- Email: [email protected]
Androgen receptor distribution in the social decision-making network of eusocial naked mole-rats
Behavioural Brain Research 256 (2013) 214–218
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Short communication
Androgen receptor distribution in the social decision-making network of eusocial naked mole-rats Melissa M. Holmes a,b,∗ , Spencer Van Mil a , Camila Bulkowski a , Sharry L. Goldman c , Bruce D. Goldman c , Nancy G. Forger d a
Department of Psychology, University of Toronto Mississauga, Mississauga, Ontario L5L 1C6, Canada Department of Cell and Systems Biology, University of Toronto, Toronto, Ontario M5S 3G5, Canada c Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT 06269, USA d Neuroscience Institute, Georgia State University, Atlanta, GA 30302, USA b
h i g h l i g h t s • • • •
Naked mole-rats are eusocial rodents, living in large social groups with strict hierarchies. Subordinate naked mole-rats have more androgen receptor protein in the social decision-making network than breeders. Naked mole-rats exhibit some unique features in neural androgen receptor distribution. Social status controls androgen receptor protein in the brain of eusocial rodents.
a r t i c l e
i n f o
Article history: Received 23 June 2013 Received in revised form 13 August 2013 Accepted 15 August 2013 Available online 21 August 2013 Keywords: Androgen receptor Anterior hypothalamus Lateral septum Naked mole-rat Preoptic area Social status
a b s t r a c t Naked mole-rats are highly social rodents that live in large groups and exhibit a strict reproductive and social hierarchy. Only a few animals in each colony breed; the remainder are non-reproductive and are socially subordinate to breeders. We have examined androgen receptor immunoreactive (AR+) cells in brain regions comprising the recently described social decision-making network in subordinate and breeder naked mole-rats of both sexes. We find that subordinates have a significantly higher percentage of AR+ cells in all brain regions expressing this protein. By contrast, there were no significant effects of sex and no sex-by-status interactions on the percentage of AR+ cells. Taken together with previous findings, the present data complete a systematic assessment of the distribution of AR protein in the social decision-making network of the eusocial mammalian brain and demonstrate a significant role for social status in the regulation of this protein throughout many nodes of this network. © 2013 Elsevier B.V. All rights reserved.
Social behaviors are broadly controlled by a complex neural network and, in turn, social interactions feed back to sculpt the brain throughout life. O’Connell and Hofmann [1] recently described those brain regions involved in the expression of motivated behaviors and their conservation across species, ultimately proposing the social decision-making (SDM) network. The SDM network essentially marries the social behavior network [2] and mesolimbic reward system to form an integrated network underlying the expression of adaptive motivated behaviors in vertebrates. It is clear that to better understand general mechanisms associated with the neural control of social interactions and concomitant
∗ Corresponding author at: Department of Psychology, University of Toronto Mississauga, 3359 Mississauga Road North, Mississauga, Ontario L5L 1C6, Canada. Tel.: +1 905 828 3956; fax: +1 905 569 4326. E-mail address: [email protected] (M.M. Holmes). 0166-4328/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbr.2013.08.025
socially controlled neural plasticity, we need a more systematic and comprehensive characterization of these brain regions in diverse species. Indeed, species that exhibit striking adaptations in their social interactions are particularly valuable for such comparisons as they better illuminate those features that are the exception versus those that are the rule. Naked mole-rats are a powerful animal model for understanding the reciprocal relationship between social interactions and brain plasticity. They are eusocial and display the most rigid social hierarchy and reproductive skew among mammals [3]. These animals live in large subterranean colonies of up to 300 animals and reproduction is restricted to a single breeding female and her 1–3 male consorts [4,5]. Breeders are socially dominant over all other members of the colony, which are non-reproductive subordinates [6]. We have previously shown various neural changes associated with transitions in social status in this species (reviewed in [7]). For example, the paraventricular nucleus, bed nucleus of the stria
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218
215
Fig. 1. Photomicrographs of AR immunoreactivity in the (A) dorsal striatum, (B) nucleus accumbens (NAc), (C) lateral septum (LS), (D) ventral pallidum (VP), (E) medial preoptic area (mPOA), (F) anterior hypothalamus (AH), (G) basolateral amygdala, ventral portion (BLV); inset is a higher magnification view of the BLV, (H) periaqueductal gray (PAG), and (I) ventral tegmental area (VTA) of a subordinate male naked mole-rat. Midline is on the right side of each image (except H). Other abbreviations: ac = anterior commissure; BLA = basolateral amygdala; ec = external capsule; ICj = Island of Calleja major; LA = lateral amygdala; LV = lateral ventricle; MS = medial septum; SN = substantia nigra; 3V = third ventricle. Scale bar at lower right: 100 m for A–F and 200 m for H and I. Scale bar in G: 200 m for larger view and 100 m for inset.
terminalis and medial amygdala, all brain regions important for reproduction in other species, are larger in breeder than subordinate naked mole-rats [8,9]. To begin to understand the putative endocrine mechanisms associated with changes in status and corresponding changes in brain morphology and social behavior, we previously evaluated the expression of androgen receptor (AR) protein in the same brain areas in which we saw morphological plasticity [10]. Social status significantly altered AR immunoreactivity in regions that increased in size in breeders, however, an inverse relationship was found whereby subordinates had more AR+ cells than breeders. The present report describes our efforts to more fully characterize the neuroendocrinology and neurobiology of these unique animals. Using the same tissue in our initial report [10], we have analyzed AR immunoreactivity in the remaining regions of the SDM network of breeding and subordinate naked mole-rats of both sexes. Brains from 5 breeding females, 5 breeding males, 4 subordinate females, and 4 subordinate males were collected; all animals were gonadally intact. Brains were removed, immersion fixed in 5% acrolein for 4 h, transferred to 30% sucrose in 0.1 M phosphate buffer for cryoprotection and sectioned into 4 series in the coronal plane using a sliding microtome. One series was processed for AR immunohistochemistry using the polyclonal AR antiserum, PG21, and counter-stained with methyl green, as fully described in [10]. In addition, one series from each of three gonadally intact adult male C57BL/6 mice was processed concurrently as a positive control and for between-species comparisons. Qualitative analyses indicated that not all brain regions involved in the SDM network contained high levels of AR+ cells. The dorsal
striatum, nucleus accumbens (NAc) and ventral pallidum (VP) had little to no AR immunoreactivity (Fig. 1A, B and D). Similarly, while there was consistent albeit sparse AR in the lateral amygdala, the basolateral amygdala (BLA) was unlabeled with the exception of the ventral portion (Fig. 1G). AR immunoreactivity was consistently present in the periaqueductal gray (PAG) and ventral tegmental area (VTA) though quantification of these regions was not possible due to insufficient tissue for all animals in all groups. Quantitative stereological analyses of the percentage of cells positive for AR immunoreactivity in the lateral septum (LS; equivalent to plates 25–30 in the mouse brain atlas of [11]; Figs. 1C and 2A), medial preoptic area (mPOA; plates 30–33; Figs. 1E and 2B), anterior hypothalamus (AH; plates 35–38; Figs. 1F and 2C), and basolateral amygdala, ventral portion (BLV; plates 40–42; Figs. 1G and 2D) were performed using StereoIoger software (Stereology Resource Center, Inc.). Outlines of each region were traced in each section, and unbiased estimates of the number of darkly labeled, lightly labeled, and unlabeled cells were obtained using the optical disector method. Counting frames were 20 m × 20 m and frame size was held constant across animals. Sampling grid size varied from 60 m × 60 m to 200 m × 200 m depending on brain region, and was held constant for all animals. All brain regions were analyzed bilaterally unless tearing or tissue artifacts were present in the region of interest. Because the pattern of results was the same whether only darkly labeled or all labeled cells were considered, we have combined dark plus light cell counts in the analyses presented here, collectively referred to as AR+. To account for possible group differences in total cell number in some brain regions [8], the percentage of AR+ cells was calculated
216
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218
Fig. 2. High magnification photomicrographs of AR immunoreactivity in the (A) lateral septum, (B) medial preoptic area, (C) anterior hypothalamus, and (D) basolateral amygdala, ventral portion. Scale bar: 10 m.
Fig. 3. Mean (±SEM) percentage of AR+ cells in the (A) lateral septum (LS), (B) medial preoptic area (POA), (C) anterior hypothalamus (AH), and (D), basolateral amygdala, ventral portion (BLV) of subordinate (Sub) and breeder naked mole-rats. Number of animals per group is noted at the base of each bar. White bars represent females and black bars represent males. Asterisks indicate significant main effects of social status.
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218 Table 1 Androgen receptor expression in brain regions of the social decision-making network of rodents. Brain region
AR in mice?a
AR in naked mole-rats?a
Social control of AR? c
AH BLA (ventral portion) BNST Hippocampus LS (ventral portion) MeA NAc PAG mPOA Striatum VMH VP VTA
++ ++ ++++ +++ ++++ ++++ −/+ ++ ++++ − +++ + ++
++ ++ +++b −/+b +++ +++b −/+ + +++ − +++b + ++
? ? Yes ? ? Yes Yes ? Yes ? Yes ? Yes
AH = anterior hypothalamus; BLA = basolateral amygdala; BNST = bed nucleus of the stria terminalis; LS = lateral septum; MeA = medial amygdala; NAc = nucleus accumbens; PAG = periaqueductal gray; mPOA = medial preoptic area; VMH = ventromedial nucleus of the hypothalamus; VP = ventral pallidum; VTA = ventral tegmental area. a Intensity of immunoreactivity (IR): −, no IR detected; +, weak IR; ++, moderate IR; +++, strong IR; ++++, intense IR. Mouse immunoreactivity was assessed in tissue from gonadally intact adult male C57Bl6 mice processed alongside the naked molerat brains. b From Holmes et al. [10]. c All regions exhibiting moderate or stronger AR-IR exhibit social effects in naked mole-rats. A “yes” in this column indicates a social effect on AR has been reported in at least one other rodent; see text for references.
for each region by dividing the number of AR+ cells by the total number of cells (AR+ and AR− cell counts combined). Percentages of AR+ cells were analyzed using two-way ANOVAs with sex and status as independent variables. To ensure that percentage scores were normally distributed, ANOVAs were repeated following an arcsine transformation [12]. Transformation did not alter the pattern of significance for any brain region and these F and p values are reported below. Consistent with our previous report, we see significant effects of social status on AR immunoreactivity in several brain regions. Subordinates have a greater percentage of AR+ cells than breeders in the LS (F1,13 = 22.64; p < 0.001; Fig. 3A); mPOA (F1,12 = 6.71; p = 0.024; Fig. 3B); AH (F1,12 = 6.70; p = 0.023; Fig. 3C); and BLV (F1,12 = 12.85; p = 0.004; Fig. 3D). No significant effects of sex were detected and no sex by status interactions were seen in any region. There was, however, a trend for males to have more AR than females in the mPOA (F1,12 = 3.85; p = 0.07). In addition, while the sex by status interaction in the mPOA was not significant, likely due to the small sample size, pairwise comparisons reveal a significant effect of status for females (p = 0.03) but not males (p = 0.40) in this region (Fig. 3B). Collectively, our data confirm that, for the most part, the distribution of AR immunoreactivity in naked mole-rats is similar to that seen in other rodents. Those regions of the SDM network that consistently express AR in rodents, including the naked molerat, are the AH, MeA, BNST, LS, PAG, POA, ventromedial nucleus of the hypothalamus (VMH), and VTA (Table 1) [[10,13–16], current study]. Furthermore, the localization of AR to the ventral portion of the BLA as seen here is also found in the mouse (M.M.H., unpublished). Also consistent are several areas that express few to no AR+ cells: as in mice, rats, and hamsters [13–15], we saw no labeling in the dorsal striatum/caudate nucleus and very sparse labeling in the NAc and ventral pallidum of naked mole-rats. Conversely, we found very little AR immunoreactivity in the hippocampus [10], which contains dense AR in other rodents [13]. Our current finding of a greater percentage of AR+ cells in subordinate naked mole-rats, regardless of sex, is consistent with previous work indicating that social status is a better predictor
217
than sex of neural morphology or neurochemistry in this species (reviewed in [7]). Reduced AR in breeders compared to subordinates is somewhat counterintuitive but is consistent with what was seen in the brain regions examined previously [10]. Androgens are capable of both up- and down-regulating AR expression in other species (e.g., [17]). However, the decreased AR+ in breeders of both sexes is unlikely to be explained simply by group differences in circulating testosterone levels because both breeder and subordinate males have higher testosterone levels than breeder or subordinate females [18]. Although the ability of hormones to modulate AR expression has been well established in diverse rodent species via castration and testosterone replacement paradigms, little is known about how more natural stimuli, including social interactions, regulate expression of this protein in mammals. Androgen receptors are critically involved in the expression of mammalian social behaviors both developmentally and in adulthood (e.g., [19]). In turn, social interactions dynamically influence circulating androgens, although the work done to date clearly demonstrates sex-, species-, brain region- and manipulation-specific effects (Table 1). For example, sex behavior decreases AR in the NAc, mPOA and VMH of male rats [20] whereas winning a territorial dispute increases AR in the BNST, NAc, and VTA of male California mice [21] and chemosensory stimuli from female conspecifics increases AR in the MeA of male hamsters [22]. Social stress early in life increases AR in the mPOA and hippocampus of female guinea pigs but decreases AR in the mPOA of male guinea pigs [23,24]. We do not know whether the difference in AR between subordinate and breeder naked mole-rats is a contributing factor to eusocial organization or the result of it. It is possible, for example, that AR must be down-regulated to allow for the changes in behavior seen as an animal assumes breeding status and may contribute to the relatively low levels of injurious aggressive behavior seen in stable colonies [6]. Alternatively, the behavioral interactions or endocrine changes associated with breeding may trigger a reduction in AR. Similarly, we noted some unusual features of the distribution of AR in naked mole-rats (e.g., the relative absence of AR in the hippocampus). This pattern of AR expression could contribute to, or be the result of, the unique social organization in this species, or may play no role at all in sociality. Understanding the distribution and social control of proteins involved in social behaviors, coupled with the recent cloning of the naked mole-rat genome [25], make it feasible to envision studies in which these questions are directly tested and expression of socially relevant genes, such as that for AR, are manipulated in this species.
Acknowledgements This work was funded by NSF grants 0344312 and 0642050 (NGF and BDG), and a NSERC Discovery Grant (MMH). Dr. Gail Prins generously provided the PG21 antibody and the AR21 and AR462 peptides.
References [1] O’Connell LA, Hofmann HA. The vertebrate mesolimbic reward system and social behavior network: a comparative synthesis. J Comp Neurol 2011;519:3599–639. [2] Newman SW. The medial extended amygdala in male reproductive behavior. A node in the mammalian social behavior network. Ann N Y Acad Sci 1999;877:242–57. [3] Jarvis JUM. Eusociality in a mammal: cooperative breeding in naked mole-rat colonies. Science 1981;212:571–3. [4] Lacey EA, Sherman PW. Social organization of naked mole-rat colonies: evidence for division of labor. In: Sherman PW, Jarvis JUM, Alexander RD, editors. The biology of the naked mole-rat. Princeton: Princeton University Press; 1991. p. 275–336.
218
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218
[5] Bennett NC, Faulkes CG. Social organisation in African mole-rats. In: Bennett NC, Faulkes CG, editors. African mole-rats: ecology and eusociality. Cambridge: Cambridge University Press; 2000. p. 87–125. [6] Lacey EA, Alexander RD, Braude SH, Sherman PW, Jarvis JUM. An ethogram for the naked mole-rat: nonvocal behaviors. In: Sherman PW, Jarvis JUM, Alexander RD, editors. The biology of the naked mole-rat. Princeton: Princeton University Press; 1991. p. 209–42. [7] Holmes MM, Goldman BD, Goldman SL, Seney ML, Forger NG. Neuroendocrinology and sexual differentiation in eusocial mammals. Front Neuroendocrinol 2009;30:519–33. [8] Holmes MM, Rosen GJ, Jordan CL, De Vries GJ, Goldman BD, Forger NG. Social control of brain morphology in a eusocial mammal. Proc Natl Acad Sci U S A 2007;104:10548–52. [9] Holmes MM, Seney ML, Goldman BD, Forger NG. Social and hormonal triggers of neural plasticity in naked mole-rats. Behav Brain Res 2011;218:234–9. [10] Holmes MM, Goldman BD, Forger NG. Social status and sex independently influence androgen receptor expression in the eusocial naked mole-rat brain. Horm Behav 2008;54:278–85. [11] Franklin KBJ, Paxinos G. The mouse brain in stereotaxic co-ordinates. 3rd ed. New York: Academic Press; 2008. [12] Ferguson GA, Takane Y. Statistical analysis in psychology and education. 6th ed. USA: McGraw-Hill, Inc.; 1989. [13] Simerly RB, Chang C, Muramatsu M, Swanson LW. Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J Comp Neurol 1990;294:76–95. [14] Clancy AN, Whitman C, Michael RP, Albers HE. Distribution of androgen receptor-like immunoreactivity in the brains of intact and castrated male hamsters. Brain Res Bull 1994;33:325–32. [15] Shah NM, Pisapia DJ, Maniatis S, Mendelsohn MM, Nemes A, Axel R. Visualizing sexual dimorphism in the brain. Neuron 2004;43:313–9. [16] Choate JVA, Resko JA. Androgen receptor immunoreactivity in intact and castrate guinea pig using antipeptide antibodies. Brain Res 1992;597:51–9.
[17] Handa RJ, Kerr JE, DonCarlos LL, McGivern RF, Hejna G. Hormonal regulation of androgen receptor messenger RNA in the medial preoptic area of the male rat. Brain Res Mol Brain Res 1996;39:57–67. [18] Zhou S, Holmes MM, Forger NG, Goldman BD, Lovern MB, Caraty A, et al. Socially regulated reproductive development: analysis of GnRH1 and kisspeptin neuronal systems in cooperatively breeding naked mole-rats (Heterocephalus glaber). J Comp Neurol 2013;521:3003–29, http://dx.doi.org/10.1002/cne.23327. [19] Bodo C, Rissman EF. The androgen receptor is selectively involved in organization of sexually dimorphic social behaviors in mice. Endocrinology 2008;149:4142–50. [20] Fernandez-Gausti A, Swaab D, Rodriguez-Manzo G. Sexual behavior reduces hypothalamic androgen receptor immunoreactivity. Psychoneuroendocrinology 2003;28:501–12. [21] Fuxjager MJ, Forbes-Lorman RM, Coss DJ, Auger CJ, Auger AP, Marler CA. Winning territorial disputes selectively enhances androgen sensitivity in neural pathways related to motivation and social aggression. Proc Natl Acad Sci U S A 2010;107:12393–8. [22] Blake CB, Meredith M. Change in number and activation of androgen receptorimmunoreactive cells in the medial amygdala in response to chemosensory input. Neuroscience 2011;190:228–38. [23] Kaiser S, Kruijver FP, Swaab DF, Sachser N. Early social stress in female guinea pigs induces a masculinization of adult behavior and corresponding changes in brain and neuroendocrine function. Behav Brain Res 2003;144: 199–210. [24] Kaiser S, Kruijver FP, Straub RH, Sachser N, Swaab DF. Early social stress in male Guinea-pigs changes social behaviour, and autonomic and neuroendocrine functions. J Neuroendocrinol 2003;15:761–9. [25] Kim EB, Fang X, Fushan AA, Huang Z, Lobanov AV, Han L, et al. Genome sequencing reveals insights into physiology and longevity of the naked mole rat. Nature 2011;479:223–7.
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Short communication
Androgen receptor distribution in the social decision-making network of eusocial naked mole-rats Melissa M. Holmes a,b,∗ , Spencer Van Mil a , Camila Bulkowski a , Sharry L. Goldman c , Bruce D. Goldman c , Nancy G. Forger d a
Department of Psychology, University of Toronto Mississauga, Mississauga, Ontario L5L 1C6, Canada Department of Cell and Systems Biology, University of Toronto, Toronto, Ontario M5S 3G5, Canada c Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT 06269, USA d Neuroscience Institute, Georgia State University, Atlanta, GA 30302, USA b
h i g h l i g h t s • • • •
Naked mole-rats are eusocial rodents, living in large social groups with strict hierarchies. Subordinate naked mole-rats have more androgen receptor protein in the social decision-making network than breeders. Naked mole-rats exhibit some unique features in neural androgen receptor distribution. Social status controls androgen receptor protein in the brain of eusocial rodents.
a r t i c l e
i n f o
Article history: Received 23 June 2013 Received in revised form 13 August 2013 Accepted 15 August 2013 Available online 21 August 2013 Keywords: Androgen receptor Anterior hypothalamus Lateral septum Naked mole-rat Preoptic area Social status
a b s t r a c t Naked mole-rats are highly social rodents that live in large groups and exhibit a strict reproductive and social hierarchy. Only a few animals in each colony breed; the remainder are non-reproductive and are socially subordinate to breeders. We have examined androgen receptor immunoreactive (AR+) cells in brain regions comprising the recently described social decision-making network in subordinate and breeder naked mole-rats of both sexes. We find that subordinates have a significantly higher percentage of AR+ cells in all brain regions expressing this protein. By contrast, there were no significant effects of sex and no sex-by-status interactions on the percentage of AR+ cells. Taken together with previous findings, the present data complete a systematic assessment of the distribution of AR protein in the social decision-making network of the eusocial mammalian brain and demonstrate a significant role for social status in the regulation of this protein throughout many nodes of this network. © 2013 Elsevier B.V. All rights reserved.
Social behaviors are broadly controlled by a complex neural network and, in turn, social interactions feed back to sculpt the brain throughout life. O’Connell and Hofmann [1] recently described those brain regions involved in the expression of motivated behaviors and their conservation across species, ultimately proposing the social decision-making (SDM) network. The SDM network essentially marries the social behavior network [2] and mesolimbic reward system to form an integrated network underlying the expression of adaptive motivated behaviors in vertebrates. It is clear that to better understand general mechanisms associated with the neural control of social interactions and concomitant
∗ Corresponding author at: Department of Psychology, University of Toronto Mississauga, 3359 Mississauga Road North, Mississauga, Ontario L5L 1C6, Canada. Tel.: +1 905 828 3956; fax: +1 905 569 4326. E-mail address: [email protected] (M.M. Holmes). 0166-4328/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbr.2013.08.025
socially controlled neural plasticity, we need a more systematic and comprehensive characterization of these brain regions in diverse species. Indeed, species that exhibit striking adaptations in their social interactions are particularly valuable for such comparisons as they better illuminate those features that are the exception versus those that are the rule. Naked mole-rats are a powerful animal model for understanding the reciprocal relationship between social interactions and brain plasticity. They are eusocial and display the most rigid social hierarchy and reproductive skew among mammals [3]. These animals live in large subterranean colonies of up to 300 animals and reproduction is restricted to a single breeding female and her 1–3 male consorts [4,5]. Breeders are socially dominant over all other members of the colony, which are non-reproductive subordinates [6]. We have previously shown various neural changes associated with transitions in social status in this species (reviewed in [7]). For example, the paraventricular nucleus, bed nucleus of the stria
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218
215
Fig. 1. Photomicrographs of AR immunoreactivity in the (A) dorsal striatum, (B) nucleus accumbens (NAc), (C) lateral septum (LS), (D) ventral pallidum (VP), (E) medial preoptic area (mPOA), (F) anterior hypothalamus (AH), (G) basolateral amygdala, ventral portion (BLV); inset is a higher magnification view of the BLV, (H) periaqueductal gray (PAG), and (I) ventral tegmental area (VTA) of a subordinate male naked mole-rat. Midline is on the right side of each image (except H). Other abbreviations: ac = anterior commissure; BLA = basolateral amygdala; ec = external capsule; ICj = Island of Calleja major; LA = lateral amygdala; LV = lateral ventricle; MS = medial septum; SN = substantia nigra; 3V = third ventricle. Scale bar at lower right: 100 m for A–F and 200 m for H and I. Scale bar in G: 200 m for larger view and 100 m for inset.
terminalis and medial amygdala, all brain regions important for reproduction in other species, are larger in breeder than subordinate naked mole-rats [8,9]. To begin to understand the putative endocrine mechanisms associated with changes in status and corresponding changes in brain morphology and social behavior, we previously evaluated the expression of androgen receptor (AR) protein in the same brain areas in which we saw morphological plasticity [10]. Social status significantly altered AR immunoreactivity in regions that increased in size in breeders, however, an inverse relationship was found whereby subordinates had more AR+ cells than breeders. The present report describes our efforts to more fully characterize the neuroendocrinology and neurobiology of these unique animals. Using the same tissue in our initial report [10], we have analyzed AR immunoreactivity in the remaining regions of the SDM network of breeding and subordinate naked mole-rats of both sexes. Brains from 5 breeding females, 5 breeding males, 4 subordinate females, and 4 subordinate males were collected; all animals were gonadally intact. Brains were removed, immersion fixed in 5% acrolein for 4 h, transferred to 30% sucrose in 0.1 M phosphate buffer for cryoprotection and sectioned into 4 series in the coronal plane using a sliding microtome. One series was processed for AR immunohistochemistry using the polyclonal AR antiserum, PG21, and counter-stained with methyl green, as fully described in [10]. In addition, one series from each of three gonadally intact adult male C57BL/6 mice was processed concurrently as a positive control and for between-species comparisons. Qualitative analyses indicated that not all brain regions involved in the SDM network contained high levels of AR+ cells. The dorsal
striatum, nucleus accumbens (NAc) and ventral pallidum (VP) had little to no AR immunoreactivity (Fig. 1A, B and D). Similarly, while there was consistent albeit sparse AR in the lateral amygdala, the basolateral amygdala (BLA) was unlabeled with the exception of the ventral portion (Fig. 1G). AR immunoreactivity was consistently present in the periaqueductal gray (PAG) and ventral tegmental area (VTA) though quantification of these regions was not possible due to insufficient tissue for all animals in all groups. Quantitative stereological analyses of the percentage of cells positive for AR immunoreactivity in the lateral septum (LS; equivalent to plates 25–30 in the mouse brain atlas of [11]; Figs. 1C and 2A), medial preoptic area (mPOA; plates 30–33; Figs. 1E and 2B), anterior hypothalamus (AH; plates 35–38; Figs. 1F and 2C), and basolateral amygdala, ventral portion (BLV; plates 40–42; Figs. 1G and 2D) were performed using StereoIoger software (Stereology Resource Center, Inc.). Outlines of each region were traced in each section, and unbiased estimates of the number of darkly labeled, lightly labeled, and unlabeled cells were obtained using the optical disector method. Counting frames were 20 m × 20 m and frame size was held constant across animals. Sampling grid size varied from 60 m × 60 m to 200 m × 200 m depending on brain region, and was held constant for all animals. All brain regions were analyzed bilaterally unless tearing or tissue artifacts were present in the region of interest. Because the pattern of results was the same whether only darkly labeled or all labeled cells were considered, we have combined dark plus light cell counts in the analyses presented here, collectively referred to as AR+. To account for possible group differences in total cell number in some brain regions [8], the percentage of AR+ cells was calculated
216
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218
Fig. 2. High magnification photomicrographs of AR immunoreactivity in the (A) lateral septum, (B) medial preoptic area, (C) anterior hypothalamus, and (D) basolateral amygdala, ventral portion. Scale bar: 10 m.
Fig. 3. Mean (±SEM) percentage of AR+ cells in the (A) lateral septum (LS), (B) medial preoptic area (POA), (C) anterior hypothalamus (AH), and (D), basolateral amygdala, ventral portion (BLV) of subordinate (Sub) and breeder naked mole-rats. Number of animals per group is noted at the base of each bar. White bars represent females and black bars represent males. Asterisks indicate significant main effects of social status.
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218 Table 1 Androgen receptor expression in brain regions of the social decision-making network of rodents. Brain region
AR in mice?a
AR in naked mole-rats?a
Social control of AR? c
AH BLA (ventral portion) BNST Hippocampus LS (ventral portion) MeA NAc PAG mPOA Striatum VMH VP VTA
++ ++ ++++ +++ ++++ ++++ −/+ ++ ++++ − +++ + ++
++ ++ +++b −/+b +++ +++b −/+ + +++ − +++b + ++
? ? Yes ? ? Yes Yes ? Yes ? Yes ? Yes
AH = anterior hypothalamus; BLA = basolateral amygdala; BNST = bed nucleus of the stria terminalis; LS = lateral septum; MeA = medial amygdala; NAc = nucleus accumbens; PAG = periaqueductal gray; mPOA = medial preoptic area; VMH = ventromedial nucleus of the hypothalamus; VP = ventral pallidum; VTA = ventral tegmental area. a Intensity of immunoreactivity (IR): −, no IR detected; +, weak IR; ++, moderate IR; +++, strong IR; ++++, intense IR. Mouse immunoreactivity was assessed in tissue from gonadally intact adult male C57Bl6 mice processed alongside the naked molerat brains. b From Holmes et al. [10]. c All regions exhibiting moderate or stronger AR-IR exhibit social effects in naked mole-rats. A “yes” in this column indicates a social effect on AR has been reported in at least one other rodent; see text for references.
for each region by dividing the number of AR+ cells by the total number of cells (AR+ and AR− cell counts combined). Percentages of AR+ cells were analyzed using two-way ANOVAs with sex and status as independent variables. To ensure that percentage scores were normally distributed, ANOVAs were repeated following an arcsine transformation [12]. Transformation did not alter the pattern of significance for any brain region and these F and p values are reported below. Consistent with our previous report, we see significant effects of social status on AR immunoreactivity in several brain regions. Subordinates have a greater percentage of AR+ cells than breeders in the LS (F1,13 = 22.64; p < 0.001; Fig. 3A); mPOA (F1,12 = 6.71; p = 0.024; Fig. 3B); AH (F1,12 = 6.70; p = 0.023; Fig. 3C); and BLV (F1,12 = 12.85; p = 0.004; Fig. 3D). No significant effects of sex were detected and no sex by status interactions were seen in any region. There was, however, a trend for males to have more AR than females in the mPOA (F1,12 = 3.85; p = 0.07). In addition, while the sex by status interaction in the mPOA was not significant, likely due to the small sample size, pairwise comparisons reveal a significant effect of status for females (p = 0.03) but not males (p = 0.40) in this region (Fig. 3B). Collectively, our data confirm that, for the most part, the distribution of AR immunoreactivity in naked mole-rats is similar to that seen in other rodents. Those regions of the SDM network that consistently express AR in rodents, including the naked molerat, are the AH, MeA, BNST, LS, PAG, POA, ventromedial nucleus of the hypothalamus (VMH), and VTA (Table 1) [[10,13–16], current study]. Furthermore, the localization of AR to the ventral portion of the BLA as seen here is also found in the mouse (M.M.H., unpublished). Also consistent are several areas that express few to no AR+ cells: as in mice, rats, and hamsters [13–15], we saw no labeling in the dorsal striatum/caudate nucleus and very sparse labeling in the NAc and ventral pallidum of naked mole-rats. Conversely, we found very little AR immunoreactivity in the hippocampus [10], which contains dense AR in other rodents [13]. Our current finding of a greater percentage of AR+ cells in subordinate naked mole-rats, regardless of sex, is consistent with previous work indicating that social status is a better predictor
217
than sex of neural morphology or neurochemistry in this species (reviewed in [7]). Reduced AR in breeders compared to subordinates is somewhat counterintuitive but is consistent with what was seen in the brain regions examined previously [10]. Androgens are capable of both up- and down-regulating AR expression in other species (e.g., [17]). However, the decreased AR+ in breeders of both sexes is unlikely to be explained simply by group differences in circulating testosterone levels because both breeder and subordinate males have higher testosterone levels than breeder or subordinate females [18]. Although the ability of hormones to modulate AR expression has been well established in diverse rodent species via castration and testosterone replacement paradigms, little is known about how more natural stimuli, including social interactions, regulate expression of this protein in mammals. Androgen receptors are critically involved in the expression of mammalian social behaviors both developmentally and in adulthood (e.g., [19]). In turn, social interactions dynamically influence circulating androgens, although the work done to date clearly demonstrates sex-, species-, brain region- and manipulation-specific effects (Table 1). For example, sex behavior decreases AR in the NAc, mPOA and VMH of male rats [20] whereas winning a territorial dispute increases AR in the BNST, NAc, and VTA of male California mice [21] and chemosensory stimuli from female conspecifics increases AR in the MeA of male hamsters [22]. Social stress early in life increases AR in the mPOA and hippocampus of female guinea pigs but decreases AR in the mPOA of male guinea pigs [23,24]. We do not know whether the difference in AR between subordinate and breeder naked mole-rats is a contributing factor to eusocial organization or the result of it. It is possible, for example, that AR must be down-regulated to allow for the changes in behavior seen as an animal assumes breeding status and may contribute to the relatively low levels of injurious aggressive behavior seen in stable colonies [6]. Alternatively, the behavioral interactions or endocrine changes associated with breeding may trigger a reduction in AR. Similarly, we noted some unusual features of the distribution of AR in naked mole-rats (e.g., the relative absence of AR in the hippocampus). This pattern of AR expression could contribute to, or be the result of, the unique social organization in this species, or may play no role at all in sociality. Understanding the distribution and social control of proteins involved in social behaviors, coupled with the recent cloning of the naked mole-rat genome [25], make it feasible to envision studies in which these questions are directly tested and expression of socially relevant genes, such as that for AR, are manipulated in this species.
Acknowledgements This work was funded by NSF grants 0344312 and 0642050 (NGF and BDG), and a NSERC Discovery Grant (MMH). Dr. Gail Prins generously provided the PG21 antibody and the AR21 and AR462 peptides.
References [1] O’Connell LA, Hofmann HA. The vertebrate mesolimbic reward system and social behavior network: a comparative synthesis. J Comp Neurol 2011;519:3599–639. [2] Newman SW. The medial extended amygdala in male reproductive behavior. A node in the mammalian social behavior network. Ann N Y Acad Sci 1999;877:242–57. [3] Jarvis JUM. Eusociality in a mammal: cooperative breeding in naked mole-rat colonies. Science 1981;212:571–3. [4] Lacey EA, Sherman PW. Social organization of naked mole-rat colonies: evidence for division of labor. In: Sherman PW, Jarvis JUM, Alexander RD, editors. The biology of the naked mole-rat. Princeton: Princeton University Press; 1991. p. 275–336.
218
M.M. Holmes et al. / Behavioural Brain Research 256 (2013) 214–218
[5] Bennett NC, Faulkes CG. Social organisation in African mole-rats. In: Bennett NC, Faulkes CG, editors. African mole-rats: ecology and eusociality. Cambridge: Cambridge University Press; 2000. p. 87–125. [6] Lacey EA, Alexander RD, Braude SH, Sherman PW, Jarvis JUM. An ethogram for the naked mole-rat: nonvocal behaviors. In: Sherman PW, Jarvis JUM, Alexander RD, editors. The biology of the naked mole-rat. Princeton: Princeton University Press; 1991. p. 209–42. [7] Holmes MM, Goldman BD, Goldman SL, Seney ML, Forger NG. Neuroendocrinology and sexual differentiation in eusocial mammals. Front Neuroendocrinol 2009;30:519–33. [8] Holmes MM, Rosen GJ, Jordan CL, De Vries GJ, Goldman BD, Forger NG. Social control of brain morphology in a eusocial mammal. Proc Natl Acad Sci U S A 2007;104:10548–52. [9] Holmes MM, Seney ML, Goldman BD, Forger NG. Social and hormonal triggers of neural plasticity in naked mole-rats. Behav Brain Res 2011;218:234–9. [10] Holmes MM, Goldman BD, Forger NG. Social status and sex independently influence androgen receptor expression in the eusocial naked mole-rat brain. Horm Behav 2008;54:278–85. [11] Franklin KBJ, Paxinos G. The mouse brain in stereotaxic co-ordinates. 3rd ed. New York: Academic Press; 2008. [12] Ferguson GA, Takane Y. Statistical analysis in psychology and education. 6th ed. USA: McGraw-Hill, Inc.; 1989. [13] Simerly RB, Chang C, Muramatsu M, Swanson LW. Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J Comp Neurol 1990;294:76–95. [14] Clancy AN, Whitman C, Michael RP, Albers HE. Distribution of androgen receptor-like immunoreactivity in the brains of intact and castrated male hamsters. Brain Res Bull 1994;33:325–32. [15] Shah NM, Pisapia DJ, Maniatis S, Mendelsohn MM, Nemes A, Axel R. Visualizing sexual dimorphism in the brain. Neuron 2004;43:313–9. [16] Choate JVA, Resko JA. Androgen receptor immunoreactivity in intact and castrate guinea pig using antipeptide antibodies. Brain Res 1992;597:51–9.
[17] Handa RJ, Kerr JE, DonCarlos LL, McGivern RF, Hejna G. Hormonal regulation of androgen receptor messenger RNA in the medial preoptic area of the male rat. Brain Res Mol Brain Res 1996;39:57–67. [18] Zhou S, Holmes MM, Forger NG, Goldman BD, Lovern MB, Caraty A, et al. Socially regulated reproductive development: analysis of GnRH1 and kisspeptin neuronal systems in cooperatively breeding naked mole-rats (Heterocephalus glaber). J Comp Neurol 2013;521:3003–29, http://dx.doi.org/10.1002/cne.23327. [19] Bodo C, Rissman EF. The androgen receptor is selectively involved in organization of sexually dimorphic social behaviors in mice. Endocrinology 2008;149:4142–50. [20] Fernandez-Gausti A, Swaab D, Rodriguez-Manzo G. Sexual behavior reduces hypothalamic androgen receptor immunoreactivity. Psychoneuroendocrinology 2003;28:501–12. [21] Fuxjager MJ, Forbes-Lorman RM, Coss DJ, Auger CJ, Auger AP, Marler CA. Winning territorial disputes selectively enhances androgen sensitivity in neural pathways related to motivation and social aggression. Proc Natl Acad Sci U S A 2010;107:12393–8. [22] Blake CB, Meredith M. Change in number and activation of androgen receptorimmunoreactive cells in the medial amygdala in response to chemosensory input. Neuroscience 2011;190:228–38. [23] Kaiser S, Kruijver FP, Swaab DF, Sachser N. Early social stress in female guinea pigs induces a masculinization of adult behavior and corresponding changes in brain and neuroendocrine function. Behav Brain Res 2003;144: 199–210. [24] Kaiser S, Kruijver FP, Straub RH, Sachser N, Swaab DF. Early social stress in male Guinea-pigs changes social behaviour, and autonomic and neuroendocrine functions. J Neuroendocrinol 2003;15:761–9. [25] Kim EB, Fang X, Fushan AA, Huang Z, Lobanov AV, Han L, et al. Genome sequencing reveals insights into physiology and longevity of the naked mole rat. Nature 2011;479:223–7.