The X-quisite X-ception: Sex Differences with Immune Responses

Abstract

There are significant sex-based differences in immune responses to pathogens and self-antigens, with human females exhibiting increased susceptibility to various autoimmune diseases, and human males displaying preferential susceptibility to some viral, bacterial, parasitic, and fungal infections. While sex hormones clearly contribute to sex differences in immune cell composition and function, the presence of two X chromosomes in human females suggests that differential gene expression of numerous X-linked immune-related genes may also influence sex-biased innate and adaptive immune cell function in health and disease. Here, we review the sex differences in immune system composition and function, examining how hormones and genetics influence the immune system. We focus on the genetic and epigenetic contributions responsible for altered X-linked gene expression, and how this impacts sex-biased immune responses in the context of pathogen infection and systemic autoimmunity.

INTRODUCTION

Biological sex contributes to physiological differences between human males and females that can influence pathogen exposure, recognition and clearance, and replication. Recent studies investigating how mammals respond to immunogenic challenges have revealed numerous differences between male and female susceptibility to pathogens such as bacteria, fungi, viruses, and parasites, likely via sex-biased differences in immune responses (Figure 1). In general, human females have greater innate and adaptive immune responses compared to males. Females usually clear infections faster than males, and males exhibit increased mortality rates across all ages, including in pre-term births and infants, after infection¹. Females often exhibit stronger serologic responses compared to males following vaccination. Even in mice, females are more resistant to bacterial and viral infections and generate stronger and longer-lasting immune responses². This seemingly advantageous female sex bias in the potency of immune responses is juxtaposed by the female-biased proclivity to autoimmunity. Human females are at increased risk for some autoimmune diseases, including systemic lupus erythematosus (SLE), Sjogren’s Syndrome, and systemic sclerosis³. However, the diverse molecular mechanism underlying these sex biased immune responses remain incompletely understood, and clarifying their origin may help provide new insights into disease pathogenesis.

Figure 1. Sex differences with pathogen infections and autoimmune disease.

The sex chromosomes and the sex hormones in biological males (XY) and females (XX) are responsible for the observed sex differences with responses to various pathogens (a) and in autoimmune disease (b). (a) The overall sex bias associated with a particular class of pathogen infection (and associated disease) is shown. Size of the symbol reflects amount of sex bias. (b) Autoimmune diseases are strongly female-biased.

The molecular origins for sex differences in immune responses originate from the X and Y chromosomes, the reproductive organs, and variations in levels of sex hormones over time. Sex hormones, specifically androgens and estrogens, can have either pro- or anti-inflammatory effects, and many genes with immune functions are regulated by the presence or absence of nuclear hormone receptor binding (BOX 1). The sex chromosomes provide the genetic and epigenetic foundations for altered gene expression in a heritable, sex-specific fashion in immune cells. The X chromosome contains about 50 immune-related genes, including some important for immune cell identity (FOXP3), cellular activation and intracellular signaling (CD40LG, TLR7, IRAK1, IL13RA1/2, NEMO, TASL, IL-9R), leukocyte trafficking (CD99, CXCR3), immune cell differentiation and proliferation (IL-2RG, BTK), and cellular metabolism (OGT, CYBB). Thus, the X chromosome is likely an important contributor to the molecular basis of female-biased antimicrobial responses and autoimmunity. In this Review, we review current literature on sex differences in immune system composition and function, examining how hormones and genetics of the X-chromosome influence the immune system. We focus on genetic and epigenetic contributions to altered X-linked gene expression and how this impacts female-biased autoimmune disorders and highlight examples of altered X-linked gene expression that influence pathogen susceptibility and disease progression.

Text Box 1: Sex hormones and inflammation.

Estrogens (estrone, 17β-estradiol, estriol), progesterone, and androgens (testosterone, dihydrotestosterone) are the predominant gonadal sex hormones. The sex hormone receptors – estrogen receptors (ER alpha, beta), progesterone receptor (PR), and androgen receptor (AR) – are hormone-activated transcription factors that bind to hormone response elements to regulate gene expression. ERs, PRs, and ARs are expressed by a variety of immune cells (T cells, B cells, dendritic cells, monocytes and macrophages, NK cells, type 2 innate lymphoid cells, and granulocytes) and cells at the interface of skin and mucosal barrier sites, such as thymic epithelial cells¹²⁸. Sex-biased gene expression can result from sex-specific differences in steroid production. For example, numerous ER response elements in the IFN-gamma promoter allow for increased downstream gene expression when 17β-estradiol levels are increased, thereby contributing to female biased increased production of IFN-gamma¹²⁹.

Estrogens can exhibit either anti-inflammatory (at high concentrations) or pro-inflammatory effects (at low concentrations)¹³⁰. During pregnancy or periovulation when estrogen concentrations are high, estrogens exert inhibitory effects on T cells (specifically Th1 and Th17-polarized cells), M1 macrophages, dendritic cells, neutrophils, and microglia through NF-kB inhibition, and can also increase Treg function. ERalpha in T cells can promote T cell activation induced apoptosis and downregulates Foxp3 expression¹³¹, and can also suppress TFH cell development¹³². Estrogens can also signal through ER-beta in Treg cells in the intestine to regulate immune suppression¹³³. Both ER-alpha and ER-beta in CD4+ T cells are important for ligand-mediated suppression of autoimmunity of the central nervous system (Multiple Sclerosis; MS), specially targeting pathogenic Th17 cells^134,135. ER-beta is also required for Tregs to regulate macrophage proinflammatory responses for resolving lung inflammation resulting from pneumonia¹³⁶.

The impact of estrogen on the immune response is also influenced by the immune microenvironment. In animal models of lupus-like disease, estrogens can be pro- or anti-inflammatory, depending on the proinflammatory milieu responsible for the observed lupus-like phenotypes. Deletion of ER-alpha in B cells reduced autoantibody production and nephritis in lupus-prone NZB/W F1 mice¹³⁷. In the MRL/lpr model of spontaneous lupus-like disease, exogenous estrogen administration resulted in accelerated immune-complex glomerulonephritis, reflecting a proinflammatory effect via B-cell activation and autoantibody production; however, T-cell mediated disease, including periarticular inflammation, focal sialadenitis, and renal vasculitis, was significantly reduced^130,138. Inflammatory cytokines including TNF-alpha, IL-1, and IL-6 can stimulate the activity of aromatase, an enzyme required for estrogen biosynthesis. Thus, the inflammatory milieu may confer contextual effects by altering the local concentrations of estrogen¹³⁹. The concentration- and context-dependent effects of estrogen are also apparent in examples of systemic autoimmune disease. In rheumatoid arthritis and MS, disease activity often spontaneously decreases during pregnancy, a phenomenon which has been attributed to the anti-inflammatory effects of high concentrations of estrogen^134,135,140. Nevertheless, the cumulative effects of local estrogen abundance are not so straightforward, as disease activity in systemic lupus erythematosus often spontaneously increases during pregnancy¹⁴¹. These observations from disease activity in pregnancy highlight the complex interactions between sex hormones, immune cells, immune tolerance, and the developing fetus.

Androgens have anti-inflammatory effects on immune responses in vivo and in vitro, through diverse mechanisms. In the context of their effects on the innate immune system, they have been shown to modify signal transduction through pattern recognition receptors, suppressing both NF-kB and Tlr4 expression¹⁴². Androgens also lower cytokine secretion of TNF-alpha, IL-1β, and IL-6 from monocytes and macrophages, and IL-33 production from mast cells^41,143,144. Androgens impact adaptive immune responses, and can inhibit both humoral and cellular immune responses by reducing T and B cell proliferation and decreasing immunoglobulin and cytokine production^{35 41,143,144}. In terms of B cells, androgens have inhibitory effects on B cell lymphopoiesis and specifically affect the production of B cell precursors, which express AR; conversely, mature B cells, which lack this receptor, are not directly impacted by androgens⁴¹. CD4+ T cells from female mice treated with dihydrotestosterone (DHT) produce more IL-10 compared to untreated female animals ³⁶.

Sex differences in immune cell composition

The observed sex-biased differences in innate, humoral, and cellular immunity in humans are accompanied by differential immune composition in males and females. For example, flow cytometric immunophenotyping of healthy volunteers in Europe and Asia have consistently identified female-specific elevations of CD4+ T-cells, specifically naïve cells, which may reflect enhanced thymic function^4,5. A recent whole blood transcriptional profiling of multiple international and intercontinental cohorts of healthy individuals similarly identified a higher proportion of CD4+ T cells in human females ⁶. In addition, human females have more circulating CD19+ B-cells, plasma cells, regulatory T cells (defined by CD25hiCD127lo), and both naïve CD8+ and mucosa-associated invariant T-cells relative to males^4,7-10. Conversely, human females have lower proportions of both CD14+ and CD16+ monocytes, a generally lower proportion of myeloid cells ^6,8 and fewer NK cells ^4,7,8,11 relative to males. Differences in humoral immunity have also been observed between the sexes, with human females having higher quantities of IgM antibodies¹².

Sex differences in immune cell composition have also been observed across mouse strains ^13,14. Within resident immune populations of C57BL/6 mice and Wistar rats, there are greater total leukocyte quantities, and specifically more B and T cells and F4/80+ macrophages in female animals¹⁵. Consistent with these findings, mesenteric tissues from female animals express higher levels of both innate and adaptive immune cell chemokines and chemokine receptor genes that could enhance recruitment. A recent study found that male mice have higher numbers of splenic NK cells, yet have reduced effector function due to reduced expression of the X-linked gene Kdm6a (Utx), a histone lysine demethylase. Notably, Kdm6a (Utx) is expressed in higher levels in female NK cells¹⁶. Curiously, sex hormones do not significantly influence NK number or effector function, as a female-specific Utx deletion in NK cells increases NK cell frequency and reduces effector function through aberrant upregulation of Bcl2 expression, an anti-apoptosis factor, and downregulation of IFN-gamma¹⁶. The identification of additional X-linked genes that are capable of influencing both innate and adaptive immune cell frequencies may help further clarify the basis of the observed differences in immune cell composition between the sexes.

Sex hormones influence immune cell function

Sex differences in immune cell function arise from differential gene expression in immune cell subsets of males and females, and sex hormones can contribute towards these observed gene expression differences. Transcriptional profiling of innate and adaptive immune cells – including monocytes, naïve B cells, and various T cell subsets – identified about 1875 transcripts exhibiting sex biased expression, the majority of which were autosomal¹⁷. Most of these transcripts exhibited sex-biased expression in one type of immune cell, underscoring the cellular specificity of sex-specific gene expression and simultaneously highlighting the broad impact of sex hormonal and trans-acting sex chromosomal elements on the observed gene expression differences¹⁷. Estrogens can have either pro-inflammatory or anti-inflammatory functions (BOX 1), depending on both the local hormone concentration and the immune cell examined. For example, transcription of AICDA, which encodes the AID enzyme which is important for somatic hypermutation and class switch recombination in B cells, is regulated by binding of the estrogen-ER-alpha complex to its promoter region of AICDA. In autoimmune disease, AID contributes to the production of high-affinity IgG auto-antibodies¹⁸. Estrogens also increase the expression of endosomal Toll like Receptors (TLR) 3, 7, 9, which are important regulators of Type I interferons (IFNs)¹⁹. Co-culture of the SLE-derived estrogen-treated CD4+ T cells with B cells from healthy controls resulted increased antibody production in vitro²⁰. Type I IFN production and antiviral responses downstream of TLRs 3, 7, and 9 are regulated by TRIM21, whose expression is increased by estrogen. TRIM21 expression also induces IL-23, which promotes Th17-differentiation, an adaptive immune pathway integral to the pathogenesis of several systemic autoimmune diseases, including SLE, SSc, SS, and RA²¹. The female-bias for enhanced type I IFN signaling has also been observed in human plasmacytoid dendritic cells (pDCs), which produce IFNα following TLR stimulation. Female pDCs treated with TLR7 and TLR8 agonists, but not TLR9 agonists, produce more IFNα compared to male pDCs, and more mRNA transcripts of the IFN-stimulated gene, IRF5^22-24. Although a sex-bias with type I IFN production was not originally observed for TLR9, this could be due to the type of CpG used in the original study²², as sex biased type I IFN production by pDCs stimulated through TLR9 has been recently reported^25,26. Both sex chromosomal and sex hormonal mechanisms likely contribute towards this female bias because TLR7 and TLR8 are X-linked, and IRF5 transcription is regulated by estrogen receptor 1 (ESR1)²⁷.

Sex differences in immune functions, particularly within the context of microbial infection, also reflect sex differences in the cellular expression of pattern recognition receptors. Peritoneal macrophages from female rodents express higher levels of Tlr2, Tlr3, and Tlr4, and exhibit enhanced phagocytosis and NADPH-mediated bacterial elimination compared to their male counterparts¹⁵. Accordingly, peritoneal-derived macrophages from female mice also exhibit increased expression of interferon-stimulated genes, and female mice exhibited blunted sepsis severity and fewer recoverable live bacteria in circulation^28,29. These findings were abrogated by prior ovariectomy and support a role for sex hormones in shaping the tissue resident immunophenotype. The B1 cell-dependent neutralizing antibody response to LPS-O-antigens of enteropathogenic E. coli (EPEC) occurred in sexually mature females, resulting from estrogen-mediated upregulation of BAFF and APRIL cytokines secreted by peritoneal macrophages²⁹. Bone marrow-derived macrophages from female mice have higher levels of Tlr8 compared to male mice³⁰, and peripheral blood mononuclear cells from human females express more TLR7 compared to males³¹. Vaccination or viral challenge increases expression of TLR pathway and pro-inflammatory genes in female PBMCs from humans and rats, and it has been suggested that estrogen binding at sex hormone receptor elements in the promoter regions of these genes is responsible for increased gene expression ³². However, neutrophils from human males have higher expression of TLR4 and this expression increases following activation with LPS, which results in higher pro-inflammatory cytokine production and may contribute towards the male bias observed with endotoxic shock³³. Notably, in line with its recognized generally anti-inflammatory role, androgen treatment of male mice decreases Tlr4 expression in macrophages, and it can also suppress the invasion and colonization of uropathogenic E.coli through inhibition of the JAK-STAT signaling pathway and decreased production of IL-1β, IL-6, and IL-8 ³⁴. Androgens have anti-inflammatory effects (BOX 1) on both humoral and cellular immune responses by reducing T and B cell proliferation and decreasing immunoglobulin and cytokine production³⁵. Female mice treated with dihydrotestosterone (DHT) produce more IL-10 and less IL-12 compared to untreated female animals, resulting from increased AR signaling in CD4+ T cells³⁶.

Androgens also play an important role in central tolerance through regulated expression of AIRE in the thymus, the expression of which exhibits a sex bias. While androgens recruit ARs to the AIRE promoter to increase transcription in male mice and humans, estrogens inhibit AIRE expression^37,38. Androgens can also influence expression of FOXP3, a transcription factor important for CD4+ regulatory T cells (Tregs) that play an important role in tolerance. Treg-intrinsic AR signaling increases Treg suppression during an allergen challenge, and human Tregs treated with androgen reduced IL-33 induced ST2 expression and human bronchial epithelial cells produced less IL-33³⁹. There is evidence that androgen treatment can increase FOXP3 expression in Tregs from human females during the ovulation phase of the menstrual cycle yet had no effect on Tregs from males⁴⁰. Androgens have inhibitory effects on B cell lymphopoiesis and specifically impact B cell progenitors, which express AR while mature B cells lacking this receptor were not affected⁴¹. Unlike estrogen which upregulates expression of B cell activating factor (BAFF), BAFF expression is inhibited by androgens⁴². Thus androgens block BAFF-dependent B cell clonal expansion and class switch recombination, and downregulate GC responses, acting through B cell extrinsic mechanisms⁴³. Androgens also increase expression of PTPN1, which functions in cell growth, differentiation, mitosis and immune cell function. AR-signaling prevents Th1 development by upregulating Ptpn1, which is an inhibitory phosphatase of Tyk2 and Jak2 kinases upstream of Sta4, thereby inhibiting IL12/Stat4 signaling ⁴⁴.

Genetic and epigenetic contributions to sex differences in immune responses: X-Chromosome Inactivation (XCI)

Female mammals regulate X-linked gene expression using XCI (BOX 2), in which one X chromosome (the inactive X chromosome, or Xi) is transcriptionally silenced during development, and a memory of this silencing event is maintained with each cell division and persists into adulthood. Various epigenetic modifications, as well as the long noncoding RNA Xist, are enriched across the Xi, and function to maintain transcriptional repression of most, but not all, of the genes on the Xi. Xist expression is required to form the Xi and loss of Xist in early development is lethal due to the requirement for an appropriate dosage of X-linked genes in both embryo and placental development ⁴⁵. Xist deletion or Xist gene silencing reduces, and often eliminates, enrichment of heterochromatic modifications from the Xi⁴⁶ and results in reactivation of some X-linked genes on the Xi, depending on the cell type and timing relative to XCI initiation. Mouse models with conditional Xist deletion in post-XCI somatic cells yields variable phenotypes, including myeloproliferative neoplasia (deletion in hematopoietic stem cells), increased polyp formation (intestinal deletion, following azoxymethane/dextran sulfate treatment), or even the lack of a perceptible phenotype (deletion in neurons, B cells, epithelial cells, intestinal cells) where animals are viable and have normal lifespans^47-49. Collectively, these studies highlight three fundamental observations: (1) that the identify and number of X-linked genes exhibiting reactivation upon Xist deletion varies by cell type; (2) X-linked dosage imbalances are, in fact, variably tolerated in a cell-type-specific manner; (3) inflammatory stress may exacerbate Xi reactivation.

BOX 2: X-Chromosome Inactivation (XCI) & ‘Escape’ from XCI.

Female eutherian mammals use XCI for dosage compensation of X-linked genes between the sexes¹⁴⁵. XCI is initiated during early embryonic development, during which one X chromosome is randomly selected for transcriptional silencing. The process begins with upregulation of the long non-coding RNA Xist, which spreads across the future inactive X chromosome (Xi) in cis together with Polycomb Repressive Complexes (PRCs)^146-151. X-linked gene promoters and enhancers rapidly lose active histone acetylation marks concurrently with transcriptional repression^148-150. The repressive histone modifications H2AK119Ub and H3K27me3 are deposited on the Xi by PRC1 and PRC2, respectively, and the histone variant macroH2A becomes enriched^152-155, together creating a heterochromatic environment across the Xi. As a final layer of regulation to maintain transcriptional repression of the Xi, DNA methylation becomes enriched at promoters and enhancers^156-160. Thus, in most somatic cells, transcriptional silencing of the Xi is maintained with each cell division through Xi enrichment of Xist RNA, heterochromatic histone marks, and DNA methylation, some of which can be visualized cytologically using RNA FISH and immunofluorescence.

Although most genes on the Xi are silenced, about 15-23% of X-linked genes in humans and ~3-7% of X-linked genes in mice are transcribed from the Xi and ‘escape’ XCI, either constitutively or in a cell type-specific manner^161-164. The promoter regions of XCI escape genes lack Xist RNA enrichment^148,149,151, heterochromatic histone marks^148,150, and DNA methylation^159,160, and are enriched for RNA Polymerase II, DNAse I hypersensitivity sites, and CTCF occupancy, which collectively allow for gene expression from a highly heterochromatic environment. X-linked genes with a Y-linked counterpart (XY gene pairs) commonly escape XCI silencing and are expressed from the active (Xa) and inactive X (Xi) chromosomes, although transcript levels from the Xi are typically less than levels from the Xa¹⁶⁵. Importantly, X-linked genes lacking a Y-linked homolog can also escape XCI. Both the magnitude and extent of XCI gene escape vary considerably by cell type, as well as between individual cells, and among individuals. Additionally, some escape genes display sex-biased expression ^162,166-168. Of the approximately 54 immunity-related genes located on the mouse and human X chromosomes, 20.4% are expressed more highly in females and 16.7% are expressed more highly in males¹⁶⁹. However, these values are likely underestimates due to the methodological challenges associated with measuring transcription specifically from the Xi. Because XCI is random and because expression from the Xi is typically less than that of the Xa, the identification of XCI escape genes requires single-cell approaches with high sequencing depth, or bulk cell analyses using F1 hybrid mice with skewed XCI^70,170,171. Nevertheless, ongoing efforts to identify additional escape genes are likely to provide important insights into sex-biased immune responses—to date, several proinflammatory genes have been found to aberrantly escape XCI in SLE and other autoimmune diseases, suggesting a key role for the transcription of X-linked genes from the Xi in the context of female-biased immunological processes (see Table 1).

Female mammals are mosaic for the X chromosome, as each individual female cell will contain an inactive X (Xi) that is either maternal or paternal in origin. Most somatic cells examined to date have similar epigenetic features enriched on the Xi, including Xist RNA, heterochromatic histone modifications, and histone variant macroH2A, which can be visualized cytologically^50-53 . However, immune cells lack some hallmark epigenetic features of canonical XCI maintenance. Specifically, splenic B and T cells from mice and circulating lymphocytes from humans lack cytologically visible Xist RNA and heterochromatic mark enrichment at the Xi, despite any significant changes with Xist transcription^54-57. Thus, Xist transcription and its localization to the Xi are genetically independent processes. In vitro activation of B and T cells stimulates the return of Xist RNA and heterochromatic modifications to the Xi, prior to the first cell division^54,56,58,59. This dynamic localization of Xist RNA and heterochromatic marks to the Xi is also observed during lymphocyte development, as Xist RNA is lost from the Xi after the common lymphoid progenitor stage, at the pro-B cell and DN1 thymocyte stages, and Xist transcription remains constant ^56,58.

Surprisingly, non-canonical XCI maintenance is a feature of both adaptive and innate immune cells. Xist RNA exhibits substantial diversity in the robustness of its relocalization to the Xi depending on the type of immune cell. Neutrophils and pDCs lack detectable Xist RNA transcripts at the Xi; in contrast, NK cells exhibit clusters of Xist RNA pinpoints within the Xi territory⁶⁰. Bone marrow-derived macrophages stimulated with either LPS or CpG have dispersed Xist RNA patterns and about half of the nuclei have a H3K27me3 focus that co-localizes with Xist signals. In vitro stimulation of mouse pDCs with either CpG or a Tlr7 agonist did not result in detectable Xist RNA localization, despite persistent Xist transcription, and both resting and in vitro activated pDCs have few H3K27me3 foci⁶⁰.

The paucity of epigenetic features on the Xi in many immune cells, including lymphocyte progenitors, supports the hypothesis that the chromatin of the Xi in immune cells is more euchromatic than in fibroblasts, and therefore prone to aberrant reactivation at certain loci. The facultative nature of the chromatin of the Xi may facilitate the increased expression of proinflammatory X-linked genes in response to pathogen infections, which would provide a favorable advantage to females, and yet simultaneously disrupt the balance of self-tolerance resulting in the development of female-biased autoimmune disease (Figure 1).

Female-biased systemic autoimmune disease and the X chromosome

Autoimmune diseases predominantly affect human females, and 80% of all individuals with autoimmune disease are female^61,62,63. For some autoimmune diseases, such as multiple sclerosis (MS), this sex bias is modest (~65% female), while for other autoimmune diseases such as primary biliary cirrhosis, the sex bias is strongly skewed (~90% female)⁶⁴. The female sex bias with some autoimmune diseases has increased over time. For example, in the mid-twentieth century, MS originally exhibited equal prevalence between the sexes; by the 1980s the sex ratio was 2:1 (female:male); currently, the sex ratio is 3:1⁶⁵. The observed increase in the female-bias of disease likely reflects advancements in the medical classification and diagnosis of these diseases but could also reflect environmental changes. While the female bias for many autoimmune diseases is quite high, the incidence and prevalence across rheumatic autoimmune diseases can vary. For example, the sex bias with SLE is typically ~66-93% female, and the prevalence ranges from 5 - 241 per 100,000 people in the United States; for dermatomyositis (DM) and polymyositis (PM), the sex bias is ~60-75% female yet the prevalence is only about 6 per 100,000 (Table 1). Importantly, beyond disease prevalence, there are also sex differences in disease severity, specifically with respect to the degree of organ involvement, and symptom burden/disability. Male patients with MS, SLE, and SSc typically exhibit greater disease severity^66,67, and male patients with SLE are more likely to exhibit renal and cardiovascular co-morbidities⁶⁸. However, this male-biased increase in autoimmune disease severity is not universal; in RA for example, biological sex does not correlate with articular disease severity⁶⁹. Additional translational studies systematically examining the relationship between biological sex and both disease manifestations and disease severity are necessary to better understand how biological sex may impact the phenotype, pathogenesis, and natural history of disease.

Table 1. Prevalence, incidence, and sex bias for the female-biased autoimmune diseases SLE, SS, scleroderma, inflammatory myopathies, and RA.

Autoimmune Disease	Prevalence (per 100,000 people in United States)	Incidence (per 100,000 person- years in United States)	Sex Bias (% of Affected Individuals who are Women)
Systemic lupus erythematosus (SLE)	5-2411	1.0 – 23.21	66%-93%1 83.71%2
Sjogren’s syndrome (SS)	22-1033	3.94	90.54%2 96.2%4
Systemic sclerosis/scleroderma (SSc)	27.65	1.935	83.7%5 75%-93.5%6 83.80%2
Inflammatory myopathies: dermatomyositis and polymyositis (DM/PM)	6.37	0.116-0.67	65.08%2 60%-75%7
Rheumatoid arthritis (RA)	10708 10009	75.38	73.4%8

Dosage imbalances of some X-linked genes are detrimental for cell function and can result in features of systemic autoimmune disease⁶⁴ . Because XCI ‘escape’ (BOX 2) is variable across cell types, it is likely that some of these X-linked genes are biallelically expressed in specific cells (or escape silencing in a portion of the total cell population) in healthy individuals, where they accordingly exhibit sex-biased expression⁷⁰. The X chromosome contains many genes that function, either directly or indirectly, in immune processes¹. It is therefore likely that these genes are dosage sensitive and therefore subject to XCI silencing in immune cells^59,71, although direct evidence from specific cell types in mice and human samples is lacking. Also unclear is whether any of these genes can become reactivated from the Xi in response to either antigen-mediated stimulation or pathogen infection, or as a result of chronic inflammation during autoimmunity. Aberrant XCI maintenance, where XIST RNA and some heterochromatic histone modifications (H3K27me3, H2AK119-ubiquitin) are missing from the Xi in circulating lymphocytes from female SLE patients, both pediatric and adults, and also in various mouse models of spontaneous lupus-like disease which exhibit a female bias ^56,59,72,73. We have proposed the hypothesis that autoimmune disease, such as SLE, results in reduced enrichment of epigenetic features across the Xi which permits reactivation of immunity-related genes, resulting in the observed abnormal increased expression in lymphocytes (Figure 3). Additional work is necessary to determine whether perturbed XCI maintenance is a feature of other autoimmune diseases that predominantly affect human females.

Figure 3. Impairments with dynamic XCI maintenance result in aberrant overexpression of X-linked genes in female-biased autoimmune disease.

(a) The Xi in naïve lymphocytes lack cytological enrichment of Xist RNA (pink curvey lines) and heterochromatic marks (colored circles), and these modifications return to the Xi in activated cells. (b) Activated lymphocytes from patients with autoimmune disease have dispersed Xist RNA and heterochromatic marks from the Xi, and increased expression of some X-linked genes (denoted with red arrows). Prevention of Xist RNA tethering to the Xi reduces enrichment of histone heterochromatic marks on this chromosome, and persistent absence of these epigenetic modifications across multiple cell divisions may increase abnormal overexpression across the Xi.

Individuals with more than one X chromosome, including XX females, patients with Klinefelter Syndrome (XXY), and individuals with polysomy X (XXX), have increased susceptibility to some female biased autoimmune diseases (Figure 2), including SLE, Sjogren’s syndrome (SS), SSc, and PM/DM^74,75. The association between female-biased autoimmunity and X chromosome dosage may also apply more broadly to other autoimmune diseases pending additional karyotype-stratified case-control studies. The contribution of multiple X chromosomes to systemic autoimmune disease risk (independent of sex hormones) is supported by mouse models using the ‘four core genotypes’: (1) ovary-bearing XX mice , (2) ovary-bearing XY mice deficient in the Y-linked Sry gene required for sex determination and male gonad formation (XYSry mice), (3) testes-bearing XYSry⁻ and (4) testes-bearing XXSry transgenic mice⁷⁶. Pristane-induced SLE-like disease, with elevated type I interferon production, and induced experimental autoimmune encephalomyelitis (EAE) results in increased onset and disease severity in XX and XXSry animals⁷⁷, reflecting genetic and epigenetic contributions from the X, although the causal genes responsible are unknown. Bone marrow chimera experiments in which hematopoietic stem cells or fetal liver cells from ovariectomized female NZB/W F1 mice, a classic spontaneous mouse model of SLE-like disease exhibiting a female bias (see BOX 3), are injected into hormonally intact lethally irradiated NZB/W F1 males results in the development of lupus-like disease with increased numbers of germinal center B cells, memory B cells, and plasma cells in all male recipients, similar to NZB/W F1 females⁷⁸. Thus X-linked genes expressed in female (XX) immune cells, independent of prior exposure to female sex hormones, accelerate SLE-like disease onset in male animals predisposed to develop autoimmune disease, underscoring the significance of genetic contributions from the X chromosome in autoimmunity.

Figure 2. Number of X chromosomes and risk for autoimmune disease.

Increased numbers of X chromosomes is associated with higher risk for the female-biased autoimmune diseases SLE, SS, scleroderma, polymyositis, and dermatomyositis.

Box 3: Mouse models of autoimmune disease exhibiting a sex bias.

Spontaneous and induced mouse models of lupus-like disease recapitulate some features of human autoimmune disease (primarily autoantibody production, lymphoid activation and hyperplasia, and lupus nephritis), and some of these models exhibit a sex bias. Male BXSB/Yaa bear a translocation of the telomeric end of the X chromosome containing Tlr7 region and 15 neighboring genes onto the Y-chromosome (Yaa mutation)^172,173 and acquire a lupus-like phenotype over time. Notably, Tlr7 knockdown in BXSB/Yaa mice abrogates disease development¹⁷⁴, and transgenic overexpression of Tlr7 on a non-lupus prone background results in lupus-like disease⁹⁶. The BXSB strain also contains mutations in the MHC locus, and loci on chromosomes 1, 3, and 13 (Bsx1-6) that contribute to disease activity¹⁷⁵. The F1 hybrid NZB/W mouse model develops spontaneous lupus-like disease that exhibits a female bias (100% females and < 40% males develop disease phenotypes by 1 year)^176,177. NZB/W mice also develop features of SS and female mice exhibit more extensive salivary gland lesions compared to male mice¹⁷⁸. Both X chromosome number and hormones appear to influence disease onset and severity in NZM/W F1 mice, as estrogen accelerates disease and testosterone provides anti-inflammatory protection in some studies, but not others ^179,180. In vitro activated lymphocytes from female NZB/W F1 mice with lupus-like disease exhibit mislocalized Xist RNA and reduced enrichment of H3K27me3 at the Xi, and sex biased gene expression differences^56,73. Backcrosses between NZB/W F1 and NZW generated various recombinant inbred strains of New Zealand Mixed (NZM) mice, among which NZM2328 and NZM2410 have earlier disease onset, and NZM2328 displaying a stronger female bias than NZM2410¹⁸¹. Lupus-like disease can be induced in various mouse strains (with variable efficiency) using intraperitoneal injections of pristane, during which females develop more severe disease phenotypes including anti-Sm, anti-dsDNA, anti-ribosomal P, anti-Su autoantibodies, arthritis, immune complex-mediated glomerulonephritis, and pulmonary capillaritis^182,183. Finally, there is a female bias with the experimental autoimmune encephalomyelitis model of MS ¹⁸⁴.

X-linked immunity genes that are dosage sensitive in autoimmune disease

There are a number of X-linked immunodeficiencies, thus underscoring the importance of appropriate X-linked gene dosage for immune function and health.⁷⁹ Accordingly, X-linked genes that are aberrantly expressed in immune cells of patients with female-biased autoimmune disease suggest perturbations in the regulation of XCI contributes to this bias. Abnormal X-linked gene expression has been observed in both adaptive and innate immune cells from patients with autoimmune rheumatic diseases, including SLE, SS, and SSc (Table 2). Interestingly, transgenic mouse models overexpressing some X-linked immune related genes have altered immune function and result in features of autoimmune disease (see Table 2). The X-linked gene CD40LG is primarily expressed by activated CD4⁺ T cells, and encodes a receptor that binds to CD40, which is expressed by dendritic cells, B cells, and endothelial cells. CD40LG is aberrantly overexpressed in T and B cells from female SLE^80,81 and SSc⁸² patients relative to healthy female controls. While the biological significance of this overexpression in these patients is not well-established, transgenic overexpression of Cd40lg in mice results in autoimmune disease, with increased IgG antibodies, chronic inflammation, glomerulonephritis, thymic atrophy, and increased lethality⁸³. CXCR3 is an inducible chemokine receptor that functions in adaptive immune responses by regulating Th1 cells and T cell trafficking. CXCR3 is overexpressed in circulating CD4⁺ T cells of female patients with SLE and the proportion of CXCR3⁺ CD4⁺ T cells is increased in the urine and kidneys from patients with lupus nephritis^81,84. In SS patients, there are also elevated numbers of infiltrating CXCR3⁺ CD3⁺ T cells in salivary gland tissue, and the CXCR3 ligands CXCL9 and CXCL10 are overexpressed⁸⁵. BTK is a kinase associated with the B cell receptor and its activation induces a signaling cascade required for B cell proliferation, activation, and survival. SLE patients (samples were not distinguished by sex) with active lupus nephritis overexpress BTK in peripheral blood mononuclear cells⁸⁶, and Btk overexpression in mice increases germinal center B cells, positive ANA staining, and immune complex deposition in kidneys ⁸⁷. Decreasing the activity of Btk through small molecule inhibition in both a spontaneous (NZB/W F1 mice) and an inducible mouse model of lupus reduces lupus-like disease phenotypes⁸⁸. FOXP3 is a transcription factor important for differentiation and maintenance of regulatory T cells (Tregs). SLE and SSc patient Tregs express less FOXP3, which likely compromises Treg function, and while there are conflicting reports regarding the abundance of Tregs in the peripheral blood of patients with SLE and SSc, the available evidence suggests that Tregs may be dysfunctional in these diseases ^89-91. Tregs in the skin of SSc patients exhibit abnormal production of Th2 cytokines, indicative of tissue specific Treg cellular effects that may influence localized dysfunction and contribute towards fibrosis⁹². Overexpression of Foxp3 is protective against renal dysfunction in a mouse model of accelerated crescentic glomerulonephritis due to increased Treg number and function, which blocks Th1, Th2, and Th17 responses systemically⁹³.

Table 2. Immunity-related X-linked genes that are dosage-sensitive for autoimmune diseases.

Examples of X-linked genes that function in the adaptive and innate immune systems are shown.

Dosage sensitive X-linked gene	Function	Evidence for XCI escape	Gain/loss of function mouse models & phenotypes	Aberrant expression in autoimmune patients
Adaptive Immune System
CD40LG	Type II membrane protein expressed primarily on CD4+ T cells Binds to CD40+ on B cells, ultimately leading to B cell activation, differentiation, and antibody production1	Exhibits variable XCI escape in human fibroblasts2 CD40LG locus is regulated by DNA methylation, and CD4+ T cells treated with DNA methylation inhibitors increases CD40LG expression, suggesting reactivation from the Xi (direct evidence lacking)1,3	Cd40lg overexpression in mice results in chronic inflammation, apoptosis-mediated thymic atrophy, glomerulonephritis, increased IgG antibodies, and increased lethality4,5	Aberrantly overexpressed in SLE patient CD4+ T cells, CD8+ T cells, and B cells with active disease compared to healthy controls and patients in remission1,6 Female SLE patient CD4+ T cells have increased CD40LG expression compared to male SLE CD4+ T cells 1,6 Patient with CD40LG duplication had autoimmune disease (splenomegaly, autoantibodies) 7
CXCR3	Chemokine receptor expressed on effector T cells that plays an important role in T cell trafficking to sites of infection8 During inflammation, Cxcr3 is overexpressed in infiltrating CD4+ and CD8+ T cells9	Silenced by XCI in human fibroblasts2 CXCR3 regulated by DNA methylation, and CD4+ T cells treated with DNA methylation inhibitors may reactivate CXCR3 on Xi (only indirect evidence)10 Biallelic Cxcr3 expression in CD3+ T cells from mice infected with Leishmania mexicana11		CXCR3 is overexpressed in CD4+ T cells in urine and kidneys from patients with active lupus nephritis 12 Female SLE CD4+ T cells have increased CXCR3 mRNA and protein levels13 Elevated infiltrating CXCR3+ CD3+ T cells in salivary gland tissue of SS patients14
BTK	Essential signaling component of BCR, where Btk activation initiates signaling cascade for cell proliferation, activation, and survival	Silenced by XCI in human fibroblasts2	Btk inhibition in NZB/W F1 mice or inducible mouse model of lupus eliminates lupus-like disease symptoms: lower anti-dsDNA antibodies, reduced germinal center and plasma B cells, reduced complement deposition in kidneys, and reduced inflammatory cytokines15,16 Btk overexpression in mice increases numbers of germinal center B cells, plasma cells, increases ANAs, and immune complexes in kidneys17,18	BTK is overexpressed in PBMCs from patients with active lupus nephritis19
FOXP3	Expression required for establishment and maintenance of Tregs, for Treg function and identity	Silenced by XCI in human fibroblasts2	Foxp3 overexpression is protective again renal dysfunction20	SLE patients with increased disease severity exhibit greater numbers of CD4+ FOXP3+ T cells, possibly due to increased T cell activation 21 SLE and SSc patient Tregs express lower mRNA and protein levels of FOXP3, which may compromise Treg function22,23 SSc patients have reduced numbers of Tregs23
OGT	Glycosyltransferase enzyme that catalyzes the addition of O-GlcNAc modification to proteins Required for T and B cell activation		Ogt overexpression in B cells results in enhanced activation, through increased O-GlcNAcylation of BCR dependent transcription factors24 Ogt deletion in B cells impaired B cell homeostasis, activation, and antibody production25	Elevated OGT mRNA and protein levels (and hypomethylation of OGT promoter) in female SLE patient CD4+ T cells compared to male SLE T cells 13
Innate Immune System
TASL (Cxorf21)	IFN response gene that is part of the endolysosomal TLR machinery26,27	Variable XCI escape in human lymphoblastoid cells2		Female biased increased expression in lymphoblastoid cell lines and thyroid tissue 26 TASL is risk variant for SLE28 Positive correlation between TASL protein abundance and SLEDAI score in SLE patients >35 years of age26
TLR7	Endosomal pattern recognition receptor that binds guanosine- and uridine-rich single-stranded RNA and induces MyD88-dependent activation of type I Interferon (IFN)29,30	Variable escape (40-55%) in healthy human B cells, monocytes, plasmacytoid dendritic cells (DCs)31 Biallelic expression in SLE patient B cells32 Variable escape in human pDCs; biallelic TLR7 cells also had elevated expression of IFN -beta and IFN-alpha family members 33,34 Variable escape in mouse plasmacytoid DCs; increased biallelic expression in disease NZB/W F1 mice35	Male BXSB-Yaa mice have two copies of Tlr7, and exhibit accelerated lupus-like disease, with antinuclear antibodies (ANAs), glomerulonephritis, and splenomegaly36,37 Tlr7 overexpressing mice develop spontaneous autoimmunity, with increased autoantibody production, autoreactive lymphocytes, splenomegaly and lethality.38 Tlr7 gain-of-function mice (TLR7Y264H) have increased responsiveness to guanosine and 2’,3’-cGMP, enhanced TLR7 signaling, increased survival of activated B cells, and cell-intrinsic expansion of age-associated B cells and germinal center B cells 39	Biallelic expression of TLR7 enhances cell responsiveness to TLR7 ligands and leads to an increase in class switching events (human B cells)31 Increased TLR7 expression in female SLE patients relative to controls40 SLE patients with a TLR7 gain-of-function mutation (Y264H) exhibit elevated ANAs, and features of autoimmune disease including refractory autoimmune thrombocytopenia, neuromyelilits optica, and inflammataory arthralgias41

X-linked immune genes also exhibit dosage sensitivity and are overexpressed in female-biased autoimmune diseases. TLR7 encodes an endosomal pattern recognition receptor that induces Type I IFN production and activates IFN signature genes. Increased TLR7 expression was observed among female pediatric SLE patients⁹⁴, perhaps in part due to its ability to escape XCI and exhibit biallelic expression in immune cells, such as B cells³¹. Indeed, TLR7 exhibits variable escape from XCI in human immune cells and B cells from female NZB/W F1 mice⁷³, suggesting that aberrant overexpression resulting from XCI escape is a mechanism that may contribute to the pathogenesis of female-biased autoimmune disease. Biallelic expression of TLR7 in human female B cells increases cell responsiveness to TLR7 ligands and increased class switching compared to monoallelic TLR7 expressing B cells³¹. The resulting increase in TLR7 signaling is likely biologically significant. Recently, a gain-of-function mutation in TLR7 was identified in a young female with SLE. Introduction of this mutation into BL6 mice results in the spontaneous development of lupus-like disease. ⁹⁵. Mouse models with two or more copies of Tlr7, including the BXSB-Yaa strain (BOX 3), have increased autoantibody production, autoreactive lymphocytes, and glomerulonephritis^96,97. TRL7 exhibits variable escape from XCI in human immune cells and B cells from female NZB/W F1 mice⁷³, suggesting that aberrant overexpression resulting from XCI escape is a mechanism that may contribute towards female-biased autoimmune disease. TASL (CXorf21), an X-linked adaptor protein critical for Tlr7 signaling is aberrantly overexpressed in SLE patient lymphoblastoid cell lines compared to healthy controls and may escape XCI in immune cells ⁹⁸. Thus, abnormal overexpression of TLR7 and TASL due to potential perturbations with XCI escape could represent a mechanism which contributes towards increased IFN signaling in female individuals with autoimmune disease. Retrospective identification of additional disease-related genes abnormally expressed from the X chromosome is challenged by the omission of reads from the sex chromosomes in microarray and transcriptional profiling datasets when normalizing samples between the sexes in many publications. However, many patient datasets for autoimmune diseases with a strong female bias include a female-majority of patient samples and therefore retain X-linked transcript information, permitting future re-analyses of the data to identify additional differentially expressed X-linked genes in patient samples^56,59. Future translational investigations in well clinically characterized patient cohorts will be instrumental in further defining the spectrum of X-linked genes exhibiting dysregulated expression in a sex-specific and disease-specific manner.

Sex differences with immune responses to pathogens

Sex differences in the incidence or severity of infectious diseases have also been observed across a variety of distinct pathogens (Table 3). Biological sex impacts pathogen replication and transmission, as well as the host immune response to the pathogen. The sex bias of infectious disease is typically explained by the stronger female immune response, which results in higher levels of inflammation. While sexual dimorphism among infectious diseases is likely explained in part by behavioral differences (lifestyle choices, exposures, access to healthcare, etc), underlying physiological differences, both hormonal and genetic, may have an even greater impact. The hormonal and genetic contributions responsible for the sex-biased responses to a variety of pathogens are summarized in Table 3 but have not been well characterized (particularly the genetic component). Mouse and rodent models often recapitulate the observed sex bias in humans (Table 3) and have been used to identify sex-specific molecular pathways resulting from infection of some pathogens. Pathogen type also influences infection incidence and disease severity, often exhibiting a sex bias in both humans and rodent models. Males typically exhibit greater incidence for most bacterial, parasitic, some viral, and fungal infections (Table 3); females have greater incidence of parasitic worm infections^99-101. Some pathogens, including HIV, Ebola, M. tuberculosis, Cryptosporidium, and Sars-CoV-2, exhibit altered rates of replication, transmission, or incidence of multiple infections of different pathogens between the sexes, which likely influence incidence rates ^102-107. Continued use of relevant mouse models, when they recapitulate some features of human disease, with these different pathogens is likely to reveal genetic and epigenetic mechanisms that contribute to the sex differences following infection by these specific pathogens.

Table 3. Sex differences across various classes of pathogens (virus, bacteria, worms, parasites, fungi) for humans and mouse/rodent models.

Pathogen prevalence and incidence, pathogen loads and disease severity, and mouse models exhibiting sex bias with infection or disease are shown in each column. Information regarding hormonal or genetic influences on the sex bias is provided (when available) for each pathogen.

	Pathogen	Pathogen Prevalence and Incidence	Pathogen Intensity/Load; Resulting Disease Severity	Mouse Models & Sex Biased Mechanisms/Pathways
VIRUS	Coronaviruses; SARS-CoV; SARS-CoV-2	Males have higher incidence1-10	Adult males (45-79yrs) have higher mortality5,6 9 Females more likely to be diagnosed with Long Covid syndrome11,12	Male mice are more susceptible to SARS-CoV infection13-15 Male mice have increased accumulation of inflammatory macrophages and neutrophils in lung (SARS-CoV)3,13 Estrogen receptor signaling is protective after SARS-CoV-infection13,16 SARS-CoV-2 entry receptor Ace2 is biallelically expressed in females, which might reflect its dual role in mediating viral replication vs renin-angiotensin-aldosterone system 4,17
	Influenza	Infant males and older adult males have increased incidence18-23	Males (pre-pubescent, elderly) have high mortality.19,23 Females (pre-menopausal; pregnancy) also have high mortality18,24-26	Male mice have greater disease severity and this increases with age27 Female mice have greater mortality28 Female mice are more protected following influenza vaccination29,30 Low levels of testosterone in male mice correlate with poorer protection27,31-33 but estriol protects female mice from severe disease and decreases influenza replication34,35
	Hepatitis A	Males have more hospitalizations36	Males have higher mortality36	n/a
	Hepatitis C	Similar incidence rates between males and females37	Males have greater disease severity (HCV-associated cirrhosis) 37 Females more likely to clear virus37	n/a
	West Nile Virus	Higher percentage of affected males in this case study38	Similar initial viremia38 Females have more symptoms38 Males have longer lived cytokine response38 Males have increased mortality38 Males more likely to develop neuroinvasive disease38	n/a
	Human Immunodeficiency Virus (HIV)	Females have higher incidence 39 Females have higher levels of immune activation and interferon signature gene expression40	Females have lower viral loads in early stages of infection (but comparable in advanced stage)40,41 No sex difference with disease progression or clinical outcomes	Male-to-female transmission appears more efficient than female-to-male transmission42
	HCMV	Females (post-puberty pre-menopausal) have higher incidence of HCMV seroprevalence43		n/a
	Herpes Simplex Virus (HSV)	Females have higher prevalence44,45		Female mice more susceptible to infection28 Female mice have higher HSV titers in brain tissue28 Higher mortality in male mice46 Ovariectomy of female mice or estrogen treatment of male mice eliminated sex differences after infection28 Sex-biased survival differences depend on type I IFN signaling and DAP12 signaling28
	Coxsackievirus			Male mice have increased mortality28,47 Males develop more severe cardiac inflammation due to Th1 response47 Females are more resistant; exhibit predominantly Th2 responses47
	Ebola		Males have higher mortality48	n/a
	Measles	Females (age 45-64) have higher incidence49; Males (age 0-45) have higher incidence50	Females (ages 0-49) have higher mortality, particularly post-puberty pre-menopausal49.	n/a
	Respiratory Syncytial Virus (RSV)	Males have higher incidence51,52, but a metanalysis of acute respiratory infections in Africa did not identify sex as a factor in RSV prevalence53	Males have higher rates of hospitalizations51	Male neonatal mice have higher viral gene expression after RSV infection, and delayed viral resolution54 After early-life RSV infection, male mice exposed to allergen have severe allergic exacerbation (female mice are protected). TSLP pathway (which impacts IFN-beta production) alters male immune environment after neonatal infection54
BACTERIA	Heliobactor pylori	Males have greater H. Pylori sero-prevalence55,56 Infection has a male bias55	Males have more severe inflammation, atrophy, and intestinal metaplasia57	Male mice are more susceptible Males have higher colonization levels for babA virualence factor of H. pylori58 Male mice treated with estradiol produce less IFN-gamma and IL-1-beta, and increased IL-10 and Th2 associated IgG1 levels58 Estrogen is protective against gastric lesions; ovarectomy increases severity of gastritis and gastric cancer58
	Pseudomonas aeruginosa	Males have higher prevalence	Female CF patients have worse disease prognosis59	Female mice more susceptible to infection60 Females mount strong inflammatory response in lungs60 Estradiol upregulates expression of secretory leucoprotease which inhibits Tlr-dependent IL-8 release in bronchial epithelial cells during P.aeruginosa infection
	Salmonella	Higher incidence rates in male children for salmonellosis (up to age 15) 61 Females have higher incidence rates (ages 15–44 and 45–64)		n/a
	Chlamydia trachomatis; Chlamydia pneumoniae	Males have greater prevalence (C. pneumoniae)62 Females have higher prevalence (C. trachomatis)63	Males have higher levels of C. pneumoniae64 Females have higher infection rates because they are more likely to be screened (C.trachomatis)65 Estrogen levels correlate with chlamydial load64 Chlamydia-induced arthritis more common in men66	n/a
	Brucella spp.	Males have higher incidence67 No sex bias with prevalence	Males more likely to develop Brucellosis68	n/a
	Borrelia burgdorferi (Lyme disease)	Males have higher incidence (USA 1992-1998)69 Females >45 greater incidence (Sweden 1992-1993)70,71 Females more likely to be re-infected after 5 years.	Males have more hospitalizations and likelihood for disseminated disease72 Lyme neuroborreliosis is more common in female patients73 Females have increased production of IFN-gamma, IL-4, IL6, IL-10, TNF-alpha70	Male mice have more infected tissues and higher spirochete loads74
	Mycobacterium tuberculosis	Males have higher incidence (male/female 1.7)75	Males exhibit higher mortality rates (global)76 Pregnancy increases risk of disease complications77 Females usually have less symptoms64	Male mice have accelerated disease progression, increased morbidity and mortality78 Males have higher M. tuberculosis loads78 Testosterone treatment increases susceptibility to infection
	Mycoplasma pulmonis			Male mice more susceptible79 Male mice develop more severe disease in lung parenchyma79 Removal of reproductive organs reduced disease severity79
	Coxiella burnettii	Males have higher incidence80	Human males more likely to become sympotomatic with Q fever (symptoms include fever, granulomatous hepatitis, myocarditis, pericarditis, pneumonia)64,80 Pregnancy increases risk for persistent infections, and impaired immunity negatively impacts pregnancy64	Male mice have higher bacterial loads64 Estrogen treatment of ovariectomized mice reduces bacterial loads and granulomas81 C. burnetti infection results in sex-specific gene expression profiles: males upregulate IL-10 and interferon-gamma production; females exhibt altered expression of circadian rhythm genes.82
	Campylobacter spp.	Males have higher incidence64		Males are more susceptible to infection and colonization83 Males have higher shedding rates
	Clostridiodes difficile	Females have higher incidence84 Females have increased risk of recurrent infection57	Increased disease severity in pregnant and peripartum females84	Progesterone and estrogen intermediates can inhibit spore germination in mice57
	Listeria monocytogenes	Females have higher incidence rates of invasive listeriosis85 Pregnant females have higher incidence64 Among older individuals, males have 2-4 higher incidence rates85	Pregnant females and older males have greater incidences of invasive disease85 Older males have increased fatality rates85	Female mice more susceptible to infection and exhibit greater lethality86 Females have higher bacterial load; Infected females have increased IL-10, which inhibits Th1 differentiation and Th1-derived cytokines86 Estrogen treatment reduced IL-12, IFN-gamma, TNF-alpha; increased IL-4 and IL-10; reduced monocytes and lymphocyte accumulation at infection87
	Legionella pneumophila	Males have higher incidence	Males more likely to develop legionellosis and males more likely to have poor prognosis64	n/a
	Leptospira spp.	Males have higher incidence64		n/a
	Francisella tularensis	Males have higher incidence65		No sex difference with susceptibility Vaccinated female mice are more resistant to infection, with lower bacterial burdens, less tissue inflammation, and less proinflammatory cytokine production, and more Ft-specific antibodies in serum and lung85
	Escherichia coli	Females have higher incidence64		No sex difference with enterohemorrhagic E.coli disease in mice
	Treponema pallidum (syphilis)	Males have higher incidence88,89		n/a
	Neisseria gonorrhea	Males have higher incidence64 Infected males may also have increased expression of gonococcal antimicrobial resistance genes90	Most females lack symptoms64 Complications in males include epididymitis, infertility, prostatitis, seminal vesiculitis91 Elevated progesterone promotes gonococcal infection (human cervical epithelial cells)	Estrogen treated mice have increased susceptibility to gonococcal infection92
	Streptococcus pneumoniae	Males have higher incidence for all types of pneumonia64 Males (pre-puberty) have higher incidence	Males have greater hospitalization rates and increased mortality64 Males more frequently diagnosed with Legionellosis (1.7:5 male to female)93	Male mice are more susceptible & have more severe disease94 Males exhibit increased pro-inflammatory cytokines (IL-6, IL-17A, IFN-gamma)94 Estrogen is protective, regulating macrophage activity (for pneumococcal pneumonia)95
	Yersinia enterocolitica	Males have higher incidence for Yersiniosis96	Males have higher levels of IgG4 antibodies for Yersinia outer membrane proteins, which is associated with anti-inflammatory response that is resistant to treatment57	n/a
	Sepsis: Staphylococcus, Escherichia coli, Pseduomonas, etc	Males have higher rates of sepsis and septic shock65 Males more likely to develop sepsis after trauma or surgery65	Conflicting results for a sex bias with mortality97	Male mice develop greater inflammatory response, producing more pro-inflammatory cytokines64 Males have more severe sepsis-induced cardiac dysfunction85 Estrogen is protective, and female mice produced protective antibodies in response to estrogen; estrogen-driven antibodies were maternally transferrable to offspring98
WORMS	Pork tapeworm (Taenia solium) Neurocysticercosis	Females have higher incidence in some countries (Nigeria, Tanzania, Guatemala)99 Females have more transitional cysts in brain (Ecuador)100 No sex difference with incidence in Vietnam99	Female patients have greater number of transitional cysts100	Estrogen increases parasite loads and androgens decrease loads in mice, either acting directly on the worm’s reproduction or by altering host’s immune response to favor Th2 or Th1 pathways (Taenia crassiceps)101
	A. Lumbricoides	Females have higher incidence102		n/a
	Schistosoma masoni	Males have higher prevalence for infection103		Female and castrated male mice have greater morbidity after Schistosoma infection104 Female mice have higher worm loads104 Testosterone is protective for Schistosoma mansoni infections; female mice treated with testosterone had reduced worm burdens (if treated before infection)104
PARASITES	Plasmodium falciparum (malaria)		Male patients have greater disease severity105	n/a
	Cryptosporidium	Males have higher incidence106	Male patients have greater incidence of hospitalizations107	n/a
	Entamoeba histolytica (amoebiasis)	Asymtopmatic infection rates are the same across sexes108	Invasive amebiasis predominantly affects males; males have higher rates of invasive disease108 Males have higher incidence of hepatic amebiasis109	Testosterone treatment induces proinflammatory responses in mouse (& human) classical monocytes, with increased production of CXCL1 and TNF109
	Leishmania	Males have higher incidence even when accounting for exposure110 Adult males have higher incidence of cutaneous leishmaniasis111 Childhood cutaneous leishmaniasis does not exhibit a sex bias112 Males have higher incidence and greater risk ratio of visceral leishmaniasis113 No sex bias for childhood cutaneous leishmaniasis114	Male patients exhibit higher rates of treatment failure and adverse effects110	Male mice have higher parasite burdens following infection (L.infantum)115 Male mice express higher levels of IL-10 and TNF after infection and exhibit greater disease severity115 Male mice (BALB/c congenic strains) are more susceptible to subcutanteous L.major, and exhibit more severe disease110,116 Female mice heal small lesions following L. Mexicana infection, yet male mice exhibit persistent lesions, dependent on IL-4 levels117 Male hamsters have increased disease severity and parasite burden with L. viannia infection. Testosterone-treated female animals had larger lesions than untreated females. Disease severity correlated with increased expression of IL-4, IL-10, and TGF-beta118 X-linked Cxcr3 is biallelically expressed in T cells of female mice and contributes to increased cytokine production119
	Toxoplasma gondii		Sex differences with infection-induced behavioral changes and personality shifts120	Female mice are more susceptible to infection and have higher cyst burdens121 Female mice exhibit higher mortality after acute infection121 Male mice produce higher TNF-alpha after day 10 of infection; female mice mortality did not correlate with lower TNF-alpha levels. Male mice produce higher IFN-gamma and IL-10 early during infection121
FUNGI	Aspergillus fumigatus	Males have higher incidence (invasive pulmonary aspergillosis)122 Male bias with prevalence, incidence and severity123 Males more susceptible to infection123		Female mice have higher levels of immune components (antibody titers, neutrophil eosinophil, and lymphocyte cells) after infection124
	Cryptococcus neoformans	Males have higher incidence125 Males more affected than females125		Female mice express more cytokines in plasma and increased expression of TNF-alpha, interferon-gamma in spleen126 Increased lethality for young male mice126 Survival and fungal loads are similar between male and female mice126
	Paracoccidioides brasiliensis	Males have greater incidence (10:1 male to female, Latin America)127	Male patients have faster disease progression128	Male mice are more susceptible129 Macrophages from infected female mice exhibit greater fungicidal activity, with higher nitric oxide production129 Estrogen is protective following P. brasiliensis infection, as castrated male treated with estradiol have higher levels of IFN-gamma and lower levels of IL-10 compared to normal males. Ovariectomized female mice treated with testosterone produce less IFN-gamma and more IL-10 compared to normal female mice after infection129
	Microsporum, Trichophyton, epigermophyton (Tinea or Dermatophytosi)	Males have higher incidence 130		n/a
	Candida albicans	Females have higher incidence (oral candidiasis)131 Females have higher incidence (candida onychomycosis), with 3/4 females (childbearing age) infected at least once in their life; and 1/10 females having a recurring event131	Male patients with seropositivity for C.albicans have increased odds for schizopherenia132 Female patients with seropositivity for C.albicans have increased odds for lower cognitive scores132	n/a

Sex differences in response to vaccination

Females often have greater antibody responses, and typically experience more adverse reactions to vaccinations¹⁰⁸. Activation of the innate immune system immediately following vaccination often results in localized inflammation of the injection site. Subsequent activation of the adaptive immune system is also critical for generating an effective memory response to inactivated viruses or viral particles. Females often develop more inflammation around the vaccine injection site, which may result from sex differences with innate immune activation and produce stronger class-switched antibody profiles for vaccines against influenza, smallpox, measles mumps and rubella, yellow fever, hepatitis A and B, and herpes simplex (Table 3). Female mice injected with inactivated virus produce higher antibody titers and higher numbers of germinal center B cells and CD8+ and CD4+ T cells in lymph nodes compared to male mice¹⁰⁹. Because females often have stronger responses to viral and bacterial vaccines and have more severe reactions, it has been suggested that female-specific reductions with vaccine doses should be considered³².

X-linked genes involved in sex-biased responses to pathogen infections and vaccination

Accumulating evidence suggests that XCI escape of some X-linked immunity genes in innate and adaptive immune cells contributes towards increased female protection from bacteria and parasites. Transcription of ACE2, one of the receptors utilized by coronaviruses for cellular entry, is influenced by both genetic and hormonal factors^{106,107,110,111}, and ACE2 escapes XCI in human cells and mouse AT2 cells inside alveoli of the lung^71,112. One study using Cxcr3 dual reporter mice found that Cxcr3 escapes XCI in activated T cells, and that following Leishmania infection, biallelic Cxcr3-expressing cells produce more IFN-gamma, IL-2, and CD69 than do monoallelic Cxcr3 expressing cells¹¹³. In general, males have less potent antiviral responses (Table 3), which may be influenced by sex-dependent differences in UTX/Utx expression levels in NK cells that result in increased numbers of NK cells with reduced functionality in males compared to females¹⁶. NK cells are necessary for antiviral responses to various viruses including cytomegalovirus (CMV), and Utx deletion increases lethality following CMV infection¹⁶. It is likely that in addition to Utx and Cxcr3, other X-linked genes that escape XCI in immune cells impact the response to pathogens that exhibit sex biases in infection (Table 3). The identification of these genes may reveal one pathway which underlies the mechanistic basis for sex-biased immune protection in response to a variety of pathogens.

The importance of X-linked immunity related genes for sex-biased responses to pathogens is underscored by observations of sex-specific phenotypes in genetic deletion models or in patients with specific mutations following infections. For example, the X-linked gene Ddx3x, an RNA helicase that impacts RNA processing and transcription and regulates type I IFN production following viral and bacterial infections¹¹⁴, escapes XCI in female cells. Although male cells contain a Y-linked functional homolog (Ddx3y), conditional deletion of Ddx3x using VAV-Cre impairs the ability of male mice (female-specific deletion is embryonic lethal) to respond to listeria monocytogenes¹¹⁵. Consistent with these findings, female bone marrow derived macrophages (BMDMs) with homozygous Ddx3x deletion are unable to restrict listeria growth¹¹⁵. Ddx3x deletion in BMDMs exhibits sex-specific gene expression patterns, and female BMDMs lacking Ddx3x have greater reductions in cytokine (IL-1, IL-6, Il-12, and TNF-alpha) and chemokine expression compared to male Ddx3x mutants¹¹⁵. Additional research investigating other XCI escape genes, especially those that encode for chromatin modifying enzymes, through the use of gain and loss of function mutations in specific immune cell types will reveal other potential genetic contributors to the immune responses to pathogens that exhibit a sex bias (Table 3).

Tuberculosis exhibits a male bias in humans and mouse models, and TB is a leading cause globally of a disease caused by a bacterial pathogen (Table 3). Mutations in two X-linked genes, CYBB and IKBKG (NEMO), increase susceptibility to mycobacterial disease. Missense mutations in CYBB, which encodes for the gp91 subunit of the NADPH oxidase complex, results in a hypomorphic protein that impairs respiratory burst activity in macrophages necessary for protection from mycobacteria¹¹⁶. IKBKG/NEMO is the regulatory subunit of the inhibitor of kappa B kinase (IKK) complex, which regulates canonical NF-kappa B activation of genes involved in inflammation and immunity. While IKBKG/NEMO mutations result in incontinentia pigmenti, ectodermal dysplasia, and various immunodeficiencies, there are some NEMO mutations that increase predisposition of male patients to mycobacterial infection through reduction of IL-12 and IFN-gamma production¹¹⁷.

Despite ample evidence for sex biased responses to vaccinations, few X-linked genes that contribute to this finding have been identified. B cells from female immunized mice do express higher levels of Tlr7¹⁰⁹, which may result from increased levels of XCI escape and expression from the Xi in response to flu vaccines. Tlr7 can recognize single-stranded viruses including SARS-CoV-2, and may contribute to Type I IFN production^118,119. Loss of function TLR7 mutations in human males reduce Type I IFN levels and prevent interferon-stimulated gene (ISG) induction, resulting in severe COVID-19 disease^118,120-122. RNA vaccines for SARS-CoV-2 induce adverse which exhibit sex biases ¹²³, as human females are more susceptible to thrombosis thrombocytopaenia syndrome (from adenoviral vectors)¹²⁴, yet young human males appear to be more susceptible to myocarditis and pericarditis from mRNA vaccines¹²⁵.

CONCLUSIONS AND PERSPECTIVES

Biological sex is an important factor for immune responses and immune health, and the importance of understanding hormonal and genetic contributions to sex differences in immunity is increasing. In 2009, about >60% of immunology-related research publications using animal models lacked information about biological sex, and over half of human immunology publications included male and female samples, yet >90% of these publications lacked sex-specific analyses¹²⁶. Inclusion rates for both sexes in immunological research was 16% in 2009, and increased to 46% by 2019, but suggests that more work is needed to ensure that sex as a biological variable is addressed in future immunologic research. While the influence of sex hormones on sex-specific immune cell function and cytokine production following infection has been investigated for some of the pathogens in Table 3, the contribution from the X chromosome in biased immune responses is not well understood. Future experiments examining XCI escape genes in specific immune cell populations, and use of genetic gain and loss of function experiments of immunity-related X-linked genes will reveal important mechanisms of sex differences in infection susceptibility and resulting disease severity. While the sensitivity of X-linked gene dosage for some female-biased autoimmune diseases has been examined, whether X-linked gene dosage influences immune responses to pathogens is not well known. Understanding the origins of the sex biases using rodent models that recapitulate sex differences observed in clinical studies of infectious and autoimmune disease are likely to inform sex-specific treatment strategies that could improve patient outcomes.

The interplay of the sex chromosomes and how they influence sex hormones is likely to influence sex differences in immune responses and sex-biased autoimmune disease. In addition to the Y-linked Sry gene, which functions in primary sex determination and the formation of the male gonads responsible for testosterone production, the influence of other X- and Y-linked genes on sex hormones in the context of sex-biased immune responses has not been carefully investigated. Given the large number of steroid receptor binding sites across the genome, including on the X chromosome, it is likely that inflammatory pathways resulting from infections and autoimmunity will promote sex hormone receptor binding to promoters of X-linked immunity-related genes, perhaps promoting increased XCI escape in female immune cells. Moreover, AR and ER expression levels change with cellular activation in immune cells¹²⁷, potentially impacting X-linked gene expression and contributing to sex-biased gene expression. Hormonally-induced expression changes of X-linked genes are likely to occur on the active X, yet it is possible that XCI escape genes on the Xi could be differentially regulated by AR and ER in the context of immune activation. Investigation of allele-specific transcriptional changes in response to cellular activation and inflammation are necessary to determine the complex interplay of sex hormones and X-linked gene expression for sex differences in immune responses. Understanding the genetic and hormonal contributions for sex differences with immune health will undoubtedly result in novel therapeutic approaches for effective precision medicine.