Sex-Specific Effects of Microglia-Like Cell Engraftment during Experimental Autoimmune Encephalomyelitis

Multiple sclerosis (MS) is a chronic neuroinflammatory disorder of the central nervous system (CNS) that usually presents in young adults and predominantly in females. Microglia, a major resident immune cell in the CNS, are critical players in both CNS homeostasis and disease. We have previously demonstrated that microglia can be efficiently depleted by the administration of tamoxifen in Cx3cr1CreER/+Rosa26DTA/+ mice, with ensuing repopulation deriving from both the proliferation of residual CNS resident microglia and the engraftment of peripheral monocyte-derived microglia-like cells. In this study, tamoxifen was administered to Cx3cr1CreER/+Rosa26DTA/+ and Cx3cr1CreER/+ female and male mice. Experimental autoimmune encephalomyelitis (EAE), a widely used animal model of MS, was induced by active immunization with myelin oligodendrocyte glycoprotein (MOG) one month after tamoxifen injections in Cx3cr1CreER/+Rosa26DTA/+ mice and Cx3cr1CreER/+ mice, a time point when the CNS niche was colonized by microglia derived from both CNS microglia and peripherally-derived macrophages. We demonstrate that engraftment of microglia-like cells following microglial depletion exacerbated EAE in Cx3cr1CreER/+Rosa26DTA/+ female mice as assessed by clinical symptoms and the expression of CNS inflammatory factors, but these findings were not evident in male mice. Higher major histocompatibility complex class II expression and cytokine production in the female CNS contributed to the sex-dependent EAE severity in mice following engraftment of microglia-like cells. An underestimated yet marked sex-dependent microglial activation pattern may exist in the injured CNS during EAE.


Introduction
Multiple sclerosis (MS) is a complex neuroinflammatory disorder of the central nervous system (CNS) associated with progressive and irreversible neurological dysfunctions [1]. As specialized

Long-Term Engraftment of Microglia-Like Cells in Cx3cr1 CreER/+ Rosa26 DTA/+ Mice
In order to investigate the kinetic changes of microglial depletion and repopulation Cx3cr1 CreER/+ Rosa26 DTA/+ and Cx3cr1 CreER/+ mice were terminated at different time points (day 1, 3, 7 and 1 month) after three consecutive subcutaneous tamoxifen injections (experimental design depicted in Figure 1A), both male and female mice being used. Flow cytometric and immunohistochemical analyses of brain and spinal cord tissues were performed at each time point. Approximately 92% of CD11b + CD45 low Ly6C − Ly6G − microglia in the brain (gating strategy presented in Supplementary Figure S1A) could be depleted effectively 7 days after tamoxifen injections. The repopulation of microglia-like cells with two subpopulations, deriving from both the proliferation of residual CNS resident microglia and the engraftment of peripheral monocyte-derived microglia-like cells [21], was evident in the brain one month later ( Figure 1B,D). Consistent with the findings in the brain, spinal cord CD11b + CD45 low Ly6C − Ly6G − microglia (gating strategy presented in Supplementary Figure S1C) in Cx3cr1 CreER/+ Rosa26 DTA/+ mice could also be depleted effectively 7 days after tamoxifen injections, and repopulating microglia-like cells were noted one month later ( Figure 1C,E). No depletion of microglia was noted in either the brains or spinal cords of Cx3cr1 CreER/+ mice, indicating their suitability as a control group ( Figure 1B,C). We further confirmed these findings by double immunofluorescent staining with Iba1 and Tmem119 antibodies in both brain and spinal cord tissues at different time points (Figure 2A,B). In summary, the engraftment of microglia-like cells occurs in both the brain and spinal cord of Cx3cr1 CreER/+ Rosa26 DTA/+ mice one month after administration of tamoxifen.
Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 3 of 16 resident microglia and the engraftment of peripheral monocyte-derived microglia-like cells [21], was evident in the brain one month later ( Figure 1B,D). Consistent with the findings in the brain, spinal cord CD11b + CD45 low Ly6C − Ly6G − microglia (gating strategy presented in Supplementary Figure S1C) in Cx3cr1 CreER/+ Rosa26 DTA/+ mice could also be depleted effectively 7 days after tamoxifen injections, and repopulating microglia-like cells were noted one month later ( Figure 1C,E). No depletion of microglia was noted in either the brains or spinal cords of Cx3cr1 CreER/+ mice, indicating their suitability as a control group ( Figure 1B,C). We further confirmed these findings by double immunofluorescent staining with Iba1 and Tmem119 antibodies in both brain and spinal cord tissues at different time points (Figure 2A,B). In summary, the engraftment of microglia-like cells occurs in both the brain and spinal cord of Cx3cr1 CreER/+ Rosa26 DTA/+ mice one month after administration of tamoxifen.    EAE was induced by active immunization with MOG one month after tamoxifen injections in Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice, a time point when the CNS niche was colonized by microglia derived from both CNS microglia and peripherally-derived macrophages. In female Cx3cr1 CreER/+ Rosa26 DTA/+ mice the disease onset was normal but a higher chronic disease severity was recorded than in male mice ( Figure 3A,B, * p < 0.05). In contrast, similar clinical scores during the whole EAE observation period were recorded in male and female Cx3cr1 CreER/+ control mice ( Figure 3A). Furthermore, days 17-20 correspond to the peak of disease and female Cx3cr1 CreER/+ Rosa26 DTA/+ mice had higher peak scores than did male mice ( Figure 3B, * p < 0.05), while cumulative scores and day of disease onset were similar between groups. Overall, female Cx3cr1 CreER/+ Rosa26 DTA/+ mice developed a more severe EAE course following engraftment of microglia-like cells. In order to further explore if the engraftment of microglia-like cells contributed to disease severity during the EAE recovery period, tamoxifen was administered at days 8-10 post-immunization, allowing microglia-like cells to gradually repopulate after the clinical peak of EAE ( Figure 3C). Again, female Cx3cr1 CreER/+ Rosa26 DTA/+ mice also experienced a higher severity of the disease than did male mice, while this phenomenon was not evident in Cx3cr1 CreER/+ control mice of either gender ( Figure 3C). To exclude that this sex-specific effect of the engraftment of microglia-like cells was not due to a baseline sex difference in EAE development, we also compared the disease courses of male and female mice from the different strains. Our results demonstrate that both males and females, irrespective of their strain background, developed similar EAE courses and cumulative scores ( Figure 3D). Taken together, we conclude that engraftment of peripherally-derived microglia-like cells in females exacerbates EAE disease.

CNS Myeloid Cell Compositions during Chronic EAE
We further confirmed that microglia depletion followed similar dynamics between male and female Cx3cr1 CreER/+ Rosa26 DTA/+ mice (Supplementary Figure S2A) and the sex differences in EAE severity were not dependent on the overall number of repopulating microglia in Cx3cr1 CreER/+ Rosa26 DTA/+ mice (Supplementary Figure S2B). In order to analyze sex differences of CNS myeloid cell compositions during the chronic EAE stage, brain and spinal cord tissues were dissected 1 month post-immunization (schematic overview is depicted in Figure 4A). Myeloid cell compositions in the brain and spinal cord tissues of Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice were analyzed using flow cytometry. The gating strategies of brain and spinal cord myeloid cells during EAE are depicted in Supplementary Figure S1B,D, respectively. We observed that infiltrating CD11b + CD45 hi Ly6C − Ly6G − macrophages, Ly6C + monocytes and Ly6G + neutrophils were still present in the CNS tissues during the chronic EAE stage ( Figure 4B,C,F). Newly repopulated CD11b + CD45 low Ly6C − Ly6G − microglia in the brains were lower in number than in control groups ( Figure 4D). A similar trend was also recorded in the spinal cord ( Figure 4E). Infiltrating CD11b + CD45 hi Ly6C − Ly6G − macrophages in the CNS tissues did not differ among different groups ( Figure 4D,E). The percentage of infiltrating Ly6C + monocytes and Ly6G + neutrophils in chronic EAE brains and spinal cords did not differ between Cx3cr1 CreER/+ Rosa26 DTA/+ and Cx3cr1 CreER/+ mice or between sexes ( Figure 4F,G). exclude that this sex-specific effect of the engraftment of microglia-like cells was not due to a baseline sex difference in EAE development, we also compared the disease courses of male and female mice from the different strains. Our results demonstrate that both males and females, irrespective of their strain background, developed similar EAE courses and cumulative scores ( Figure 3D). Taken together, we conclude that engraftment of peripherally-derived microglia-like cells in females exacerbates EAE disease. Figure 3. Female Cx3cr1 CreER/+ Rosa26 DTA/+ mice develop more severe experimental autoimmune encephalomyelitis (EAE) after engraftment of microglia-like cells (A) Tamoxifen was first administered and myelin oligodendrocyte glycoprotein (MOG)-EAE was subsequently induced when microglia were repopulated (one month later). Clinical scores of neurological deficits post-immunization up to 29 days are indicated. Female Cx3cr1 CreER/+ Rosa26 DTA/+ mice (n = 11) with newly repopulated microglia (dark pink color) had a similar disease onset but experienced a higher disease severity during the chronic stage than did male Cx3cr1 CreER/+ Rosa26 DTA/+ mice (n = 13) with newly repopulated microglia (dark blue color, * p < 0.05). Female Cx3cr1 CreER/+ mice (n = 12) with resident microglia (light pink color) experienced similar clinical scores during the whole EAE period when compared with male Cx3cr1 CreER/+ mice (n = 13) with resident microglia (light blue color). Data are representative of two independent experiments. (B) Peak disease score, day of disease onset and cumulative scores of both male and female groups are depicted during EAE. (C) EAE was first induced and tamoxifen was subsequently administered on days 8, 9 and 10 post-immunization. Clinical scores of neurological deficits post-immunization up to 25 days are presented (n = 6, 8, 9, 8, respectively). (D) Different mouse strains (Cx3cr1 CreER/+ Rosa26 DTA/+ male and female mice, Cx3cr1 CreER/+ male and female mice and C57BL/6 female wild-type mice, n = 3, 9, 5, 5, 7, respectively) without tamoxifen injections developed similar clinical scores and cumulative EAE scores. Statistical significance is indicated as * p < 0.05. One month after tamoxifen injections, EAE was induced by active immunization with MOG in Cx3cr1 CreER/+ Rosa26 DTA/+ mice with newly repopulated microglia and in Cx3cr1 CreER/+ mice with resident microglia (both male and female mice). Brain and spinal cord tissues were dissected on day 29 post-immunization (chronic EAE period). (B,C) Myeloid cell subset analysis in the brain (B) and spinal cord tissues (C) of the Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice were performed using flow cytometry. (D) The numbers of CD11b + CD45 low Ly6C − Ly6G − brain microglia and CD11b + CD45 hi Ly6C − Ly6G − macrophages in repopulated and control groups. n = 5 mice in each group. (E) The numbers of CD11b + CD45 low Ly6C − Ly6G − spinal cord microglia and CD11b + CD45 hi Ly6C − Ly6G − macrophages in repopulated and control groups. n = 3, 5, 4 and 4 mice in female repopulation, female control, male repopulation and male control group, respectively. (F) Representative flow cytometry plots of infiltrating Ly6C + monocytes and Ly6G + neutrophils in EAE brain tissues are depicted during the chronic EAE period. (G) Infiltrating Ly6C + monocyte and Ly6G + neutrophil proportions in EAE brains did not differ among different groups during chronic EAE period. Statistical significance is indicated as * p < 0.05, ** p < 0.01 and *** p < 0.001. One month after tamoxifen injections, EAE was induced by active immunization with MOG in Cx3cr1 CreER/+ Rosa26 DTA/+ mice with newly repopulated microglia and in Cx3cr1 CreER/+ mice with resident microglia (both male and female mice). Brain and spinal cord tissues were dissected on day 29 post-immunization (chronic EAE period). (B,C) Myeloid cell subset analysis in the brain (B) and spinal cord tissues (C) of the Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice were performed using flow cytometry. (D) The numbers of CD11b + CD45 low Ly6C − Ly6G − brain microglia and CD11b + CD45 hi Ly6C − Ly6G − macrophages in repopulated and control groups. n = 5 mice in each group. (E) The numbers of CD11b + CD45 low Ly6C − Ly6G − spinal cord microglia and CD11b + CD45 hi Ly6C − Ly6G − macrophages in repopulated and control groups. n = 3, 5, 4 and 4 mice in female repopulation, female control, male repopulation and male control group, respectively. (F) Representative flow cytometry plots of infiltrating Ly6C + monocytes and Ly6G + neutrophils in EAE brain tissues are depicted during the chronic EAE period. (G) Infiltrating Ly6C + monocyte and Ly6G + neutrophil proportions in EAE brains did not differ among different groups during chronic EAE period. Statistical significance is indicated as * p < 0.05, ** p < 0.01 and *** p < 0.001.

Higher MHC II Expression of Infiltrating Ly6C hi Monocytes during Peak EAE in Female Microglia-Repopulated Mice
We next explored CNS myeloid cells during the acute EAE stage. Brain and spinal cord tissues were thus dissected on day 18 post-immunization (schematic overview is depicted in Figure 5A) and sex differences were also noted in this setting (Supplementary Figure S3). Consistent with previous EAE results, there was a trend that following engraftment of microglia-like cells female Cx3cr1 CreER/+ Rosa26 DTA/+ mice developed worse clinical symptoms than in Cx3cr1 CreER/+ mice with resident microglia (Supplementary Figure S3A). Cx3cr1 CreER/+ Rosa26 DTA/+ female mice had an earlier disease onset than that of female control mice, and both peak score and cumulative scores were significantly higher than in female control mice (Supplementary Figure S3B), while the corresponding male groups were comparable (Supplementary Figure S3B). Furthermore, CD11b + CD45 low Ly6C − Ly6G − microglia in both the brain and spinal cord were less numerous in microglia-like cell engrafted groups than in control groups ( Figure 5B,C). Infiltrating CD11b + CD45 hi Ly6C − Ly6G − macrophages in the brain EAE tissues did not differ among groups ( Figure 5B,C), while infiltrating Ly6G + neutrophils in EAE brains was significantly increased in the female repopulated microglia group than in the female control group ( Figure 5B). Proportions of infiltrating Ly6C + monocytes and Ly6G + neutrophils in EAE spinal cord tissues did not differ among groups during the acute EAE period ( Figure 5C). Importantly, MHC II expression of infiltrating Ly6C hi monocytes was significantly increased in female repopulated microglia mice than in the male group during the acute EAE period, but not during the chronic EAE period ( Figure 5D,E).

Elevated Cytokine Production during EAE Peak in Female Microglia-Repopulated Mice
In concert with the higher clinical score, we observed that the expression of Interferon-γ (IFN-γ) in CD4 + T cells (CD4 + IFN-γ + T cell subgroup) was significantly greater in female Cx3cr1 CreER/+ Rosa26 DTA/+ brains than in both the female control group and male Cx3cr1 CreER/+ Rosa26 DTA/+ brains during the acute EAE stage ( Figure 6A,B). Furthermore, the expression of IL-17 in CD4 + T cells (CD4 + IL-17 + T cell subgroup) was also significantly higher in female Cx3cr1 CreER/+ Rosa26 DTA/+ brains when compared to female control mice. (Figure 6C,D). Furthermore, no sex differences in proportions and proliferation of infiltrating CD4 + and CD8 + T cells in the brain (18 days post-immunization of EAE) were noted in Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice (Supplementary Figure S4). These data suggest that higher cytokine production in the female brains with microglia-like cells contributed to the sex-dependent exacerbation of EAE.
( Figure 5B,C), while infiltrating Ly6G + neutrophils in EAE brains was significantly increased in the female repopulated microglia group than in the female control group ( Figure 5B). Proportions of infiltrating Ly6C + monocytes and Ly6G + neutrophils in EAE spinal cord tissues did not differ among groups during the acute EAE period ( Figure 5C). Importantly, MHC II expression of infiltrating Ly6C hi monocytes was significantly increased in female repopulated microglia mice than in the male group during the acute EAE period, but not during the chronic EAE period ( Figure 5D,E).

Figure 5.
Higher MHC II expression of infiltrating Ly6C hi monocytes during peak EAE in female microglia-repopulated mice (A) Schematic overview of the experimental design. One month after tamoxifen injections, EAE was induced by active immunization with MOG in Cx3cr1 CreER/+ Rosa26 DTA/+ mice with newly repopulated microglia and Cx3cr1 CreER/+ mice with resident microglia (both male and female mice). Brain and spinal cord tissues were dissected on day 18 post-immunization (acute EAE period). Myeloid cell subset analysis in the brain (B) and spinal cord tissues (C) of Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice were performed using flow cytometry during the acute EAE disease course. MHC II expression (%) of infiltrating Ly6C hi monocytes was higher in the female repopulated microglia group than in the male repopulated group during the acute EAE period (D), while significant difference was not evident during the chronic EAE period (E). Acute EAE experiment: n = 6 mice/group in female repopulation, female control and male repopulation groups, and n = 7 mice in male control group. Chronic EAE experiment: n = 5 mice in each group. Statistical significance is indicated as * p < 0.05 and ** p < 0.01. when compared to female control mice. (Figure 6C,D). Furthermore, no sex differences in proportions and proliferation of infiltrating CD4 + and CD8 + T cells in the brain (18 days post-immunization of EAE) were noted in Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice (Supplementary Figure S4). These data suggest that higher cytokine production in the female brains with microglia-like cells contributed to the sex-dependent exacerbation of EAE. Representative numbers indicate the percentage of each compartment. n = 6, 5, 6 and 7 mice in female repopulation, female control, male repopulation and male control groups, respectively. Statistical significance is indicated as * p < 0.05 and *** p < 0.001.

Discussion
In this study we utilized the Cx3cr1 CreER/+ Rosa26 DTA/+ mouse model which has the advantage of no need of invasive administration of diphtheria toxin, as is the case in the Cx3cr1 CreER :iDTR animal model [19]. Our previous studies using this model demonstrated that newly repopulating microglia are a combination of repopulation from both self-renewing microglia and of infiltrating monocytes that undergo transcriptional modification to become microglia-like cells [22,24]. In the present study we addressed the functionality of these engrafted microglia-like cells in the setting of neuroinflammatory EAE disease, noting their reduced ability to limit pathology compared to self-renewing microglia. The functional loss of microglial homeostasis or microglial trophic factors is thus not beneficial for EAE disease recovery. The significantly increased injury was sex-dependent, with females experiencing worse clinical symptoms. This suggests that there are sex-specific differences in microglia-like cells. Recent data obtained from additional preclinical models also indicated that the number and phenotype of microglia may differ between females and males [25,26]. Microglial functional differences may predispose to previously underestimated but marked sex-dependent microglial activation patterns and signaling cascades in settings of CNS damage [27]. However, there is an argument that MOG-EAE is considered less valid as an animal model for MS since MOG-antibody-associated disease is now considered as a new antigen-specific neuroimmunological disorder [28]. It should be noted that the complex pathological MS processes depend on a combination of the genetic diversity of individuals, a wide range of environmental factors and different immune cell types in humans, while its animal model EAE may not cover the entire spectrum of these characteristics. As the implementation of microglial depletion and repopulation is currently not initiated in patients with MS or MOG-antibody-associated disease, the EAE model provides an opportunity to address the concept. Our study extends the understanding of the distinct functions of engrafted male and female microglia-like cells following depletion, and might provide valuable insights for the decision-making for female vs. male MS patients when microglial-depletion therapies eventually become available.
The timing of microglial ontogeny and development places them in a special position compared with other tissue resident macrophages. Microglia can be self-maintained by local proliferation and apoptosis with little contribution from peripherally circulating monocytes in physiological conditions. We and others have previously demonstrated that peripherally-derived macrophages can give rise to engrafted long-lived microglia-like cells in the brain parenchyma using experimental models, with this new phenotype retaining distinct gene signatures and functional differences such as phagocytic capacity and responses to in vitro challenges [20,22,29]. The functional differences between these engrafted microglia-like cells and yolk sac-derived resident microglia in vivo thus have clinical implications for the treatment of diverse neurological diseases, as we have previously reviewed [30]. It is also important to note that CX3CR1 + resident macrophages such as perivascular, subdural meningeal and choroid plexus macrophages within the CNS might also be affected in Cx3cr1 CreER/+ Rosa26 DTA/+ mice after treating with tamoxifen [31,32]. These cell populations play a complex role in the pathogenesis of EAE, but it is difficult to distinguish them from infiltrating macrophages using surface markers. Nevertheless, by using our gating strategy we referred to the CD11b high CD45 high population as 'infiltrating macrophages' since in the context of EAE, the contribution of infiltrating macrophages is more dominant compared to that of the resident CNS macrophage pool; CNS-resident macrophages expand modestly compared to the considerable infiltration of peripheral monocytes/macrophages. Although the CNS tissues were perfused thoroughly with PBS and dissected carefully from the meninges for flow cytometrical analysis to minimize the potential contribution from CNS-resident macrophages, the results still need to be interpreted with caution. Technically, applying advanced methods including single-cell RNA-sequencing, high-dimensional cytometry and fate-mapping or using transgenic tools such as Cx3cr1 CreERT2 R26td Tomato mice to label resident macrophages may provide valuable information regarding the relative contribution of these border-associated macrophages in the context of EAE between male and female.
Recent data indicated that microglial repopulation following experimental depletion using pharmacological colony-stimulating factor 1 receptor (CSF1R) inhibition could effectively resolve inflammation and promote disease recovery after brain injury, without any apparent adverse effects [33]. Using the same approach, age-related spatial memory impairment, microglial cell densities and morphologies could all be reversed by replacing primed microglia in aged mice with newly repopulated microglia [34], and they are also crucial for CNS regeneration following pro-inflammatory microglial necroptosis in a focal demyelinated animal model [35]. Furthermore, bone marrow-derived microglia-like cells may be more efficient in clearing amyloid beta deposits than their CNS endogenous counterparts [36]. CSF1R inhibition-mediated repopulation largely promotes repopulation by self-renewing microglia, while engraftment of microglia-like cells occurs in our Cx3cr1 CreER/+ Rosa26 DTA/+ mice. However, we did not observe a neuroprotective effect followed by microglial depletion and repopulation in our case. Different experimental depletion methods might explain some of the discrepancies. Long-term systematic administration of CSF1R inhibitors may affect other immune cells expressing CSF1R in the periphery, then generally suppressing immune responses [18]. In Cx3cr1 CreER/+ Rosa26 DTA/+ mice, tamoxifen injections may thus modulate systemic immune responses during EAE course ( Figure 3C). Furthermore, myeloid cells may also play regulatory roles in EAE and in our model, the overall number of microglia was fewer than the control group one month after experimental depletion. The relative importance of microglial subpopulations such as resident microglia, disease-associated microglia and peripheral-engrafted microglia-like cells during microglial depletion and repopulation periods at different disease stages needs to be explored and validated using different depletion methods in the future.
Tamoxifen, a non-steroidal estrogen receptor modulator, is used as a hormone therapy drug for treating breast cancer. Previous studies have reported that low doses of tamoxifen may promote microglial polarization toward an anti-inflammatory phenotype in male mice following chronic hypoperfusion [37]. Some may argue that tamoxifen itself may exert confounding effects on EAE independent of microglia depletion and repopulation. We induced MOG-EAE almost one month after injecting tamoxifen, and this should minimize any potential impact of tamoxifen on disease development. In our study we also included male and female EAE mice that were not treated with tamoxifen, and these developed comparable clinical scores during EAE, suggesting no sex difference in wild-type mice. Taken together, our data support the notion that the replacement of microglia by engraftment of microglia-like cells has a sex-specific consequence on development of autoimmune neuroinflammation.
Our results revealed that no sex differences in proportions and proliferation of infiltrating CD4 + and CD8 + T cells in the brain (18 days post-immunization of EAE) were evident between Cx3cr1 CreER/+ Rosa26 DTA/+ mice and Cx3cr1 CreER/+ mice. It is now well accepted that activated microglia and infiltrating T cells interact during disease conditions, and our results suggested that the sex differences in EAE severity might not depend on the changes of CD4 + T cells or CD8 + T cells after the engraftment of microglia-like cells. Antigen can be presented by microglia via MHC II. Resident microglia and peripheral-derived microglia-like cells may perform specialized distinct functions in vivo. In our study the gender-dependent EAE severity in Cx3cr1 CreER/+ Rosa26 DTA/+ mice with engraftment of microglia-like cells may be partially due to their higher MHC II expression and cytokine production in the female CNS. Correlation with enhanced CD4 + IFN-γ + and CD4 + IL-17 + functionality in female mice supports this increased disease-promoting immune activity.
Our study has several limitations, such as low yield of cells from spinal cords following enzymatic digestion for intracellular/cytokines staining. It also hampers the future investigation of the potential differences in cellular phenotypes and biological functions of repopulated microglia between brain and spinal cord, as suggested by other researchers [38]. Further investigating inflamed spinal cords in EAE may provide additional cues of the sex-specific effects of peripheral microglia-like cell engraftment during neuroinflammation. Although we still noted the sex-specific effects during acute EAE, a larger sample size will be needed to validate our findings in the future. With the engrafted microglia-like cells inflammatory mediators may be changed differentially between males and females during the acute period, then contributing to an obvious sex-specific effect during the disease recovery period.
There is a current notion that peripherally-derived macrophages could be used to replace defective microglia and thus may have therapeutic potential for a range of CNS pathologies [34,36]. Our study indicates that careful consideration of gender effects and of desired in vivo functionality should be taken in design of future microglial replacement therapies.

Ethics Statement
All experiments in this study were approved and performed in accordance with the guidelines from the Swedish National Board for Laboratory Animals and the European Community Council Directive (86/609/EEC) and the local ethics committee of Stockholm North under the ethical permits 9328-2019. All efforts were made to minimize animal suffering and discomfort.

Animals
Cx3cr1 CreER (Jax Stock: 021160) and Rosa26 DTA (Jax Stock: 010527) mice were purchased from the Jackson Laboratory and bred to obtain both Cx3cr1 CreER/+ Rosa26 DTA/+ and Cx3cr1 CreER/+ mice that were used for experiments. C57BL/6NTac mice (Taconic) were bred in the Comparative Medicine Department at Karolinska University Hospital, Sweden. All experimental mice were maintained under a regulated light/dark schedule and temperature conditions of this specific pathogen-free animal facility. All experimental mice were aged between 5 and 13 weeks-old and had free access to standard rodent chow and water.

EAE Induction and Clinical Evaluation
EAE was induced based on the standard protocol in our lab [39]. Briefly, MOG (amino acids 1-125 from the N terminus) was expressed in Escherichia coli and purified to homogeneity by chelate chromatography. Purified MOG dissolved in 6M urea was then dialyzed against sodium acetate buffer (10 mM, pH 3.0) to obtain a soluble preparation. Mice were anesthetized with isoflurane (Forane; Abbott Laboratories, Abbot Park, IL, USA) and injected subcutaneously in the dorsal tail base with 35 µg of MOG in phosphate-buffered saline emulsified in Complete Freund's Adjuvant (CFA, Chondrex, Inc, 7027, Redmond, WA, USA) containing 100 µg heat-killed Mycobacterium tuberculosis H37Ra (Difco Laboratory, Detroit, MI, USA). On the day of immunization and 48h later, mice were injected intraperitoneally with 200 ng pertussis toxin (Sigma, P7208-50UG, St Louis, MO, USA). Body weight and paralytic symptoms were assessed daily from 8 days post-immunization. The clinical signs of EAE were scored according to the following criteria: 0, no clinical signs of EAE; 1, tail weakness or tail paralysis; 2, hindlimb paraparesis or hemiparesis; 3, hindlimb paralysis or hemiparalysis; 4, tetraplegia or moribund; and 5, death (0.5 being assigned for intermediate clinical signs). The cumulative score was calculated as the sum of the daily scores of individual mouse from day 0 until the day of termination. The clinical symptoms were scored in a blinded manner by 3 independent researchers.

Preparation of Single Cell Suspensions from CNS Tissues
Mice were deeply anaesthetized by injecting 100 µg pentobarbital intraperitoneally. Mice were perfused through the left cardiac ventricle using ice-cold PBS. Whole brain and spinal cord were removed and minced using a surgical disposable scalpel (MediCarrier AB, Stockholm, Sweden), followed by enzymatic digestion using Collagenase (11088866001, Roche, Sweden) and DNAse (000000010104159001, Roche, Sweden). Myelin was removed using a 38% Percoll gradient (P1644-1L, Sigma, Sweden).

Flow Cytometry
Single cell suspensions were plated into 96-well V-bottom plates and stained at 4 • C for 20 min. Dead cells were removed using Live/Dead Fixable Near-IR Dead Cell Stain Kit (Invitrogen, Thermo Fisher Scientific, Stockholm, Sweden) in each panel. The following antibody panels were used: 1.