The developing mammalian brain is destined for a female phenotype unless exposed to gonadal hormones during a perinatal sensitive period. It has been assumed that the undifferentiated brain is masculinized by direct induction of transcription by ligand-activated nuclear steroid receptors. We found that a primary effect of gonadal steroids in the highly sexually dimorphic preoptic area (POA) is to reduce activity of DNA methyltransferase (Dnmt) enzymes, thereby decreasing DNA methylation and releasing masculinizing genes from epigenetic repression. Pharmacological inhibition of Dnmts mimicked gonadal steroids, resulting in masculinized neuronal markers and male sexual behavior in female rats. Conditional knockout of the de novo Dnmt isoform, Dnmt3a, also masculinized sexual behavior in female mice. RNA sequencing revealed gene and isoform variants modulated by methylation that may underlie the divergent reproductive behaviors of males versus females. Our data show that brain feminization is maintained by the active suppression of masculinization via DNA methylation.
Subscribe to Journal
Get full journal access for 1 year
only $4.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Gene Expression Omnibus
NCBI Reference Sequence
McCarthy, M.M. Estradiol and the developing brain. Physiol. Rev. 88, 91–124 (2008).
McCarthy, M.M. & Arnold, A.P. Reframing sexual differentiation of the brain. Nat. Neurosci. 14, 677–683 (2011).
Mangelsdorf, D.J. et al. The nuclear receptor superfamily: the second decade. Cell 83, 835–839 (1995).
Kudwa, A.E., Michopoulos, V., Gatewood, J. & Rissman, E. Roles of estrogen receptors α and β in differentiation of mouse sexual behavior. Neuroscience 138, 921–928 (2006).
Sato, T. et al. Brain masculinization requires androgen receptor function. Proc. Natl. Acad. Sci. USA 101, 1673–1678 (2004).
Amateau, S.K. & McCarthy, M.M. A novel mechanism of dendritic spine plasticity involving estradiol induction of prostaglandin-E2. J. Neurosci. 22, 8586–8596 (2002).
Bakker, J., Honda, S.I., Harada, N. & Balthazart, J. The aromatase knock-out mouse provides new evidence that estradiol is required during development in the female for the expression of sociosexual behaviors in adulthood. J. Neurosci. 22, 9104–9112 (2002).
Xu, X. et al. Modular genetic control of sexually dimorphic behaviors. Cell 148, 596–607 (2012).
Simerly, R.B. Wired for reproduction: organization and development of sexually dimorphic circuits in the mammalian forebrain. Annu. Rev. Neurosci. 25, 507–536 (2002).
Morris, J.A., Jordan, C.L. & Breedlove, S.M. Sexual differentiation of the vertebrate nervous system. Nat. Neurosci. 7, 1034–1039 (2004).
Guo, J.U. et al. Neuronal activity modifies the DNA methylation landscape in the adult brain. Nat. Neurosci. 14, 1345–1351 (2011).
Yoo, C.B., Cheng, J.C. & Jones, P.A. Zebularine: a new drug for epigenetic therapy. Biochem. Soc. Trans. 32, 910–912 (2004).
Roth, T.L. & Sweatt, J.D. Regulation of chromatin structure in memory formation. Curr. Opin. Neurobiol. 19, 336–342 (2009).
Siedlecki, P. et al. Discovery of two novel, small-molecule inhibitors of DNA methylation. J. Med. Chem. 49, 678–683 (2006).
Satoh, A. et al. Neurabin-II/spinophilin. J. Biol. Chem. 273, 3470–3475 (1998).
Wright, C.L., Burks, S.R. & McCarthy, M.M. Identification of prostaglandin E2 receptors mediating perinatal masculinization of adult sex behavior and neuroanatomical correlates. Dev. Neurobiol. 68, 1406–1419 (2008).
Everitt, B.J. Sexual motivation: a neural and behavioural analysis of the mechanisms underlying appetitive and copulatory responses of male rats. Neurosci. Biobehav. Rev. 14, 217–232 (1990).
Cardoso, M.C. & Leonhardt, H. DNA methyltrasferase is actively retained in the cytoplasm during early development. J. Cell Biol. 147, 25–32 (1999).
LaPlant, Q. et al. Dnmt3a regulates emotional behavior and spine plasticity in the nucleus accumbens. Nat. Neurosci. 13, 1137–1143 (2010).
Jeltsch, A. & Jurkowska, R.Z. New concepts in DNA methylation. Trends Biochem. Sci. 39, 310–318 (2014).
Wedemeyer, J. & Galli, S.J. Mast cells and basophils in acquired immunity. Br. Med. Bull. 56, 936–955 (2000).
Silverman, A.J., Asarian, L., Khalil, M. & Silver, R. GnRH, brain mast cells and behavior. Prog. Brain Res. 141, 315–325 (2002).
Amateau, S.K. & McCarthy, M.M. Induction of PGE2 by estradiol mediates developmental masculinization of sex behavior. Nat. Neurosci. 7, 643–650 (2004).
Lauber, M.E., Sarasin, A. & Lichtensteiger, W. Transient sex differences of aromatase (CYP19) mRNA expression in the developing rat brain. Neuroendocrinology 66, 173–180 (1997).
McCarthy, M.M. et al. Sex differences in the brain: the not so inconvenient truth. J. Neurosci. 32, 2241–2247 (2012).
Bowers, J.M., Perez-Pouchoulen, M.R. & McCarthy, M.M. Foxp2 mediates sex differences in ultrasonic vocalization by rat pups and directs order of maternal retrieval. J. Neurosci. 33, 3276–3283 (2013).
Maunakea, A.K. et al. Conserved role of intragenic DNA methylation in regulating alternative promoters. Nature 466, 253–257 (2010).
Takahashi, H. et al. Drebrin-dependent actin clustering in dendritic filopodia governs synaptic targeting of postsynaptic density-95 and dendritic spine morphogenesis. J. Neurosci. 23, 6586–6595 (2003).
Luco, R.F. et al. Epigenetics in alternative pre-mRNA splicing. Cell 144, 16–26 (2011).
Guo, J.U. et al. Distribution, recognition and regulation of non-CpG methylation in the adult mammalian brain. Nat. Neurosci. 17, 215–222 (2014).
Lister, R. et al. Global epigenomic reconfiguration during mammalian brain development. Science 341, 1237905 (2013).
Miller, C.A. et al. Cortical DNA methylation maintains remote memory. Nat. Neurosci. 13, 664–666 (2010).
Sabatino, M.A. et al. Zebularine partially reverses GST methylation in prostate cancer cells and restores sensitivity to the DNA minor groove binder brostallicin. Epigenetics 8, 656–665 (2013).
Champion, C. et al. Mechanistic insights on the inhibition of C5 DNA methyltransferases by zebularine. PLoS ONE 5, e12388 (2010).
Globisch, D. et al. Tissue distribution of 5-hydroxymethylcytosine and search for active demethylation intermediates. PLoS ONE 5, e15367 (2010).
Williams, K., Christensen, J. & Helin, K. DNA methylation: TET proteins-guardians of CpG islands? EMBO Rep. 13, 28–35 (2012).
Niehrs, C. & Schafer, A. Active DNA demethylation by Gadd45 and DNA repair. Trends Cell Biol. 22, 220–227 (2012).
Tan, L. & Shi, Y.G. Tet family proteins and 5-hydroxymethylcytosine in development and disease. Development 139, 1895–1902 (2012).
Gavin, D.P. et al. Growth arrest and DNA-damage-inducible, beta (GADD45b)-mediated DNA demethylation in major psychosis. Neuropsychopharmacology 37, 531–542 (2012).
Sodersten, P. Mounting behavior in the female rat during the estrous cycle, after ovariectomy, and after estrogen or testosterone administration. Horm. Behav. 3, 307–320 (1972).
Kimchi, T., Xu, J. & Dulac, C. A functional circuit underlying male sexual behaviour in the female mouse brain. Nature 448, 1009–1014 (2007).
Uhlenhaut, N.H. et al. Somatic sex reprogramming of adult ovaries to testes by FOXL2 ablation. Cell 139, 1130–1142 (2009).
Dodge, J.E. et al. Inactivation of dnmt3b in mouse embryonic fibroblasts results in dna hypomethylation, chromosomal instability, and spontaneous immortalization. J. Biol. Chem. 280, 17986–17991 (2005).
Todd, B.J., Schwarz, J.M. & McCarthy, M.M. Prostaglandin-E2: a point of divergence in estradiol-mediated sexual differentiation. Horm. Behav. 48, 512–521 (2005).
McCarthy, M.M., Schlenker, E.H. & Pfaff, D.W. Enduring consequences of neonatal treatment with antisense oligodeoxynucleotides to estrogen receptor messenger ribonucleic acid on sexual differentiation of rat brain. Endocrinology 133, 433–439 (1993).
R Core Team. R: a language and environment for statistical computing. http://www.r-project.org (R Foundation for Statistical Computing, 2014).
Glaser, E.M. & Van der Loos, H. Analysis of thick brain sections by obverse-reverse computer microscopy: application of a new, high clarity Golgi-Nissl stain. J. Neurosci. Methods 4, 117–125 (1981).
Kimchi, T., Xu, J. & Dulac, C. A functional circuit underlying male sexual behavior in the female mouse brain. Nature 448, 1009–1014 (2007).
Park, J.H., Bonthuis, P., Ding, A., Rais, S. & Rissman, E.F. Androgen- and estrogen-independent regulation of copulatory behavior following castration in male B6D2F1 mice. Horm. Behav. 56, 254–263 (2009).
Krueger, F. & Andrews, S.R. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics 27, 1571–1572 (2011).
Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Akalin, A. et al. MethylKit: a comprehensive R package for the analysis of genome-wide DNA methylation profiles. Genome Biol. 13, R87 (2012).
Trapnell, C., Pachter, L. & Salzberg, S.L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009).
Anders, S. & Huber, W. Differential expression analysis for sequence count data. Genome Biol. 11, R106 (2010).
Warnes, G.R. et al. gplots: various R programming tools for plotting data. R package version 2.14.1. http://CRAN.R-project.org/package=gplots (2014).
We thank B.K. Krueger and S.M. Thompson for their helpful input on this manuscript. We thank G. Fan (University of California, Los Angeles) for kindly providing the Dnmt3aloxP/loxP mice. This work was supported by grant R01 MH052716 to M.M.M. and F31NS073545-01 to B.M.N., and R21 MH099562 to S.J.R. This work was conducted as part of the doctoral thesis requirements of B.M.N.
The authors declare no competing financial interests.
Integrated supplementary information
On PN0, PN1, and PN4 females had significantly higher levels of Dnmt activity in the POA than males (PN0 t(8) = 2.28, p = 0.05; PN1 t(8) = –3.92, p = 0.0044; PN4 t(5) = 2.65, p = 0.0453). Prior to (on Embryonic day 20 (E20; t(5) = 2.23, p = 0.0766)) and after the perinatal androgen surge, there were no significant sex differences in Dnmt activity in the POA (PN7 t(5) = 0.94, p = 0.3919; PN14 t(9) = 0.67, p = 0.5193). Twenty four hours after estradiol injection there was a significant reduction in Dnmt activity in females on PN1 (t(8) = –2.52, p = 0.036), but not PN14 (t(6) = 0.42, p = 0.6928). Data are graphed as percent change from the control males. Statistics were performed on the actual data and no comparisons were made across time points because the values are relative and therefore not comparable across assays. Data shown here were obtained via colormetric assay. Time course was subsequently repeated (Fig.1) with a more-sensitive fluorometric assay. E20: n = 3 female, 4 male. PN0: n = 5 female, 5 male. PN1: n = 5 female, 5 male, 5 female + e; PN4: n = 4 female, 3 male; PN7: n = 3 female, 4 male; PN14: n = 4 female, 6 male, 4 female + e.*p < 0.05, data are represented as mean ± s.e.m.
(a) Most CpG sites in the rat POA were highly methylated (2-Way ANOVA, F%meth(9,60) = 1244, p < 0.0001), and females had significantly more CpG sites with methylation at 90–100% compared to males and females treated with e (Fsex(2,60) = 2.885, p = 0.0636; Tukey's HSD, 90–100% female vs. male p < 0.0001, female vs. female + e p < 0.0001. Fint(18,60) = 7.395, p < 0.0001). Males (Tukey's HSD, p < 0.0001) and estradiol-treated females (Tukey's HSD, p = 0.0014) had a greater number of CpG sites 80–90% methylated compared to females. Estradiol-treated females had fewer sites that are 0–10% methylated compared to control females (Tukey's HSD, p = 0.0055). (b) Conversely, most CHG sites had low levels of methylation (F%meth (9,60) = 8959, p < 0.0001), and all groups differed in the number of sites that are 0-10% methylated (Fsex (2,60) = 12.73, p < 0.0001; Tukey's HSD, 0–10% female vs. male p < 0.0001, female vs. female + e p < 0.0001, male vs. female + e p < 0.0001. Fint(18,60) = 23.8, p < 0.0001). Males and females treated with e had more CHG sites methylated 10–20% compared to control females (Tukey's HSD, p < 0.0001). Only sites with at least 3 reads across all 9 samples were included in this analysis. n = 3 female, 3 male, 3 female + e. **p < 0.01, ***p < 0.001, data are represented as mean ± s.e.m.
Percent methylation in males is plotted relative to percent methylation in females by chromosome. There were no overall differences in CpG methylation by chromosome (χ2 = 30.97, n = 20, p = 0.0556), average percent methylation is based on n = 3 female, n = 3 male.
Quantification of Dnmt protein levels via western blot on PN1 following 24 hours of oil or e indicates no significant hormonal modulation of enzyme levels (ANOVA, Dnmt1 F(2,13) = 1.188, p = 0.336; Dnmt3a F(2,13) = 0.387, p = 0.687; Dnmt3b F(2,13) = 0.622, p = 0.552). Cropped representative western bands are shown and full length blots are provided in Supplementary Figure 10. Ponceau S staining of proteins was used as a loading control..n= 6 male, 7 female, 3 female + e. Data are represented as mean ± s.e.m,
Supplementary Figure 5 Neonatal Dnmt inhibition masculinizes adult copulatory behavior and Neurabin II protein levels in the POA
(a) RG108 administration on PN0 and PN1 significantly increased the number of mounts (ANOVA, F(3,26) = 5.31, p = 0.0054) and number of thrusts (F(3,27) = 19.97, p < 0.0001) displayed by females in adulthood. Latency to display first mount was not significantly reduced by neonatal RG108 treatment (F(3,26) = 3.104, p = 0.0439, p-value reflects sex difference, not effect of RG108), however the latency to display first thrust behavior was masculinized in females treated with the inhibitor (F(3,20) = 6.60, p = 0.0028). (b) Protein levels of the synaptic marker Neurabin II were significantly higher in RG108-treated females compared to control females (a priori t-test; t(14)=2.301, p = 0.0373.), although there were no main effects of treatment or sex by ANOVA (F(3,26) = 2.008, p = 0.1375). Cropped representative Neurabin II western blot bands with corresponding Ponceau S loading control shown. Full blots are provided in Supplementary Figure 10. (c) Animals treated with RG108 showed no impairments in locomotor activity as measured by grid crosses (F(3,28) = 1.21, p = 0.33) and rears (F(3,28) = 2.15, p = 0.17), however time spent in the center of the open field arena was masculinized by RG108 (F(3,26) = 2.86, p = 0.049). Mounts: n = 9 male, 8 male + RG108, 7 female, 8 female + RG108. Thrusts: n = 9 male, 6 male + RG108, 8 female, 8 female + RG108. Mount latency: n = 9 male, 6 male + RG108, 7 female, 8 female + RG108. Thrust latency: n = 9 male, 6 male + RG108, 3 female, 6 female + RG108 (a), n = 8 male, 7 male + RG108, 8 female, 8 female + RG108 (b), Grid crosses: n = 8 male, 6 male + RG108, 8 female, 7 female + RG108. Rears: n = 8 male, 6 male + RG108, 8 female, 7 female + RG108. Time in center: n = 8 male, 6 male + RG108, 8 female, 7 female + RG108 (c), *p < 0.05, **p < 0.01, ***p < 0.001 compared to control female by ANOVA. #p < 0.05 by t-test. Data are represented as mean ± s.e.m.
Females were treated with either vehicle or Zeb on PN0 and PN1 and raised to adulthood. (a) Estrous cycle stage was noted for 15 consecutive days. (b) There were no differences in the number of days spent in diestrus (t-test, t(9) = 1.761, p = 0.112), proestrus (t(9)=0.869, p = 0.407), or estrus (t(9) = 1.649, p = 0.1335) across treatment groups and Zeb-treated females appeared to cycle normally. n = 4 vehicle, 7 Zeb.
Supplementary Figure 7 qPCR confirmation of sex and estradiol-mediated differences in aromatase and Neurabin II gene expression patterns in the POA
Rat pups were treated with estradiol or vehicle on PN0 and PN1. mRNA from the POA was collected 6, 48, 96 or 168 hours later. (a) There was a significant effect of sex/hormone treatment on Cyp19a1 (aromatase) mRNA expression in the POA (2-way ANOVA, FSex(2, 49) = 7.264, p = 0.0017), whereby control females had lower levels of aromatase on PN2 compared to males and estradiol-treated females (p = 0.05). On PN7, estradiol-treated females had significantly higher levels of aromatase mRNA compared to control females (p = 0.0325). There was also a significant effect of age on aromatase expression in the POA (FAge(3, 49) = 5.487, p = 0.0025), however there were no significant statistical interactions between age and sex on aromatase expression (FInt(6, 49) = 1.518, p = 0.1920). (b) There were main effects of both sex and age on Ppp1r9b (Neurabin II) mRNA levels in the developing POA (2-way ANOVA, FSex(2, 52) = 7.817, p = 0.0011; FAge(3, 52) = 4.258, p = 0.0092; Fint(6, 52) = 0.8048, p = 0.5708). On PN2, females had significantly lower levels of Neurabin II compared to males and females treated with 100μg E for 24 hours (p = 0.0255) and at PN4 females treated with estradiol had significantly higher levels than both males and control females (p = 0.0109). Cyp19a1 PN0: n = 6 female, 4 male, 5 female + e; PN2: n = 6 female, 5 male, 6 female + e; PN4: n = 5 female, 4 male, 6 female + e; PN7: n = 5 female, 6 male, 3 female + e (a). Ppp1r9b PN0: n = 6 female, 5 male, 5 female + e; PN2: n = 6 female, 5 male, 5 female + e; PN4: n = 5 female, 4 male, 6 female + e; PN7: n = 5 female, 6 male, 5 female + e (b), *p < 0.05 compared to vehicle treated female. Data are represented as mean ± s.e.m.
The methylation status of CpG islands within the promoters of multiple candidate genes was quantified by 454 sequencing following bisulfite conversion on PN2 POA DNA. Most genes exhibited the level of CpG methylation typical of promoter regions, in the 5–10% range, whereas both Ppp1r9b (Neurabin II) and Cyp19a1 (aromatase) were notable for the high levels of CpG methylation, exceeding 50% in the case of aromatase (n = 9; data are the mean of 3 males, 3 females and 3 estradiol-treated females, there were no significant differences between groups in the amount of promoter CpG methylation between groups for a particular gene).
We used qPCR on PN2 POA mRNA to confirm the expression patterns of a subset of genes identified as having sex differences by RNA-Seq. Confirming male-biased gene expression patterns, Nr2f2 (t(10) = 2.175, p = 0.054), Ebf2 (t(10) = 1.976, p = 0.076), and Ebf3 (t(10) = 2.178, p = 0.054), were marginally higher in males compared to females. There were no statistically significant differences in the female-biased gene Darp-32 when measured by qPCR (t(7) = 1.487, p = 0.180). An additional female-biased gene, Adora2a was significantly higher in the female POA compared to males (t(9) = 2.482, p = 0.0349). Nr2f2: n = 6 male, 6 female. Ebf2: n = 6 male, 6 female. Ebf3: n = 6 male, 6 female. Darp-32: n = 4 male, 5 female. Adora2a: n = 5 male, 6 female. *p < 0.05 compared to female. Data are represented as mean ± s.e.m.
Western blot films and Ponceau S stained membranes are shown for each protein quantification experiment referenced throughout the paper. Details of western blot protocol, antibody catalog numbers and concentrations used are described in the Methods section. Neurabin II molecular weight ~120 kDa, Dnmt1 ~84 kDa, Dnmt3a ~130kDa, Dnmt3b ~100kDa.
Supplementary Figures 1–10 (PDF 3145 kb)
(a) Genes with higher expression in control males vs. control females. Male (n=3) and female (n=2) rat pups from two litters were treated with vehicle (1% DMSO) on PN0 and PN1 and RNA was collected from the POA on PN2. RNA-Seq showed that males had 34 genes with significantly higher expression in the POA compared to females. (b) Genes with higher expression in control females vs. control males. We detected 36 genes with significantly higher expression in the female POA compared to males. (XLSX 52 kb)
Male and female rat pups from two litters were treated with vehicle (1% DMSO; n=3 females, n=3 males) or Zeb (300ng, n=3 females) on PN0 and PN1 and RNA was collected from the POA on PN2. Of the 34 genes that were significantly higher in control males relative to control females (Supplementary Table 1a), 24 were significantly increased by Zeb treatment in females. (XLSX 52 kb)
(a) Gene isoforms exclusive to the male POA. Male (n=3) and female (n=2) rat pups from two litters were treated with vehicle (1% DMSO) on PN0 and PN1 and RNA was collected from the POA on PN2. RNA-Seq showed that there were 32 gene isoforms expressed in the male POA, but not the female POA. (b) Gene isoforms exclusive to the female POA. We detected 37 gene isoforms expressed in the female POA, but not the male POA. (XLSX 61 kb)
(a) Male biased gene isoforms increased in females following DNMT inhibition. Male and female rat pups from two litters were treated with vehicle (1% DMSO; n=3 females, n=3 males) or Zeb (300ng, n=3 females) on PN0 and PN1 and RNA was collected from the POA on PN2. Gene isoforms that were significantly higher in males compared to females, which were significantly increased in females with Zeb are listed. (b) Female biased gene isoforms decreased in females following DNMT inhibition. Gene isoforms that were significantly higher in females compared to males, which were significantly decreased in females with Zeb are listed. (XLSX 59 kb)
Male (n=3) and female (n=3) rat pups were treated with vehicle (sesame oil), and an additional group of females (n=3) were treated with estradiol (100 μg) on PN0 and PN1. POA DNA was collected on PN4, bisulfite converted and subject to whole genome bisulfite sequencing. Bismark processing statistics are reported for each library. (XLSX 34 kb)
About this article
Cite this article
Nugent, B., Wright, C., Shetty, A. et al. Brain feminization requires active repression of masculinization via DNA methylation. Nat Neurosci 18, 690–697 (2015). https://doi.org/10.1038/nn.3988
Clinical Epigenetics (2021)
Journal of Neuroinflammation (2021)
British Dental Journal (2021)
Sex-differential DNA methylation and associated regulation networks in human brain implicated in the sex-biased risks of psychiatric disorders
Molecular Psychiatry (2021)
Biology of Sex Differences (2020)