Sex Differences in Poststroke Inflammation: a Focus on Microglia Across the Lifespan

Microglia exists in region, time, and context-dependent states in the healthy adult central nervous system, presenting differential phenotypic characteristics and specific transcriptomic profiles in different brain areas.^11,12 However, knowledge on sex differences in microglia states in healthy brain is scarce.

Most of the information on sex differences in microglia are derived from studies performed in rodents. Sex-related differences in microglial density and morphology exist already in early the development and, although these features change during the lifespan, some of the differences remain in the adult brain. Only few studies have addressed sexual dimorphism in adult rodents in different brain regions. Some studies show that females have a higher density of microglia in specific hippocampal regions compared with males,^13,14 whereas others show a higher density of microglia in the cortex, amygdala, and hippocampus of male rodents.^15–17 Not only the density but also microglial morphology (eg, number of branches, processes per branch, and branch thickness) is region-dependent. For instance, microglia in the dorsal hippocampi have a higher total number of processes and more processes per branch compared with the prefrontal cortex in male rats. On the contrary, microglia are reported to have a higher total number of processes and more processes per branch in the prefrontal cortex compared with the dorsal hippocampi in female mice.¹⁸ Microglia are more ramified in the prefrontal cortex in female rats compared with males,¹⁹ while microglia in hippocampus are more ramified in males compared with females.²⁰ The contradictory results may be partly due to differences in the analyzed brain region, age, and species, but also due to the microglia markers and techniques that were used.²¹ This fact limits the comparison between studies, urging the researchers to use harmonized techniques to explore the sex-differences in microglia (Table 1).

Of note, analysis of the expression of molecular patterns or transcriptomic programs in rodents reveals sex differences that cannot be quantified by morphometric analysis. Single-cell transcriptomics emerges as a powerful tool to analyze changes in cellular populations and states, yet, studies evaluating sex-differences at the single cell level are largely missing. In the recent years, a growing number of studies have demonstrated differences in the bulk transcriptomic and proteomic profiles of microglia depending on the brain region^12,22 or age.²³ Focusing on sex differences, transcriptomic analysis revealed 1109 differentially expressed genes in the hippocampus, 55 in the cerebral cortex and 46 in both regions between male and female mice, demonstrating brain region to be an important factor in microglial sex-diversity.¹⁷ Whole-brain microglia showed increased expression of 204 genes related to inflammatory processes in male mice compared with females, including regulation of cell migration, chemotaxis, and cytokine production.²⁴ In agreement with transcriptomic studies, NF-κB (nuclear factor kappa B) is transcriptionally more activated in male microglia compared with female microglia.²⁵ This suggests that male microglia are inherently more prone for inflammatory activation in pathological conditions compared with female microglia, consistently with higher expression of interferon-stimulated genes in males.²⁶ On the other hand, females show higher expression of genes associated with development or morphogenesis that are under the control of transcription factors linked to the inhibition of the inflammatory response.²⁴ Female microglia also express more sensome-related genes,²⁶ a cluster of transcripts that code for proteins involved in sensing of endogenous ligands and microbes.²⁵ In addition, proteomic profiling of microglia has demonstrated 263 proteins with significantly higher expression in males and other 96 proteins, preferably expressed in females.¹⁷ All these studies provide evidence that female and male microglia express different protein and RNA patterns even in the healthy brain which may contribute to the sex-differential response of microglia to stroke in rodent models (Table 2).

In addition, there are differences in microglia across species. Microglial morphology is relatively conserved between humans, macaque, and rodents, but human microglia are more heterogeneous and display an organization in different subtypes, as revealed by single-cell RNA-seq.^12,29 These subtypes do not correlate with sex-specific features but add an additional layer of complexity to the analysis of sex differences.

Microglia are essential in responding to brain insults and play a central role in initiating and gearing the neuroinflammatory response.⁷ In the presence of a stimulus, such as stroke, these cells can exhibit a wide spectrum of activation states.³⁰ Traditionally, microglia were divided into M1 and M2 phenotypes, with M1 type cells secreting pro-inflammatory cytokines, proteins of the complement system, reactive oxygen species and reactive nitrogen species, and the highly phagocytic M2 type microglia releasing anti-inflammatory and immunosuppressing cytokines. While this classification has been used extensively in the literature, we omit from using this over-simplistic dichotomy, and rather refer to microglial states that are both stimulus and context dependent and form a continuity between different states.^30,31

The abovementioned sex differences in the gene transcription contribute to their differential responses to ischemic stroke. Differential stroke outcomes (short- and long-term) between sexes, as well as sex-specific inflammatory response are reported, which are critical to understand the consequences of sex-differences in microglial response to stroke and in the related outcomes (Table 3).

Sex hormones have a profound impact on the outcome of stroke.⁵⁶ There is clinical evidence demonstrating a higher stroke incidence and severity in women after the menopause,⁵ and for this reason, female sex hormones, specifically estrogen, have been attributed a cytoprotective role in ischemic stroke.^57,58 Several studies consistently demonstrate that estrogens drive the sex-differential protective effect in young females, which has been confirmed in ovariectomized female rodents subjected to stroke and the protection by estrogen therapies.^57,59,60 While the effect of estrogens is consistent across studies, the role of testosterone remains controversial with evidence of both detrimental and beneficial effects.^61,62 The fact that the beneficial effect of testosterone is dose-dependent⁶² suggests that it is the balance between the sex hormones that matters.⁶³ Besides this controversy, it has been suggested that estrogens have an anti-inflammatory role after stroke in which microglia may be involved.⁶⁴

Age is the main contributor for the outcome of stroke and has a complex and interactive effect with sex. Aging by itself is associated to changes in the immune system, such as a rise in inflammatory markers and diminished response to cytokine stimulation, which can modulate the stroke progression and outcome.⁷⁰ In experimental stroke, while young females show smaller infarct volume compared with males, the outcome is inverted in aged males.^16,40,48 It is still under discussion whether the worsened outcome of aged females compared with males is caused by the lack or dysregulation of estrogens, or due to a sex chromosome–linked features. Some studies highlighting the relevance of the X chromosome in aging suggest that these differences could rely on X chromosome dosage. The additional X chromosome is silenced in young female animals,⁷¹ thus no differences in the infarct volumes are observed between females with 1 and 2 X chromosomes (XX and XO).⁷² However, aging affects the stability of the silencing, and thus the X chromosome may escape from the inactivation, giving an extra dosage of X-related genes.⁷³ The Four Core Genotype mouse model is a useful tool to dissect the effect of X-dosage from sex hormones along lifespan. In this model, the testis-determining gene sry is deleted from the Y chromosome and inserted on an autosome, allowing the generation of XXM mice (a gonadal male with XX chromosomal complement) and XYF mice (a gonadal female with XY chromosome complement).⁷⁴ Manwani et al⁶⁰ utilized this model to demonstrate that gonadal hormones are responsible for the cytoprotection in young females compared with males. The authors show that both XXF and XYF (gonadal females) display reduced infarct volumes compared with XXM and XYM (gonadal males), which is not observed when comparing males with ovariectomized females. Moreover, estradiol treatment not only reverses this but also exerts protective effect in males suggesting that circulating estrogen is the main driver for the protection during reproductive period of life. However, age is a game-changer. McCullough et al⁴⁰ used the Four Core Genotype model in aged mice to demonstrate that aging changes the relative contribution of XX/XY chromosomes to ischemic damage when compared with young mice. XXF and XXM aged mice have significantly larger infarct volumes. Moreover, these mice show a more robust stroke-induced microglial activation pattern compared with XY female and XY male mice, revealing that the chromosomal complement shapes the inflammatory response as well as the outcome of stroke in aged mice.⁴⁰ The combination of these 2 studies provide a wide perspective of the relative contribution of XX/XY chromosomes along lifespan.

Focusing on microglia specific features in aged animals, sex differences are scarcely studied. Transcriptomic analysis through the lifespan show activation of inflammatory genes in both males and females, but these changes are more evident in females, resulting in a sharper sex divergence in the elderly.^27,28 These data suggest that the predisposition of pro-inflammatory signaling observed in young males is compensated by age in females, possibly via X chromosome escape as described above.⁵³ In fact, microglia from aged female animals express higher levels of pro-inflammatory markers (MHCII [major histocompatibility II]) compared with male microglia, together with an increase of pro-inflammatory cytokines in the serum (TNF-α and IL-1β).⁴⁰ Another example of this sexual divergence along the age is the differential expression of complement proteins in microglia of old females and males. The complement system is induced upon ischemic stroke and modulates of the poststroke inflammatory response.⁷⁵ Microglia are described to release several complement effectors after stroke, such as C1qa (complement 1qa)⁷⁶ and complement 3 (C3).⁷⁷ Aging induces the transcription of C1qa and complement 1qc (C1qc) genes and the translation of C1q protein at higher levels in females compared with males.²⁷ Similar to the complement system, there is a higher upregulation of the sensome genes in female microglia compared with males.²⁷ In addition, the importance of X chromosome escape has been highlighted by a recent study on 2 X escapee genes, the lysine demethylase 5C and 6a (kdm5c and kdm6a respectively), which code for histone demethylases.⁵³ These genes regulate microglial production of cytokines and are expressed at higher levels in microglia from aged females subjected to middle cerebral artery occlusion compared with males as well as in human postmortem tissue from stroke patients, suggesting that epigenetic modifications may play a role in poststroke sex differences.⁵³

All these studies highlight the fact that sex differences in the inflammatory response are not static along lifespan and raise the need of more in-deep studies of the sex-differential inflammatory response after stroke in aged animals to achieve better therapies.

Sex differences in treatment efficacy are common in stroke, yet, only a few studies have reported sex-differences on microglia-based therapies. Minocycline and other PARP (poly [ADP-ribose] polymerase) inhibitors⁷⁸ have been reported to alleviate poststroke neuroinflammation and ischemic damage only in male rats.^79,80 Interestingly, minocycline was ineffective in ovariectomized female with estrogen levels similar to males. Consistent with experimental studies, Amiri-Nikpour et al⁸¹ showed that oral administration of minocycline was effective in male patients with stroke but not in women. Meta-analysis of 7 randomized-controlled clinical trials showed that although minocycline seems to be a promising cytoprotective agent in patients with stroke, further studies are needed to evaluate the efficacy and safety.⁸² Our cautious conclusion is that minocycline may fail to effectively modulate highly dynamic, complex and sex-specific stroke-induced microglial responses.

Indeed, to discover more effective microglia-targeted therapies for stroke, we need to gain a deeper understanding of microglia biology. An elegant work by Villa et al demonstrated a significant difference in the transcriptome profile between adult male and female microglia independent of hormonal cues. Female phenotype is retained after transfer of microglia into male brain and more importantly, female microglia protected the male brain from ischemic stroke.²⁴ The higher expression of genes involved in cell plasticity, control of inflammation and brain repair in female mice could contribute to protection. These genes may provide novel targets for sex-related approaches for effective stroke therapy.

Going beyond the acute phases of stroke is also important. Peri-infarct cortex is highly plastic undergoing reorganization through the reactivation of neurodevelopmental mechanisms to compensate lost neuronal functions.⁸³ Increase in dendritic spine production together with axonal sprouting is an integral part of remodeling and brain repair process. Here, stroke-induced microglia have a major role in elimination of spines and shaping neural networks in functionally relevant manner. In depth understanding of the sex differences in microglia modulation of the peri-infarct cortex during stroke repair process, similar to neurodevelopment, may offer novel therapeutic targets for long-term functional brain recovery.

Different molecular mechanisms evolve after stroke in time-dependent manner. These are potential targets for microglia-based therapies to reduce degeneration and alleviate neuroinflammation or later to enhance brain repair, when properly timed. High number experimental drugs and interventions are effective in stroke models, yet, they have not translated into clinic success.⁸⁴ It is expected that emerging understanding of sexually dimorphic brain and stroke pathology improves translation of basic science results to sex-specific therapies. For translational success, both sexes and aged animals should be considered and used whenever possible as suggested by STAIR committee,⁸⁵ despite budgetary constraints and complex study design. Microglial chimeric mice enable comparison of sex differences in human context.⁸⁶

Sex has been shown to contribute to the progression of poststroke inflammation and outcome. Specific molecular programs are launched to drive the process of brain differentiation between sexes already in very early stages of development and specifically in microglia. The characteristic differences acquired during development will influence the inflammatory response in adulthood and particularly in case of severe brain insults such as stroke. These inherent differences need to be investigated and understood to ensure that the novel therapeutic paradigms translate into new treatments.

Recent advances in molecular biology and single-cell analysis have provided considerable insights into dynamic microglial responses in stroke. In-depth understanding of sex-dependent microglial responses in aging animals and both acute and long-term consequences aid in driving and accelerating development of sex- and microglia-specific, personalized treatment strategies for acute protection and brain repair in stroke.