Chromosome Y Regulates Survival Following Murine Coxsackievirus B3 Infection

Coxsackievirus B3 (CVB3) contributes to the development of myocarditis, an inflammatory heart disease that predominates in males, and infection is a cause of unexpected death in young individuals. Although gonadal hormones contribute significantly to sex differences, sex chromosomes may also influence disease. Increasing evidence indicates that Chromosome Y (ChrY) genetic variants can impact biological functions unrelated to sexual differentiation. Using C57BL/6J (B6)-ChrY consomic mice, we show that genetic variation in ChrY has a direct effect on the survival of CVB3-infected animals. This effect is not due to potential Sry-mediated differences in prenatal testosterone exposure or to differences in adult testosterone levels. Furthermore, we show that ChrY polymorphism influences the percentage of natural killer T cells in B6-ChrY consomic strains but does not underlie CVB3-induced mortality. These data underscore the importance of investigating not only the hormonal regulation but also ChrY genetic regulation of cardiovascular disease and other male-dominant, sexually dimorphic diseases and phenotypes.


heart disease sexual dimorphism heterochromatin
Coxsackievirus infection of the heart can lead to myocarditis and cause unexpected death in individuals less than 40 years old (Woodruff 1980). Like most heart diseases, myocarditis predominates in males by a 2:1 ratio and displays a greater severity of disease over females. Susceptibility is greatest in men during the first year of life and after puberty. However, females are often rendered susceptible to myocarditis during pregnancy, during the postpartum period, and in individuals over 40 years of age, suggesting a role for hormonal influences on disease development (Woodruff 1980).
The coxsackievirus B3 (CVB3)-induced mouse model of experimental myocarditis exhibits many of the same characteristics as the human disease (Gauntt et al. 1979;Grodums and Dempster 1959b;Huber et al. 1982;Matsumori and Kawai 1980;Reyes and Lerner 1985;Wong et al. 1977). Adult male mice less than 40 weeks of age are more susceptible to CVB3-induced myocarditis than females of the same age. However, both males and females less than 3 weeks of age or over 40 weeks of age are equally susceptible, indicating that agerelated changes in sex-associated hormones may influence experimental myocarditis susceptibility (Grodums and Dempster 1959a;Lyden et al. 1987a). In this regard, susceptibility was significantly reduced in male mice that were orchiectomized prior to infection (Huber et al. 1982). However, when orchiectomized males received exogenous testosterone or progesterone prior to infection, disease susceptibility returned to similar levels seen in intact males (Huber et al. 1982). In contrast, treatment of orchiectomized males with estrogens provided protection from disease. Moreover, in vitro experiments revealed that male and female cardiomyocytes exhibited enhanced infectivity when cultured in the presence of progesterone or testosterone, but not estradiol, due to increased virus binding to the cell surface (Lyden et al. 1987b). Further studies demonstrated that testosterone induced an approximately 6-fold increase in virus receptor expression in cardiomyocytes (Lyden et al. 1987a). Therefore, these studies indicate that androgens enhance virus receptor expression and disease susceptibility, whereas estrogens provide protection from disease.
Sex hormones also influence the immune response mounted against CVB3 infection in mice. Susceptibility to CVB3-induced myocarditis in males is dependent on the induction of a CD4 + Th1 cellular response, and the immune bias toward a Th1 phenotype requires activated gd Tcells (Huber and Pfaeffle 1994;Huber et al. 2001). Orchiectomy led to an increase in immune regulatory cells in the heart, including regulatory T cells, suggesting that testosterone inhibits anti-inflammatory cell recruitment and enhances susceptibility to disease (Frisancho-Kiss et al. 2009). In contrast, resistance to disease in females was associated with a preferential Th2 response to infection (Huber and Pfaeffle 1994). However, treatment of female mice with testosterone in vivo caused a shift from the protective Th2 cell response toward the Th1 phenotype, leading to enhanced susceptibility in these mice (Huber and Pfaeffle 1994).
Clearly, testosterone plays a critical role in the sexual dimorphism in susceptibility to myocarditis by regulating not only viral infectivity but also differences in innate and adaptive antiviral immune responses. While a clear role for sex-associated hormones, and particularly testosterone, can be observed in the male-specific sexual dimorphism during CVB3 pathogenesis, the effect of Chromosome Y (ChrY) on disease has not been examined. Here, we provide direct evidence that ChrY regulates survival following CVB3 infection and suggest that the mechanism is unrelated to biological differences in testosterone production among the strains.

Virus and infection
The H3 variant of CVB3 was made from an infectious cDNA clone as previously described (Knowlton et al. 1996). Mice were infected by intraperitoneal injection of 0.5 ml of PBS containing 100 or 50 PFU CVB3 and killed seven days after infection.

Serum testosterone
Eight-week-old mice were bled by tail vein, and the samples were spun down for 20 min at 16,000 g to extract serum. Serum testosterone levels were measured using a testosterone enzyme immunoassay (EIA) kit according to the manufacturer's instructions (Assay Designs, Ann Arbor, MI). Briefly, serum samples were diluted with 1 part steroid displacement reagent for every 99 parts sample. Each sample was diluted 1:10 and 1:20 in PBS and added to the 96-well plate in duplicate, followed by the primary antibody, and incubated at room temperature for 1 hr while shaking. Wells were washed and conjugate was added to each well and incubated at room temperature for 1 hr while shaking. Wells were washed, and the pNpp substrate was added to each well and incubated at 37°for 1 hr without shaking. Stop solution was added, and the plate was read immediately with optical density at 405 nm.

Lymphocyte isolation
Perfused and intact liver, spleen, and thymus were extracted from 8week-old mice and placed in 50 ml tubes containing RPMI 1640 media with 10% FCS and shipped overnight on ice to Brown University. Thymocytes were minced and washed once in 1% PBSserum. To obtain splenic lymphocytes, spleens were minced, passed through nylon mesh (Tetko, Kansas City, MO), washed once in 1% PBS-serum, and then cell suspensions were layered on lympholyte-M (Cedarlane Laboratories Ltd., Canada). Hepatic lymphocytes were prepared by mincing and passing through a 70 mm nylon cell strainer (Falcon, Franklin Lakes, NJ). After washing three times in 1% PBSserum, cell suspensions were layered on a two-step discontinuous Percoll gradient (Pharmacia Fine Chemicals, Piscataway, NJ). Splenocytes and hepatic lymphocytes were collected after centrifugation for 20 min at 900 g.
Antibodies and flow cytometric analysis TCRb-FITC (or APC), CD8a-efluor 450 (or Pacific blue), CD4-APC (or perCP), CD44-efluor 780, NK1.1-PE-Cy7, and CD24-FITC were purchased from eBioscience (San Diego, CA). CD1d Tetramer-PE was obtained from the National Institute of Allergy and Infectious Disease MHC Tetramer Core Facility at Emory University (Atlanta, GA). Cells were suspended in buffer composed of PBS containing 1% FCS. Cells were first incubated with 2.4G2 mAb and stained with mAbs specific for cell surface markers for 20 min at room temperature. Depending on the experiment and the tissue, from 2.5 · 10 5 to 1 · 10 6 events were collected in the lymphoid gate on a FACSAria. The data were analyzed using FlowJo (Tree Star Inc., Ashland, OR).

Statistics
Statistical analyses were performed using GraphPad Prism version 5.0c (GraphPad Software, San Diego, CA). The specific tests used are detailed in the figure legends. A P value # 0.05 was considered significant.
B6-ChrY ST and B6-ChrY MET mice were the only strains that did not display a decrease in survival following infection with either 100 or 50 PFU CVB3 when compared to B6. B6-ChrY LEWES , B6-ChrY SWR , B6-ChrY SJL , B6-ChrY 129/S1 and B6-ChrY BUB mice only showed a decrease in survival compared with B6 when infected with 100 PFU CVB3 but not at the 50 PFU dose. Conversely, B6-ChrY RF , B6-ChrY MA , B6-ChrY AKR , B6-ChrY SWR , B6-ChrY WSB , and B6-ChrY A/J mice were Figure 1 ChrY polymorphism influences susceptibility to CVB3-induced mortality. B6 and B6-ChrY consomic strains were infected with 100 or 50 PFU CVB3 and monitored for survival. Significance of observed differences were determined using a log-rank (Mantel-Cox) test (n = 5-20). The legend labels indicate the mouse strain donating ChrY to B6. P values in graph title represent overall P values, and asterisks in the legend labels represent significant differences in survival between B6 and the B6-ChrY consomic strains at the indicated virus dose. Ã P , 0.05; ÃÃ P , 0.01; ÃÃÃ P , 0.001; ÃÃÃÃ P , 0.0001.
significantly more susceptible to CVB3-induced mortality at both the 100 and 50 PFU doses compared with B6 ( Figure 1). Transforming the data into a single susceptibility value by averaging the 100 and 50 PFU response groups for each strain indicated that B6-ChrY consomic strains exhibit a continuous distribution of CVB3-induced mortality phenotypes (Figure 2). Because the only genetic difference among these strains is the inheritance of ChrY, the differences in mortality rates can be directly attributed to polymorphic differences in ChrY. We have designated this locus as Y Cvb3 (CVB3 susceptibility region of ChrY). With respect to candidates for Y Cvb3 , there are three loci that are biologically relevant to CVB3-induced mortality: Y Nkt (iNKT-determining region of ChrY) (Wesley et al. 2007), Sry (sex determining region of ChrY), and Y Cmc (cardiomyocyte sizedetermining region of ChrY) (Llamas et al. 2007(Llamas et al. , 2009. Differences in basal invariant natural killer T cell numbers do not influence the mortality of CVB3-infected B6-ChrY consomic mice Invariant natural killer T (iNKT) cells are an important immune regulator during CVB3 infections, and the treatment of mice with a-galactosylceramide, an agonist for iNKT cells, protects the mice from CVB3-induced disease (Liu and Huber 2011;Wu et al. 2010). Consequently, Y Nkt , a locus that leads to a profound deficiency in iNKT cells in male B6-ChrY Nkt-129F11/Pas mice (Wesley et al. 2007), is a functionally relevant candidate for Y Cvb3 . Therefore, we analyzed the number of iNKT cells in the liver, spleen, and thymus of all B6-ChrY consomic strains by staining leukocytes for TCRb + and CD1dtetramer + cells and assessing the percentage of iNKT cells by flow cytometry (File S1).
As the various B6-ChrY consomic strains were analyzed throughout multiple experiments, the data derived from each consomic line was normalized to the B6 data from each independent experiment to control for interexperimental variability. Interestingly, we observed a continuous distribution of iNKT cells across the ChrY strains. The analysis of thymic iNKT cells produced three significant groupings ( Figure 3A). Similar continuous distributions of iNKT cells were also observed for liver and splenic iNKT cell numbers (data not shown).
Our data confirm that natural ChrY polymorphism results in differences in the percentage of basal iNKT cells in mice. However, it is important to note that further investigation is required to determine whether CVB3 infection itself changes the distribution of iNKT cells among the B6-ChrY consomic strains. Nonetheless, as the strain Figure 2 CVB3-induced mortality exhibits a continuous distribution across the B6-ChrY consomic strains. The survival data obtained from the infection with 100 and 50 PFU in Figure 1 was combined for each consomic strain and plotted in a single graph. The legend labels indicate the mouse strain donating ChrY to B6. Significance of observed differences measured using a log-rank (Mantel-Cox) test (n = 5-20). distribution pattern (SDP) for mortality is discordant with the SDP for iNKT cells in naïve B6-ChrY consomic strains, these data suggest that Y Cvb3 is not Y Nkt , and that the variations in basal iNKT cell numbers do not contribute to the phenotypic differences in CVB3-induced mortality. Moreover, these data indicate that the Y Nkt-129/Pas allele transmitted by male founder mice derived from 129F11/Pas embryonic stem cells is unique, as male mice that inherited the wild-type ChrY 129F11/Pas by natural breeding differed markedly in iNKT cell numbers ( Figure 3B). A direct comparison of these two ChrYs could in theory lead to the identification of the locus controlling iNKT cell numbers and possibly, as with yaa (Pisitkun et al. 2006;Subramanian et al. 2006), the identification of a unique translocation in ChrY Nkt-129/Pas giving rise to a dramatic reduction in iNKT cell numbers compared to naturally occurring Y Nkt alleles.
Sry allele-dependent variations in prenatal testosterone production do not underlie Y Cvb3 Sry is a particularly intriguing candidate for Y Cvb3 because Sry expression by the bipotential cells of the primordial gonad initiates testis differentiation by activating male-specific transcription factors, in particular Sox9, to induce Sertoli cell differentiation. This drives testis formation (Kashimada and Koopman 2010), including differentiation of fetal Leydig cells (Brennan et al. 2003;Clark et al. 2000;Gnessi et al. 2000;Yao et al. 2002), which produce the androgens required for masculinization of the male fetus during embryogenesis (Sriraman et al. 2005). Sertoli cells also contribute to the proper differentiation and function of adult Leydig cells (Loveland and Schlatt 1997;Russell et al. 2001;Yan et al. 2000) responsible for pubertal production of testosterone and sexual maturation (Ge and Hardy 2007). There is increasing evidence that testis determination may not be the only function of Sry, as it is also expressed in the brain, kidney, and adrenal glands of adult males (Ely et al. 2010). The existence of functionally significant Sry polymorphisms is well documented in studies using B6-ChrY consomic strains in which Sry alleles give rise to varying degrees of sex reversal, ranging from normal testis development to permanent sex reversal (Albrecht et al. 2003;Biddle and Nishioka 1988;Eicher and Washburn 1986;Eicher et al. 1982;Nagamine et al. 1987Nagamine et al. , 1999. Therefore, Sry polymorphisms could lead to differ-ences in neonatal and/or adult testosterone levels, thereby impacting susceptibility to CVB3-induced mortality. To test our hypothesis, we compared the SDP of Sry polymorphisms with the SDP of CVB3-induced mortality. The B6-ChrY consomic lines can be separated into two categories based on the evolutionary origin of the Mus musculus species donating ChrY, including musculus and domesticus. Within the domesticus ChrY consomic lines, two groups of mice exist based on their Sry alleles (Table 1) (Washburn et al. 2001). Domesticus Group A is defined as having Sry alleles with 13 CAG repeats in glutamine repeat cluster 3, which results in sex reversal when inherited by B6 mice heterozygous for the Chr17 TOrleans mutation. Mice within this group have reduced Sry expression in the fetal gonad, which is insufficient for normal development. Domesticus Group B is defined as having Sry alleles with 12 CAG repeats in glutamine repeat cluster 3 and a 10 bp deletion in the 59 UTR that allows for normal development in B6 mice heterozygous for the TOrleans mutation and normal Sry expression levels in the fetal gonad. However, our analysis of the consomic lines revealed that the SDP for CVB3-induced mortality is discordant with the SDP for Sry alleles (Table 1). These data suggest that the Sry allelic variations having the potential to influence testis development and neonatal testosterone production are not likely to underlie Y Cvb3 .
Variation in adult testosterone levels do not account for ChrY regulation of CVB3-induced mortality Testosterone clearly plays a critical role in the sexual dimorphism in susceptibility to myocarditis by regulating both viral infectivity and differences in innate and adaptive antiviral immune responses. Therefore, we hypothesized that allelic differences in Sry could also in theory influence the production of postpubertal testosterone in CVB3-infected B6-ChrY consomic mice and lead to differences in mortality among the strains. To test this, blood was collected from male mice at eight weeks of age that were bled between 1 and 3 PM to minimize daily fluctuations in testosterone production. Serum testosterone levels were measured using a testosterone EIA kit (File S1). First, adult testosterone levels did not correlate with Sry alleles, suggesting that Sry allelic variations with the potential to influence testis development do not influence adult testosterone levels (Table 1). Only B6-ChrY RF consomic mice exhibited significant differences in adult testosterone levels compared with B6 Figure 4. Thus, elevated testosterone levels may contribute to increased mortality in B6-ChrY RF , but an increase in adult testosterone levels does not explain the enhanced mortality seen in the other 11 B6-ChrY consomic strains. However, whether ChrY influences other steroid hormones and whether ChrY influences virus receptor expression in the heart independently of testosterone levels remain under consideration. Figure 4 Differences in CVB3-induced mortality are not the result of changes in adult serum testosterone levels among the B6-ChrY consomic strains. Serum testosterone was measured in 8-week-old B6-ChrY consomic male mice and compared with wild-type B6 by EIA. X-axis labels indicate the mouse strain donating ChrY to B6. Only B6-ChrY RF mice exhibited a significant elevation in testosterone compared with B6. Significance of observed differences was determined by one-way ANOVA, followed by a Dunnett's multiple comparison test. n = 5 mice per strain. ÃÃÃ P # 0.001. n a Increased susceptibility to CVB3-induced mortality at 100 and 50 PFU compared with B6. b Increased susceptibility to CVB3-induced mortality at 100 PFU compared with B6. c Equal susceptibility at both virus doses to CVB3-induced mortality compared with B6.
Our data suggest that Sry polymorphisms neither underlie Y Cvb3 nor control adult serum testosterone levels. Nevertheless, dependence of CVB3-induced mortality on both testosterone and ChrY polymorphism makes Y Cmc , a ChrY polymorphism that impacts the sensitivity of cardiomyocytes to the hypertrophic effects of postpubertal testosterone, a strong candidate for Y Cvb3 . Llamas et al. (2007) showed that the surface area of cardiomyocytes from B6 mice is significantly larger than cardiomyocytes from B6-ChrY A/J mice. Orchiectomy of B6 and B6-ChrY A/J consomic mice resulted in a decrease in the size of B6 cardiomyocytes but not in the size of B6-ChrY A/J cardiomyocytes (Llamas et al. 2007(Llamas et al. , 2009. Furthermore, treatment of orchiectomized mice with exogenous testosterone effectively increased the size of cardiomyocytes in B6 mice but not in B6-ChrY A/J mice (Llamas et al. 2009). Therefore, these differences are not due to inherent changes in testosterone production between the strains but, rather, due to the insensitivity of cardiomyocytes from B6-ChrY A/J mice to the hypertrophic effects of postpubertal testosterone, which is a direct consequence of Y Cmc . This same effect may contribute to the differences in CVB3-induced mortality among the B6-ChrY consomic lines, and it is being actively investigated.
Our data suggest that the functional Sry alleles influencing testis differentiation are unlikely to be responsible for Y Cvb3 or Y Nkt . Moreover, a comparison of the available ChrY sequence data (phenome.jax. org) indicates that there are no single nucleotide polymorphisms within the NPAR candidates that uniquely distinguish the Y Cvb3 alleles influencing mortality (Table 1). Thus, it is important to also consider the impact that ChrY structural polymorphism may have on phenotypic differences among the B6-ChrY consomic strains. Structural polymorphism can arise through variations in the number of repeat sequences, inverted sequences, and retroelements composing heterochromatin. As seen in Drosophila, Y-linked heterochromatin can influence the epigenetic regulation of autosomal and ChrX gene expression through its interaction with chromatin, thereby epigenetically regulating differences in males (Lemos et al. 2008(Lemos et al. , 2010. Therefore, the differences in CVB3-induced mortality observed among the consomic strains may be the consequence of epigenetic regulation by ChrY resulting from heterochromatic polymorphism. Delineating the underlying biological components regulating sex differences is critical to appropriate therapeutic treatment of sexually dimorphic diseases. Increasing evidence indicates that ChrY polymorphism can impact biological functions unrelated to male reproduction, such as autoimmune and cardiovascular diseases and hypertension, but the mechanisms behind these alternative functions remain unknown (Llamas et al. 2007(Llamas et al. , 2009Teuscher et al. 2006). Fortunately, the use of ChrY consomic strains of mice presents the opportunity to reveal the polymorphism underlying these unconventional biological functions. Clearly, defining the genetic basis of ChrY functional polymorphism is of considerable importance to human health and disease, particularly in those settings where a male-specific sexual dimorphism dominates.