Monocytic myeloid-derived suppressor cells from females, but not males, alleviate CVB3-induced myocarditis by increasing regulatory and CD4+IL-10+ T cells

Coxsackievirus group B type 3 (CVB3) is a common etiologic agent of viral myocarditis and often causes sexually dimorphic myocarditis with increased incidence and mortality in male. So far, the underlying mechanism for the high male prevalence is not well elucidated. In this study, we deciphered the role of myeloid-derived suppressor cells (MDSCs) in the gender bias in murine CVB3-induced myocarditis by comparing their frequencies, subsets as well as immune suppressive functions. We found that much more myocardial MDSCs were enriched in infected females than males, with dramatically higher percentage ratio of CD11b+Ly6G-Ly6Chigh monocytic subset (M-MDSCs) to CD11b+Ly6G+Ly6Clow granulocytic subset (G-MDSCs). Interestingly, more potent suppression on T cell proliferation was also evidenced in female-derived M-MDSCs. Consistently, adoptive transfer of female- but not male-derived M-MDSCs efficiently alleviated CVB3-induced myocarditis in male recipient mice, and this protection could be ascribed to the increased induction of regulatory and CD4+IL-10+ T cells. Our study suggested that myocardial MDSCs were distinctively induced not only in quantities but also in phenotypes and immune suppressive functions in CVB3-infected males and females; and female-derived more suppressive M-MDSCs contributed to their insensitivity to CVB3-induced myocarditis.

Viral myocarditis is a major cause of sudden death in infants and young adults under 40 years, and can further develop into dilated cardiomyopathy (DCM) and congestive cardiac failure [1][2][3] . Enteroviruses, especially coxsackieviruses, have been considered as the most common cause of viral myocarditis [4][5][6] . It is well established that sex and sex-associated hormones contribute to the susceptibility of CVB3-induced myocarditis [7][8][9] . Although similar viral infection efficiency has been measured in both genders, the estimated incidence of myocarditis in men is 2-fold or more than in women 10 . In murine models of CVB3-induced myocarditis, significantly higher occurrence and more severe myocardial inflammation are also noted in males 11 .
Increasing evidence has shown that indirect immune-mediated injury, but not direct virus-induced damage, is the prominent pathological mechanism of CVB3-induced myocarditis 12,13 . Two genders display diverse immune patterns, strength as well as immune modulations in the disease process of viral myocarditis 14,15 . It has been showed that in CVB3-infected male mice, significantly increased Toll-like receptor 4 (TLR4) led to the augmented IL-1β /IL-18 expression and reduced regulatory T (Treg) cells production; while in female, up-regulated T cell immunoglobulin mucin-3 (Tim-3) resulted in the increased expression of regulatory molecule CTLA-4 on CD4 + T cells and higher frequency of Treg cells 11 . Our previous study showed that myocardial macrophage polarization also contributed to the gender bias of viral myocarditis in mice. Type I phenotype of macrophage dominated in infected males and aggravated myocardial inflammation. While type II phenotype of macrophage dominated in infected females and preferentially alleviated myocarditis by inducing Treg 16 . These literatures suggest that various immune modulation mechanisms contribute a lot to the gender bias of CVB3-induced myocarditis.
Scientific RepoRts | 6:22658 | DOI: 10.1038/srep22658 Myeloid-derived suppressor cells (MDSCs) have been described as a heterogeneous cell population consisting of myeloid progenitor cells and immature cells with potent immune suppressive function 17,18 . Besides of their particular roles in tumor immune escape 3,19 , the functions of MDSCs in infection and inflammation have also attracted more and more attention [20][21][22][23] . It has been proven that viruses could efficiently recruit a large amount of tissue MDSCs to inhibit host defense and benefit their infection [24][25][26] . In verse, induced MDSCs could modulate the pattern and strength of innate and adaptive immune responses to avoid massive tissue immunopathology and inflammation 27,28 . Interestingly, a previous study revealed that in Trypanosoma cruzi-induced myocarditis, abundant MDSCs accumulated in heart tissues and facilitated infection by suppressing T cell proliferation 29 , indicating the participation of MDSCs in infectious heart diseases. Therefore, MDSCs may exert versatile functions in infectious disease courses, and elucidating the role of MDSCs in the CVB3-induced myocarditis may help us gain a better understanding of the mechanisms underlying the gender bias of viral myocarditis.
In this study, the frequency and phenotype of MDSCs were characterized in male and female mice with CVB3-induced myocarditis, and their potential roles in the development of viral myocarditis were deciphered.

Results
Females revealed less susceptibility to CVB3-induced myocarditis than males. Age-matched male and female BALB/c mice were intraperitoneally infected with CVB3. Seven days later, the incidence and severity of myocarditis were measured and compared in both genders. As illustrated in Fig. 1, females exhibited slighter body weight loss, lower serum CK activity as well as limited myocardial inflammatory infiltration (indicated by the arrows, Fig. 1C) compared with males, indicating less severe viral myocarditis. In consistence with limited myocardial inflammation and injury, the 7-day survival rate of females achieved 80%, significantly higher than 30% of males (p = 0.01, Fig. 1D). These results confirmed our previous observation that female mice showed less susceptibility to CVB3-induced myocarditis 16 . In addition, this gender bias was not attributed to the less efficient CVB3 replication in females, since similar myocardial viral loads were detected on day 7 post infection in both genders (Fig. 1E).
More myocardial MDSCs were enriched in females than males with CVB3-induced viral myocarditis. Mounting evidence suggests that different patterns and potencies of immune responses between males and females cause the dimorphic sensitivity to viral myocarditis 15,30 . Therefore, we speculated that the frequency and/or function of immune regulatory cells may be distinct in both genders. MDSCs represent myeloid originated immunoregulatory cells with immune suppressive functions on both innate and adaptive immunity stages 17,31 . To evaluate the influence of MDSCs on sex-based sensitivity to viral myocarditis, their frequencies in CVB3-infected male and female mice were determined. As shown in Fig. 2, no distinct sex difference in MDSCs percentages was seen in naive mice. Following CVB3 infection, the frequencies of MDSCs in the spleen and blood significantly increased in males and were much higher than females. However, the situation was completely opposite in the heart tissues, the frequency of MDSCs reached 5.4% in females, robustly higher than 1.8% in males ( Fig. 2A). Besides, depletion of MDSCs with anti-Gr-1 antibody could obviously alleviate CVB3-induced myocarditis in males, but have marginal impact in females (Fig. 2B).
Monocytic MDSCs (M-MDSCs) predominantly distributed in the myocardial tissue of infected females. MDSCs consist of two major populations: CD11b + Ly6G + Ly6C low granulocytic (G-MDSCs) and CD11b + Ly6G -Ly6C high monocytic (M-MDSCs) cells, we next detected these two subsets in the heart tissues of both genders. As shown in Fig. 3A and B, 18.2% M-MDSCs and 55.2% G-MDSCs were observed in males, with M-MDSCs/ G-MDSCs ratio to be 0.3. While in females, M-MDSCs frequency dramatically increased to 31.2% and its ratio to G-MDSCs augmented to 1.3, indicating M-MDSCs as the predominantly expanded subset in the hearts of infected females. Consistently, the absolute number of myocardial M-MDSCs in females was considerably larger than that in males (Fig. 3D). The situation of myocardial G-MDSCs was totally opposite, the number of G-MDSCs was obviously higher in males (Fig. 3E). To further investigate whether estrogen play a role in MDSCs infiltration and phenotypes, CVB3-infected male mice were treated with 17β -estradial and then total and two subsets of myocardial MDSCs were detected. As shown in Fig. 3F, estrogen could obviously promote the enrichment of CD11b + Gr-1 + MDSCs post CVB3 infection, with the priority to increase the percentage of CD11b + Ly6G -Ly6C high monocytic (M-MDSCs). Our results were further supported by the previous study which reported that 17β -estradiol promote the accumulation of myeloid-derived suppressor cells 30 .
Next, immune suppressive functions of myocardial MDSCs subsets were compared in both genders. As shown in Fig. 4, T cell proliferation provoked by anti-CD3 and anti-CD28 stimulation was not obviously influenced by both male-and female-derived G-MDSCs, but dramatically inhibited by female-, but not male-derived M-MDSCs. These data indicated that myocardial M-MDSCs subset possessed the immune suppressive function in the context of CVB3-induced myocarditis; meanwhile, female-derived myocardial M-MDSCs dominated not only in quantity but also in immune suppressive function compared with their counterpart in males.

Adoptive transfer of female-derived M-MDSCs significantly alleviated CVB3-induced myocarditis in males.
To further assess the immune suppressive ability of female-derived M-MDSCs in vivo and evaluate their role in viral myocarditis, M-MDSCs from CVB3-infected female mice were enriched by FACS sorting and adoptively transferred into infected recipient male mice on day 3, and myocarditis severity was evaluated 4 days later. In contrast to untransferred mice, males transferred with female-derived M-MDSCs exhibited much restricted myocardial inflammation foci (Fig. 5A) as well as fewer inflammatory cells (Fig. 5B), indicating the efficient alleviation of viral myocarditis. Consistently, the cumulative 7-day survival rate also enhanced from 25% to 87.5%, although no statistically significant difference was evidenced compared with males receiving male-derived M-MDSCs or no transfer (Fig. 5C). These data indicated that only female-but not male-derived Scientific RepoRts | 6:22658 | DOI: 10.1038/srep22658 M-MDSCs efficiently ameliorated CVB3-induced myocardial inflammation. In sharp contrast, no significantly changed myocarditis was shown in female recipients following male-derived M-MDSCs (Fig. 6).

Female-derived M-MDSCs transfer potently increased regulatory and CD4 + IL-10 + T cells in CVB3-infected males.
To elucidate the mechanism underlying the protective effect of female-derived M-MDSCs on CVB3-induced myocarditis, T cell responses in the recipient male mice were examined by FACS assays. As shown in Fig. 7A, adoptive transfer substantially raised CD4 + Foxp3 + Treg frequency from 0.48-6.71%, about 14-fold higher than that of control mice; meanwhile, a 6-fold higher frequency was also achieved in CD4 + IL10 + T cells, suggesting that female-derived M-MDSCs exerted their protection against viral myocarditis mainly via promoting regulatory and IL-10-secreting T cell production. In accordance with the frequency data, absolute numbers of myocardial Treg and CD4 + IL10 + T cells robustly increased following female-derived M-MDSCs transfer, and were about 12-fold and 4-fold higher than the counterparts of control mice, respectively (Fig. 7C,D). In contrast, adoptive transfer of male-derived M-MDSCs had much less impact on CD4 + T cell subsets, although slight increase in Treg percentage and number were observed compared with control mice receiving no transfer, but far less than those in males receiving female-derived M-MDSCs (Fig. 7B,C). These data are also in agree with the limited therapeutic effect of male-derived M-MDSCs transfer (Fig. 6).

Discussion
In the present study, we found that much more myocardial MDSCs were induced in female than male mice in CVB3-induced myocarditis, and a suppressive monocytic subset (M-MDSCs) was dominant in females, while a non-suppressive granulocytic subset (G-MDSCs) was dominant in males. In addition, female-derived cells possessed a more potent T cell suppressive function, and could efficiently relieve CVB3-induced myocarditis. Yet their male-derived counterpart had limited T cell suppressive capability and exerted no such protective effect.
MDSCs are a heterogeneous population of myeloid cells, broadly defined as CD11b + Gr-1 + cells that expand under the circumstances of tumor, infection, inflammation and autoimmune diseases. Several studies have proposed a link between the increased MDSCs and the development and prognosis of viral infectious diseases [32][33][34] . In line with that, here we observed a substantial enrichment of MDSCs in CVB3-infected mice, suggesting their possible involvement in the viral myocarditis. When compared with males, female mice had fewer splenic MDSCs, but much more myocardial MDSCs, suggesting distinct enrichment patterns of MDSCs in both genders. Depletion of MDSCs with anti-Gr-1 antibody (RB6-8C5, reacting with a common epitope on Ly-6G and Ly-6C) could significantly relieve myocarditis in males, which might be attributed to the depletion of cardiac dominant non-suppressive G-MDSCs subset (about 3-fold higher than M-MDSCs). However, anti-Gr-1 antibody treatment had marginal impact on myocarditis in females, which might be caused by the both deletion of  immune-suppressive M-MDSCs (about 1.3-fold higher than G-MDSCs) and infiltrating G-MDSCs. According to the different epitopes in Gr-1, MDSCs have been subdivided into two subsets: monocytic (CD11b + Ly6G -Ly6C high ) and granulocytic (CD11b + Ly6G + Ly6C low ) cells [35][36][37] . Until now, it is still controversy whether these two populations own similar biological functions or not. Some reports indicated that both M-MDSCs and G-MDSCs possess comparable immune suppressive potentials 38,39 , whereas others demonstrated that they may exert divergent activities in a series of diseases such as infectious and autoimmune diseases 40,41 , implying complicated roles of MDSCs in pathological processes. Herein, we found that in CVB3-infected mice, myocardial M-MDSCs were the predominant T cell proliferation suppressive subset, whereas G-MDSCs were not.
Further, we found that female-derived M-MDSCs possessed a more potent suppressive function on T cell proliferation than those from males. Following adoptive transfer, only female-derived M-MDSCs efficiently alleviated CVB3-induced myocarditis, reflected by limited loci of myocardial inflammation and enhanced survival rate, while no obviously improved myocarditis was seen in mice receiving male-derived M-MDSCs. Given the immunosuppressive function of male-derived M-MDSCs observed in in vitro assays, we attributed this inconsistent in vivo data to the current transfer schedule we applied. It turned out that only one time transfer of male-derived M-MDSCs at the dose of 2 × 10 6 cells on day 3 post infection could not effectively alleviate CVB3-induced viral myocarditis; however it is reasonable to deduce that when increasing the transferring cell amounts and/or frequencies, or change the transferring time points, male-derived M-MDSCs should also exhibit obvious therapeutic effects on CVB3-induced myocarditis. Besides, we observed that following female-derived M-MDSCs transfer, myocardial viral load in male recipients increased slightly with no statistical significance. When transferring male-derived M-MDSCs to female, no noticeable improvement on viral myocarditis could be seen, reflected by the comparable pathological myocardial changes and survival rates, indicating that male-derived M-MDSCs had less immune suppressive and myocarditis-alleviating abilities. As many cytokines and molecules have been reported to influence the induction and suppressive function of tissue MDSCs and their subsets 40,42,43 , these different immune modulation functions of myocardial MDSCs in both genders may be attributed to their various myocardial environments. Moreover, female-derived M-MDSCs transfer could significantly increase Treg and CD4 + IL-10 + T cell frequencies and absolute numbers, which might work together to better control myocardial immunopathological injury in CVB3-infected male mice 44,45 . Of course, herein we could not exclude the potential effects of MDSCs on innate immune responses following CVB3 infection, such as differential macrophage polarization which has been proven to play an important role in the sex difference in the sensitivity of CVB3-induced viral myocarditis 15,46 . In fact, a line of evidence suggest that under the tumor condition, MDSCs not only possess many features of M2 macrophages (such as expressing arginase 1 and NOS2), but also might For survival rate evaluation, each group contained 8 mice. Individual experiment was conducted 3 times with similar results, and the representative data were shown. ***p < 0.001. function as the progenitors of tumor-associated macrophages 23,47 . In addition, MDSCs could further modulate macrophage mediated cytotoxicity 48 . Whether G-MDSCs or M-MDSCs could further differentiate into polarized macrophages or modulate macrophage functions in the context of CVB3-induced myocarditis need to be further interrogated. In summary, our findings suggest that compared with male mice, more abundant MDSCs dominated by monocytic phenotype and with more potent suppressive function were observed in the hearts of CVB3-infeced females, which provided protection against CVB3-induced myocarditis by increasing Treg and CD4 + IL-10 + T cells. This study may not only help us better understand the role of MDSCs in CVB3-induced myocarditis, but also provide a new explanation for the gender bias in the viral myocarditis.   Hormone treatment. Mice were administrated with 17β -Estradiol (Sigma) at a dose of 10 μg per mouse at day 1 post CVB3 infection as described previously 46 . Control mice received PBS with 0.5% ethanol without hormone.
T cell proliferation suppression assay. Myocardial M-MDSCs or G-MDSCs from CVB3-infected mice were seeded at 1 × 10 6 per well. Sorted CD4 + CD25 -CD62L high CD44 low T cells (1 × 10 6 per well) were added in the presence of purified anti-mCD3 (5 μg/ml, Biolegend) and anti-mCD28 (5 μg/ml, Biolegend) and culture for 72 h. And then cells were subjected to proliferation assays using cell proliferation ELISA BrdU kit (Roche) following the manufacturer's instructions. Absorbance at 370 nm was detected on a microplate reader (Bio-Tek). Statistical analysis. Data are presented as mean ± SEM. Differences between experimental groups were analyzed for statistical significance by one-way ANOVA test followed by Tukey's post hoc test or two-way