processes affected by these genetic differences, Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment analysis were performed for differentially expressed genes (false discovery rate [FDR] <0.05, |log2(foldchange)| > 0.6). Top KEGG pathways with significantly differentially expressed genes included: “ferroptosis”, “complement and coagulation cascades”, “HIF-1 signaling pathway”, “cell cycle”, “regulation of actin cytoskeleton”, “apoptosis”, “PI3K-Akt signaling pathway”, and “metabolic pathways” (Figure 3C). GO pathway analysis complemented these findings, and included several additional pathways of interest (Figure S2A). For example, “cellular response to stress”, “regulation of macrophage activation”, and “regulation of cytokine production” were significantly over-represented.

We next mapped the differentially expressed genes to their respective chromosomes and found that >50% mapped to the sex chromosomes (Figure 3D), suggesting that the biological sex-based clustering of iMacs is primarily driven by X- and Y-linked gene expression.

To determine if the sexual dimorphism observed at the genetic level translates to functional changes, we compared cytokine production from male and female iMacs. Notably, resting female iMacs produced significantly greater monocyte chemoattractant protein (MCP)-1 relative to male iMacs (Figure 3E). Given that MCP-1 is a major chemotactic and activating factor of monocyte and macrophage recruitment during inflammation, this difference may have important clinical implications. Though there were no other statistically significant differences observed between male and female iMac cytokine production at baseline (Figures 3E and S2B), female iMacs produced slightly elevated levels of interleukin (IL)-1β, IL-6, IL-10, and interferon (IFN)-⍺2. In contrast, male iMacs produced slightly higher levels of IL-8. Analysis of chemotaxis using a Boyden chamber assay revealed no significant differences in macrophage chemotaxis after 3 h in response to chemoattractant CXCL12 (Figure 3F), though migration of female iMacs was slightly greater compared to males.

LPS stimulation exacerbates iMac sexual dimorphism

We next sought to understand the impact of iMac sexual dimorphism during an acute inflammatory response, using LPS to induce inflammation. Bulk RNA sequencing analysis of male and female LPS-stimulated iMacs were compared to resting iMac controls. Visualization of the data using PCA showed clear separation between stimulated and resting iMacs along PC1, and subsequent clustering of female and male iMacs along PC2 (Figure 4A). Together, these PCs accounted for 39.3% of the variance within the data. Male and female samples were more clearly separated in response to LPS than in the controls, suggesting a divergence in the response to inflammation by male and female macrophages. Consistent with these findings, hierarchical clustering of the control and LPS-treated samples separated iMacs stimulated with LPS from all controls, and further separated male and female samples within each condition (Figure 4B).

Figure 4.

Divergent male and female iMacs inflammatory responses to LPS

(A) PCA plot of resting control and LPS-stimulated male and female iMacs.

(B) Hierarchical clustering of all control and LPS-treated samples.

(C) Top impacted KEGG pathways of genes that change in response to LPS treatment differently in male versus female macrophages when compared to controls.

(D) KEGG pathway diagram of chemokine signaling pathway, showing the total projected perturbation across the pathway, combining observed differentially expressed genes between male and female LPS-treated iMacs when compared to controls, which are displayed beneath the map, and calculated downstream effects along the pathway, generated by AdvaitaBio iPathwayGuide.

(E) Quantitative analysis of cytokine production for LPS-treated male and female iMacs normalized to resting cell controls. Data represent mean ± SD and were assessed via two-tailed t test; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

To evaluate underlying causes of iMac sexual dimorphism during an acute inflammatory response, we identified sexually unique gene changes in response to LPS treatment between male and female iMacs, while accounting for differences in the controls. Specifically, we identified genes that either significantly changed in opposite directions between LPS-stimulated male and female iMacs (FDR ≤0.05 for both comparisons), or genes that changed significantly (FDR ≤0.05) in male cells but remained comparable (p ≥ 0.1 with 95% confidence) in female cells. KEGG analysis of the gene set that met these criteria indicated many pathways that were significantly impacted, including “chemokine signaling”, “metabolic pathways”, “cytokine-cytokine receptor interaction”, “Th1 and Th2 cell differentiation”, “apoptosis”, and “ECM-receptor interaction” (Figure 4C). Given that the “chemokine signaling” pathway was the most significantly affected, we mapped the 16 differentially expressed genes in this pathway (KEGG: 04062) to the KEGG pathway diagram (Figure 4D). This diagram combines the measured expression of differentially expressed genes (Figure S3A) with the computed accumulated perturbation propagated across the pathway from upstream impacted genes (Figure S3B) to show the total projected perturbation throughout the pathway. This analysis revealed significant upregulation of CCL15 and CCL20 in the cytokine-cytokine receptor interaction, as well as significant upregulation of SOS1, SOS2, and BRAF, and downregulation of MAPK3. Notably, the downstream effect of these differentially expressed genes involves perturbation of the Jak-STAT and MAPK signaling pathways, which would be expected to ultimately drive increased cytokine production, migration, and apoptosis.

We validated these findings at the protein level by quantifying production of a panel of inflammatory cytokines (Figures 4E and S3C). As expected, LPS treatment enhanced protein production by both male and female iMacs for all cytokines tested. However, male and female iMacs displayed distinct cytokine profiles. Specifically, production of IL-1β, TNF-⍺, IL-6, IFN-⍺2, IL-10, and IL-23 was significantly greater in male iMacs compared to females, whereas female iMacs produced significantly more IFN-ɣ, IL-8, and IL-17A compared to males in response to LPS.

The diverging transcriptomic profiles, in combination with the observed differences in cytokine production, demonstrate that sexual dimorphism in macrophages is exacerbated during acute periods of inflammation. These findings warrant further validation with a larger patient cohort.

Mapping differentially expressed genes to sex chromosomes

Using the list of genes altered in response to LPS in males versus females, we again mapped the differentially expressed genes to their corresponding chromosomes. Interestingly, we observed a wide distribution of genes across all chromosomes (Figures 5A and S4A). In contrast to our findings in resting iMacs, in which the majority of differentially expressed genes were sex linked, we found that among 304 genes differentially expressed in male iMacs compared to females in response to LPS, only 18.1% mapped to sex chromosomes (Figure S4B). Notably, when accounting for the differences in resting iMacs, the fraction of sex-linked genes altered in response to LPS was <3%, with the greatest number of genes mapped to chromosome 1.

Figure 5.

Mapping differentially expressed genes to sex chromosomes

(A) Chord diagram illustrating chromosome mapping of genes differentially expressed between male and female LPS-stimulated iMacs compared to the controls.

(B) Genes located on the sex chromosomes that are differentially expressed in male LPS-stimulated iMacs versus controls compared to females.

(C) Immunofluorescence staining of iPSC-macrophages for all 6 lines showing expression of DAPI (blue) and NOX1 (magenta) at baseline and in response to LPS stimulation. Representative images shown for n = 2 replicates per cell line per condition; scale bars, 100 μm.

(D) Quantification of NOX1 intensity per cell. Data represent mean ± SD and were assessed via two-Way ANOVA with Fisher’s LSD test; ∗p < 0.05.

(E) Significantly upregulated GO pathways for differentially expressed genes located on autosomes.

In response to LPS stimulation, we identified several genes that respond either in opposite directions in male and female iMacs compared to controls, or only in male but not in female iMacs. X-linked genes that were significantly upregulated in LPS-treated male iMacs compared to LPS-treated female iMacs included CSTF2, UXT, TMEM164, MBTPS2, HUWE1, and MECP2 (Figure 5B). X-linked genes significantly downregulated in male iMacs compared to female iMacs included NOX1, TSPYL2, SLC6A8, RTL5, MPP1, and TSR2, among others. In addition to these X-linked genes, male iMacs displayed altered expression of Y-linked genes not present in female iMacs, including significant upregulation of DDX3Y and downregulation of RPS4Y1.

We chose to further confirm the transcriptomic differences in NADPH oxidase NOX1 at the protein level based on prior evidence that oxidative stress is greater in males compared to females.²⁵ Immunofluorescence staining was performed to detect the presence of NOX1 in male and female iMacs at baseline and in response to LPS (Figure 5C). Compared to female iMacs, male iMacs produced modest but significantly greater levels of NOX1 during an inflammatory response (Figure 5D), consistent with reports of sex-specific elevated ROS production.²⁶

Since the majority of genes altered in male iMacs compared to female iMacs in response to LPS were not sex-linked, we aimed to understand the biological processes of the autosomal genes. To this end, we generated a list of genes differentially expressed in males that mapped to only chromosomes 1–22, and performed GO enrichment analysis. Among the significantly enriched pathways were “intracellular signal transduction”, “cellular response to stress”, and “negative regulation of apoptotic signaling pathway” (Figure 5E).

Taken together, these findings provide evidence of sexual dimorphism in iMacs during both homeostasis and inflammation, and reveal the combinations of sex-linked and autosomal genes underlying the measured differences. However, the generalizability of these findings is limited by the modest sample size used in this study and should be validated in future work.

Discussion

Sexual dimorphism is observed clinically across a multitude of pathophysiological conditions, leading to divergent biological sex-specific patterns in disease severity and outcomes. Historically, this divergence was considered a consequence of inherent differences arising from sex hormones. We are only recently beginning to appreciate the influence of genetic architecture, from genes not only on the sex chromosomes but also on the autosomes.²⁷ While this applies to all cells and tissues, the immune system is of particular interest given its central and systemic role in regulating human health. In this study, we probed the impact of biological sex on the response of iMacs, which regulate tissue repair and regeneration in both health and disease, and identified underlying transcriptomic differences between males and females that are present under steady state conditions and exacerbated in response to LPS.

Gene expression differences in iMacs under homeostatic conditions were predominately linked to sex chromosomes. Although these differences were limited to a small number of genes, functional assessment captured higher levels of MCP-1 production by female iMacs compared to male iMacs. Under baseline conditions, male and female macrophages had no significant migratory differences, consistent with a report by Campesi et al.,²⁸ which found that primary human monocytes isolated from female donors only exhibited increased chemotaxis when pre-treated with LPS, as compared to monocytes from male donors.

In response to LPS stimulation, divergence between male and female iMacs became much more pronounced, but the proportion of genes linked to chromosomes X or Y represented only a small percentage of all altered genes. The overwhelming proportion of differentially expressed genes located on autosomes is consistent with a meta-analysis reported by Mayne et al., which found that the biological sex-biased genes were predominantly autosomal for most tissues analyzed, and two-thirds of these genes were not under the direct influence of sex hormones.²⁹ Although these genes are not located on the sex chromosomes, it is possible that sex-linked transcriptional regulators confer epigenetic changes that contribute to these differences on autosomal genes.^30,31 Indeed, recent work uncovered hundreds of autosomal sex differences in DNA methylation using human whole blood.³²

These data suggest that the stimuli-responsive differences observed between males and females are predominately autosomal; nevertheless, the sex-linked genes include several notable functions, including epigenetic and post-translational regulation, nuclear factor kappa B (NF-κB) signaling, and production of reactive oxygen species. Consequently, many of these genes, including CSTF2 and MECP2, and NOX1 and SLC6A8, which were altered in opposite directions in male iMacs following LPS stimulation compared to female iMacs in this study, have been implicated in macrophage activation, differentiation,^33,34,35,36 and survival.³⁷ HUWE1, UXT, and TSR2 have been implicated in regulating NF-κB) activity^38,39,40 and downstream processes like apoptosis,⁴¹ though HUWE1 also impacts inflammasome activation and host defense against bacterial infection independently of NF-κB.⁴² We suspect that the collective differences in expression of these genes between males and females have important clinical consequences in regulating inflammation, though this warrants further investigation. Although we did not observe consistent patterns in up-regulation or down-regulation of genes in males or females with respect to specific gene functions, the diverging gene expression changes observed in LPS-stimulated macrophages described here offer potential study targets to improve understanding of biological sex-biased pathophysiological conditions.

At the transcript level, NOX1 was down-regulated in male iMacs in response to LPS and remained unchanged in female iMacs. This translated to elevated levels of NOX1 protein in male LPS-stimulated iMacs compared to females, which aligns with sex-specific differences in oxidative stress observed in many species, including humans. At both the transcript and protein levels, NOX1 was altered in male iMacs and remained unchanged in females. Nevertheless, the discrepancy between the directional changes of NOX1 mRNA and protein in males likely represents an early but transient spike in transcription that leads to accumulation of protein, as NOX1 associates with p22phox, NOXO1, and NOXA1, which stabilize the protein by protecting it from ubiquitination.^43,44

The observed differences in cytokine production between male and female iMacs in response to LPS are consistent with other reports demonstrating increased endotoxin-induced production of IL-1β, TNF-⍺, IL-6, and IL-10 in males and IFN-γ and IL-8 in females.^45,46,47 High levels of endotoxin activity are associated with critically ill patients, such as those with COVID-19,⁴⁸ who present with elevated plasma levels of cytokines and chemokines, including IFN-γ, IL-1β, IL-6, IL-8, IL-10, IL-17, TNF-α, and MCP-1.⁴⁹ Significant biological sex-based differences in immune cell composition and phenotype, along with cytokine and chemokine production, underlie COVID-19 disease progression.⁵⁰ Persistently elevated serum levels of IL-6 in males following hemorrhagic shock injury are associated with a higher incidence of multiple organ failure compared to females.⁵¹ Similarly, disproportionate disease severity and mortality rates in men following sepsis have been well documented for many years.^52,53,54 Collectively, these findings support the use of iMacs as a model for interrogating the impact of biological sex on endotoxin-related inflammatory disease. This model can be applied to a multitude of inflammatory diseases with clinical sex disparities, including autoimmune disorders,⁵⁵ obesity,⁵⁶ asthma,⁵⁷ cardiovascular disease,⁵⁸ and aging,⁵⁹ among others.

This study reveals the incredible potential of utilizing iMacs as a tool for studying sexual dimorphism in human systems. Our findings both under control and LPS-stimulated conditions align with other reports of sex specificity in macrophages, establishing their physiological relevance as a model, and highlighting their potential as a powerful resource for advancing discovery, exploration, and understanding of sexual dimorphism in human inflammation.

Limitations of the study

While we demonstrate here the potential of iMacs in studying biological sex-based differences in inflammation, this study has several limitations. There are differences between human iMacs and peripheral blood monocyte-derived macrophages within the transcriptional profiles and function of the cells.⁶⁰ In addition, the sample size used for this study was small, with only 3 iPSC lines per biological sex from Asian and Caucasian donors, limiting our ability to capture population diversity and patient-to-patient variability. Consequently, future work is needed to expand the patient cohort size and reliably capture the sex-based differences across a range of biological diversity and variability. With iPSC-derived cells as a model, this work can be scaled by leveraging resources like large-scale iPSC banks, which have generated iPSC lines from hundreds of donors for research applications.^61,62 These resources can further allow for stricter stratification based on donor age, as well as consideration of female menopause stage and use of hormonal treatments, which were unknown for this donor cohort and may have contributed to the large inter-donor variability observed among the female iPSC lines. Finally, our findings are limited to a single time point in response to LPS, while the inflammatory response of macrophages is highly dynamic and context dependent. Further characterization of the temporal response of male and female iMacs is warranted and should be expanded to alternative stimuli, including viral infection or injury. In future work, iMacs can be integrated into more complex systems, such as organ-on-a-chip models, enabling more cell-cell interactions to more accurately capture biological sex-specific changes when cells are not in isolation.^63,64,65

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Gordana Vunjak-Novakovic (gv2131@columbia.edu).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •Bulk RNA-sequencing data have been deposited in the NCBI Gene Expression Omnibus (GEO) under the accession number GSE328566 and are publicly available as of the date of publication. Accession numbers are listed in the key resources table.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

The authors gratefully acknowledge the funding support of their research by NIH (grant P41 EB027062 to G.V.-N. and 1F31CA275733 to D.N.T.). These studies used the resources of the Herbert Irving Comprehensive Cancer Center Confocal and Specialized Microscopy Core and Columbia Genome Center, funded in part through NIH/NCI Cancer Center Support Grant P30CA013696. We would like to thank the Columbia Stem Cell Core, under the leadership of Barbara Corneo for the FA10 and FA11 iPSC lines. We would also like to thank the staff of the Columbia Stem Cell Initiative Flow Cytometry Core Facility, under the leadership of Michael Kissner, at Columbia University Irving Medical Center for their contributions to the work presented in this manuscript. All schematics were created using BioRender.com, and chord graphs and KEGG diagrams were generated from Advaita Bio’s iPathwayGuide. We thank Katherine Cunningham for her review and editing of the manuscript.

Author contributions

Conceptualization, P.L.G., R.I.L., D.N.T., and G.V.-N.; methodology, P.L.G., D.N.T., C.C., and M.S.; formal analysis, P.L.G., R.I.L., D.N.T., and R.F.; investigation, P.L.G., R.I.L., D.N.T., E.S., N.P.H., and M.P.R.; resources, P.L.G., D.N.T., C.C., M.S., and N.P.H.; writing – original draft, P.L.G., R.I.L., D.N.T., and R.F.; writing – review and editing, P.L.G., R.I.L., D.N.T., and G.V.-N.; visualization, P.L.G., R.I.L., and D.N.T.; supervision, G.V.-N.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Mouse monoclonal anti-CD68 (clone KP1)	Abcam	Cat# ab955; RRID: AB_307338
Goat anti-mouse IgG, Alexa Fluor 647	Invitrogen	Cat# A-21235; RRID: AB_2535804
Alexa Fluor 488 Phalloidin	Invitrogen	Cat# A12379
Alexa Fluor 700 anti-human CD45 (clone HI30)	BioLegend	Cat# 304024; RRID: AB_493761
Brilliant Violet 421 anti-human CD11b (clone ICRF44)	BioLegend	Cat# 301324; RRID: AB_11219589
Brilliant Violet 605 anti-human CD14 (clone M5E2)	BioLegend	Cat# 301834; RRID: AB_2563798
PE/Cyanine7 anti-human HLA-DR (clone L243)	BioLegend	Cat# 307615; RRID: AB_493589
APC anti-human CD172a (SIRPa)	BioLegend	Cat# 372105; RRID: AB_2650862
NOX1 Polyclonal Antibody	Invitrogen	Cat# PA5-103220; RRID: AB_2852587
Goat anti-rabbit IgG, Alexa Fluor 568	Invitrogen	Cat# A-11011; RRID: AB_143157
DAPI	Thermo Scientific	Cat# 62248
Biological samples
Male and female leukopaks	New York Blood Center	nybc.org
Chemicals, peptides, and recombinant proteins
mTeSR Plus medium	STEMCELL Technologies	Cat# 100–0276
Matrigel	Corning	Cat# 354230
RPMI 1640	Gibco	Cat# A41923-01
EDTA	Invitrogen	Cat# 15575
PBS	Corning	Cat# 21–040
Y-27632 dihydrochloride	Tocris Bioscience	Cat# 1254
STEMdiff Hematopoietic Kit	STEMCELL Technologies	Cat# 05310
X-VIVO 15 Serum-free Medium	Lonza	Cat# 02-053Q
GlutaMAX	Invitrogen	Cat# 35050–061
2-mercaptoethanol	Gibco	Cat# 21985–023
Human recombinant macrophage colony stimulating factor (M-CSF)	PeproTech	Cat# 300–25
Human recombinant interleukin-3 (IL-3)	PeproTech	Cat# 200–03
Human CD14 microbeads	Miltenyi	Cat# 130–050–201
Human Serum	Millipore Sigma	Cat# H4522-100 ML
Penicillin-Streptomycin	Gibco	Cat# 15140148
4% Paraformaldehyde	Santa Cruz Biotechnology	Cat# Sc-281692
Triton X-100	Sigma	Cat# X100-100 mL
Bovine Serum Albumin (BSA)	Sigma	Cat# A7906-500G)
ProLong Glass Antifade Mountant with NucBlue Stain	Invitrogen	Cat# P36983
TrypLE	Gibco	Cat# 12605010
FcR Blocking Solution	Miltenyi	Cat# 130–059–901
Propidium Iodide	Invitrogen	Cat# P3566
Fluorescent red latex beads, carboxylate-modified polystyrene	Millipore Sigma	Cat# L3030
Lipopolysaccharides (LPS) from E. coli O26:B6	Millipore Sigma	Cat# L2654
Human recombinant CXCL12	PeproTech	Cat# AF-300-28A
Lymphoprep Density Gradient Medium	StemCell Tech	Cat# 07801
Critical commercial assays
CytoSelect 96-well Cell Migration Assay	Cell Biolabs	Cat# CBA-105
RNeasy Micro Kit	Qiagen	Cat# 74004
LEGENDplex Human Inflammation Panel 1	BioLegend	Cat# 740809
TruSeq Stranded mRNA Library Prep Kit	Illumina	Cat# 20020594
SepMate-50 Tubes	StemCell Tech	Cat# 85450
EasySep Human CD14 Positive Selection Kit	StemCell Tech	Cat# 18002
Deposited data
Bulk RNA-sequencing	This paper	GEO: GSE328566
Experimental models: Cell lines
WTC-11	Material Transfer Agreement from Bruce Conklin, Gladstone Institute	RRID: CVCL_Y803; https://gladstone.org/science/stem-cell-core
BS-2	Columbia University Stem Cell Core, IRB AAAR1017	https://www.stemcell.columbia.edu/
FA10 (FA0000010)	Columbia Stem Cell Core; Patel et al.66	TALSCTRL11.5; RRID: CVCL_EZ99
FA11 (FA0000011)	Columbia Stem Cell Core; Patel et al.66	TALSCTRL15.12; RRID: CVCL_FA00
AA1 (CW20136AA1)	FUJIFILM Cellular Dynamics CIRM iPSC Repository; https://www.fujifilmcdi.com	Cat# CW20136AA1
CC1 (CW20052CC1)	FUJIFILM Cellular Dynamics CIRM iPSC Repository; https://www.fujifilmcdi.com	Cat# CW20052CC1
Software and algorithms
ImageJ	Schneider et al.67	https://imagej.org
CellProfiler (4.2.8)	Stirling et al.68	https://cellprofiler.org
FlowJo	BD Biosciences	https://flowjo.com
GraphPad Prism (10.2.1–11.0.0)	GraphPad	https://www.graphpag.com
Bcl2fastq (version 2.19)	Illumina	https://support.illumina.com/sequencing/sequencing_software/bcl2fastq-conversion-software/downloads.html
R Studio (version 4.3.1)	Posit	https://www.rstudio.com/
R/Bioconductor	Huber et al.69	https://www.bioconductor.org
Kallisto (0.44.0)	Bray et al.70	https://github.com/pachterlab/kallisto
iPathwayGuide	Advaita Bioinformatics	https://advaitabio.com/
Other
NovoCyte Penteon	Agilent BioTek	https://www.agilent.com/en/product/research-flow-cytometry/flow-cytometers/flow-cytometer-systems/novocyte-penteon-flow-cytometer-1270342
ECLIPSE Ti2 inverted microscope	Nikon	https://www.microscope.healthcare.nikon.com/products/inverted-microscopes/eclipse-ti2-series
NovaSeq6000 with Real Time Analysis (RTA3)	Illumina	https://www.illumina.com/systems/sequencing-platforms/novaseq.html

Experimental model and study participant details

iPSC sourcing

All human iPSC lines were previously established and obtained through material transfer agreement. No new human participants were recruited for this study. WTC-11 (Asian male, age 30, GM25256, Coriell) was obtained from B. Conklin, Gladstone Institute. BS-2 (Asian female, age 20) was derived at Columbia University from a blood sample obtained from a female donor who had provided informed consent for blood donation for research use (IRB AAAR1017). The blood sample was fully de-identified in accordance with applicable institutional policies and relevant regulations, so that no information that could identify the donor was available to the research team. FA10 (Caucasian male, age 36, FA0000010) and FA11 (Caucasian female, age 49, FA0000011) were generated by and obtained from the Columbia Stem Cell Core.⁶⁶ AA1 (Asian male, age 27, CW20136AA1) and CC1 (Asian female, age 26, CW20052CC1) were obtained from the FUJIFILM Cellular Dynamics CIRM iPSC Repository. WTC-11, BS-2, FA10, and FA11 lines were karyotyped previously, and all lines were minimally expanded prior to this study.

iPSC culture

Human iPSCs were cultured in mTeSR Plus medium (STEMCELL Technologies, 100–0276) on tissue culture treated plastic well-plates coated with Matrigel (Corning, 354230) 1:100 in RPMI 1640 (Gibco, A41923-01). The cells were passaged every 3–4 days, upon reaching 70% confluency. Cells were detached using 0.5 mM EDTA (Thermo Fisher Scientific, 15575) in PBS (Corning, 21–040), then for the first 24 h after passaging, cultured in mTeSR Plus supplemented with 5 μM Y-27632 dihydrochloride (Tocris Bioscience, 1254). All iPSC lines were regularly tested for mycoplasma contamination.

iPSC-macrophage differentiation

iPSCs were replated on tissue culture plates coated with Matrigel (Corning, cat. 354230, diluted 1:175) with low agitation to allow for clusters of cells to attach at approximately 8–16 clusters/cm². Cells were first differentiated into hematopoietic stem and progenitors using a STEMdiff Hematopoietic Kit (STEMCELL Technologies, 05310), according to the manufacturer’s instructions, until day 12 of differentiation. At day 12, cells in suspension were removed and the adherent hemogenic endothelial cell fraction was cultured in X-VIVO 15 Serum-free Hematopoietic Cell Medium (Lonza, 02-053Q) supplemented with 1:100 GlutaMAX (Invitrogen, 35050061), 1:100 2-mercaptoethanol, 100 ng/mL recombinant human macrophage colony stimulating factor (M-CSF, Peprotech, 300–25), and 25 ng/mL recombinant human interleukin 3 (IL-3, Peprotech, 200–03). Hemogenic endothelial cells were maintained for over 8 weeks with weekly media exchange, with cells in the suspension fraction discarded during week 1. Myeloid- and monocytic-specific lineage cells subsequently collected in the suspension fraction were replated onto 6-well ultra-low attachment plastic (Corning) in X-VIVO 15 media containing 100 ng/mL M-CSF for at least 5 days to achieve macrophage differentiation. CD14⁺ macrophages were positively selected prior to use via CD14 microbeads (Miltenyi, 130–050–201), according to the manufacturer’s protocol.

Primary human monocyte sourcing and isolation

Mononuclear cells used in this study were isolated from commercially obtained leukopaks supplied by the New York Blood Center. The blood products were collected from three male and three female donors who had provided informed consent for blood donation and for research use in accordance with the policies of the supplier. All specimens provided to the investigators were fully de-identified, and no information that could identify individual donors was available to the research team. The research was conducted using only de-identified human biospecimens in accordance with applicable institutional policies and relevant regulations. Blood was diluted with an equal volume of PBS containing 2% FBS and overlaid on Lymphoprep Density Gradient Medium (STEMCELL Technologies, 07801) using SepMate-50 Tubes (STEMCELL Technologies, 85450) according to the manufacturer’s protocol. Following centrifugation at 1200 × g for 10 min at room temperature, the top layer containing mononuclear cells was collected and washed twice in 2% FBS in PBS. The cells were resuspended in 2% FBS and 1 mM EDTA, and CD14⁺ monocytes were subsequently isolated using an EasySep Human CD14 Positive Selection Kit (STEMCELL Technologies, 18002) according to the manufacturer’s protocol. Primary human CD14⁺ monocytes were resuspended in RPMI containing 10% human serum and 1% penicillin-streptomycin and frozen in CryoStor until use.

Primary human macrophage differentiation

Primary human male and female CD14⁺ monocytes were cultured for 5 days at 1 × 10⁶ cells/mL on ultra-low attachment plastic in RPMI 1640 supplemented with 10% human serum, 1% penicillin-streptomycin, and 20 ng/mL M-CSF. Media was replenished on day 3 and on day 5, macrophages were gently scraped and collected for use via centrifugation at 400 × g for 7 min.

Method details

Experimental design

All experiments involved macrophages derived from three healthy male (WTC-11, AA1, FA10) and three healthy female (BS-2, CC1, FA11) iPSC lines, with ≥2–3 technical replicates per cell line. Where indicated, functional validation experiments were performed using primary human monocyte-derived macrophages obtained from leukopaks for three male and three female donors.

Immunofluorescence

Chambered coverslips (Ibidi, 80827) were coated with FBS for 1 h prior to cell plating. 24 h after cell attachment, macrophages were fixed with 4% Paraformaldehyde Solution in PBS (Santa Cruz Biotechnology, Sc-281692) for 20 min at room temperature, permeabilized in 0.25% Triton X-100 (Sigma, X100-100 mL) for 20 min at room temperature, and then blocked with 5% BSA (Sigma, A7906-500G) in PBS for 2 h at room temperature. iMacs were then stained with CD68 (1:50, Abcam, ab955) in 1% BSA in PBS overnight at 4°C. Residual stain was removed by washing with PBS. Samples were then incubated in Alexa Fluor 647 (1:1000, Invitrogen, A-21235) and Alexa Fluor 488 Phalloidin (1:1000, Invitrogen, A-12379) in 1% BSA in PBS for 1 h at room temperature, then washed with PBS. Finally, samples were mounted using ProLong Glass Antifade Mountant with NucBlue Stain (Fisher, P36983).

Detection of NOX1 was performed as described above, using a 96 well glass bottom plate. iMacs were stained with NOX1 (1:200, Invitrogen, PA5-103220) in 1% BSA in PBS overnight at 4°C, and subsequently washed and staining with Alexa Fluor 568 (1:600, Invitrogen, A-11011) and DAPI (1:600, Thermo, 62248) for 2 h at room temperature.

All images were acquired using a Nikon Ti2 inverted microscope with AXR resonant spectral scanning confocal unit, then processed in ImageJ.

Flow cytometry

iMacs were detached from ultra-low attachment culture dishes by incubation with TrypLE (Thermo Fisher, 12605010) for 3 min. For flow cytometry characterization of surface marker expression, cells were centrifuged, resuspended in FACS buffer (2% FBS, 0.5 mM EDTA), and blocked in FcR blocking solution (Miltenyi, 130–059–901) for 15 min at 4°C. Cells were subsequently resuspended in FACs buffer containing AlexaFluor 700-conjugated anti-human CD45 (BioLegend, 304024, 1:50 dilution), Brilliant Violet 421-conjugated anti-human CD11b (BioLegend, 301324, 1:50 dilution), Brilliant Violet 605-conjugated anti-human CD14 (BioLegend, 301834, 1:50 dilution), PE/Cyanine7-conjugated anti-human HLA-DR (BioLegend, 307615, 1:50 dilution), and APC-conjugated anti-human CD172/SIRPa (BioLegend, 372105, 1:50 dilution). Cells were then stained with propidium iodide (Thermo Fisher), and washed twice with FACS buffer. Data were acquired on a NovoCyte Penteon flow cytometer at the Columbia Stem Cell Initiative Flow Cytometry core facility. Data analyses were performed using FlowJo software (BD BioSciences).

Phagocytosis assay

iMacs were replated onto FBS-coated wells at a density of 100,000 cells/well of a 96-well plate and incubated overnight in X-VIVO 15 media containing 100 ng/mL M-CSF. The cells were subsequently incubated with fresh media containing 20 μL/mL fluorescent red latex beads (Sigma, L3030) for 2 h at 37°C. Post-incubation, cells were enzymatically detached, centrifuged, washed twice with FACS buffer, and stained with CD45 and CD14 antibodies for flow cytometry analysis, as described above.

Primary human CD14⁺ macrophages were replated onto a 96 well ultra-low attachment plate at 1 × 10⁶ cells/mL in RPMI 1640 growth media containing 20 ng/mL M-CSF, with or without 100 ng/mL LPS, and incubated for 48 h at 37°C. Post-incubation, the media was replaced with fresh media containing 20 ng/mL M-CSF and 20 μL/mL fluorescent red latex beads (Sigma, L3030) and the cells were incubated an additional 2 h at 37°C. The cells were gently scraped and collected, and analyzed as described for iMacs.

Cell migration assay

Macrophage migration was quantified using the CytoSelect 96-well Cell Migration Assay (Cell Biolabs, CBA-105), composed of a Boyden Chamber with a polycarbonate membrane of 5 μm pore size. Macrophages were suspended in in X-VIVO 15 media containing 100 ng/mL M-CSF, supplemented with or without 100 ng/mL LPS (Sigma, L2654). Cells were seeded in the upper chamber at a density of 100,000 cells/well. The bottom chamber contained X-VIVO 15 media containing 100 ng/mL M-CSF with or without LPS, as well as 50 ng/mL CXCL12 (Peprotech, AF-300-28A) to act as the chemoattractant agent. The cells were incubated at 37°C, 5% CO₂ for 3 h, and subsequently analyzed according to manufacturer’s instructions.

RNA isolation

Cells were lysed using Buffer RLT from a Qiagen RNeasy Micro Kit (Qiagen, cat. 74004), then stored at −20°C for later use. Upon thaw, RNA isolation proceeded with the Qiagen RNeasy Micro Kit according to the manufacturer’s instructions. Resulting total RNA for each sample was then assess by an Agilent bioanalyzer, Qubit 2.0 at Columbia’s Molecular Pathology core, and samples with an RNA integrity number (RIN) > 8 were considered acceptable for bulk RNA sequencing analysis.

Bulk RNA sequencing

Total RNA samples were submitted to the JP Sulzberger Columbia Genome Center for sequencing. A poly-A pulldown was used to enrich mRNAs from total RNA samples, then a library was constructed using TruSeq Stranded mRNA Library Prep Kit (Illumina) following the manufacturer’s protocol. Libraries were sequenced to a targeted depth of ∼20M 100bp paired-end reads on a NovaSeq6000. RTA (Illumina) was used for base calling and bcl2fastq2 (version 2.19) was used for converting BCL to fastq format, coupled with adapter trimming. We conducted pseudoalignment to a kallisto index created from transcriptomes (Ensembl v96, Human: GRCh38.12) using kallisto (0.44.0).

RNA-sequencing analysis

Data processing and analysis were performed in R/Bioconductor.⁶⁹ Genes which had at least 10 reads in at least 3 samples were considered to be expressed and subject to further analysis. Multidimensional scaling,⁷¹ principal component analysis,⁷² and hierarchical clustering⁷³ were used to check for outliers. Samples were logcpm transformed with a pseudocount of 1. Multidimensional scaling was performed with the plotMDS function in limma. Principal component analysis was performed with the plotPCA command in affycoretools.⁷⁴ Hierarchical clustering was performed using the heatmap.2 function in the gplots2 package; only the top 1000 genes are included for clarity and computational tractability. A Euclidean distance metric and complete linkage clustering was used. Samples were normalized by the Trimmed Mean Method.⁷⁵ Differential expression was analyzed with limma-voom with sample weighting⁷⁶ and duplicate correlation,⁷⁷ which models the effect of 2 samples per donor iPSC line. Differential expression results were analyzed with the KEGG⁷⁸ and Biological Process Gene Ontology⁷⁹ databases, using a significance cutoff of p < 0.001 and |log2 fold change| >0.6. iPathwayGuide⁸⁰ was used to perform the KEGG analysis, using the SIPA (Signal Interaction Pathway Analysis) method⁸¹ and Gene Ontology analysis using the elim method.⁸² Genes on X and Y chromosomes were identified used the TxDb.Hsapiens.UCSC.hg38.knownGene⁸³ package.

To identify which genes changed in response to LPS treatment differently in male and female macrophages, we identified genes that either (i) changed in opposite directions upon LPS treatment in male and female statistically significantly (FDR ≤0.05 for both comparisons), or (ii) changed statistically significantly (FDR ≤0.05) for 1 sex, but were equivalent statistically significantly (p ≥ 0.10)⁸⁴ with 95% confidence in the other sex.

Protein secretion analysis

Macrophages were plated at a density of 100,000 cells/well in an ultra-low attachment 96-well plate, and cultured in X-VIVO 15 media containing 100 ng/mL M-CSF, supplemented with or without 100 ng/mL LPS (Sigma, L2654) for 48 h. The conditioned supernatant was collected and centrifugated at 400 g for 5 min to remove cellular debris, then frozen at −20°C until analysis. LEGENDplex Human Inflammation Panel 1 kit (Biolegend, 740809) was used to quantify inflammatory cytokine presence in the conditioned cell culture supernatant. Upon use, samples were thawed and the assay was performed according to the manufacturer’s instructions.

Quantification and statistical analysis

Statistical analysis was performed using GraphPad Prism 9. Outliers were identified using Grubbs’ method. Statistically significant differences were determined using a two-tailed, unpaired t test, assuming a Gaussian distribution. Assumptions of normality were tested using the Shapiro-Wilk test. For all comparisons, p < 0.05 was considered statistically significant (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001). All data are represented as mean ± SD. A sample size of n = 3 biological replicates was used throughout this study, defined as iMacs derived from 3 female iPSC lines and 3 male iPSC lines. For each cell line, ≥2–3 technical replicates were included for all assays. Data points represent the mean for each biological replicate.