In the United States, methamphetamine (MA) was reported to be abused by approximately 439 000 people in 2011, 133 000 of which were new users (SAMHSA 2012). MA abuse results in devastating consequences for those who use the drug as well as for the society as a whole. MA is estimated to cost the United States approximately $23.4 billion annually (Nicosia et al. 2009). There are sex differences in the use and subsequent effects of MA. Women begin using MA at an earlier age than men (Dluzen and Liu 2008) and have a shorter duration from first use to regular use (Brecht et al. 2004). Furthermore, women are more dependent on and committed to MA (SAMHSA 2010), whereas men tend to use other drugs when MA is unavailable to them (Brecht et al. 2004). Rodent models also indicate that females show greater MA seeking and reinstatement of MA use following a period of withdrawal (Roth and Carroll 2004; Reichel et al. 2012).
Chronic hypothalamic–pituitary–adrenal (HPA) axis activation can lead to lasting alterations in HPA axis function (Herman et al. 1995; Gómez et al. 1996). Dysregulation of HPA axis function is strongly linked to changes in emotional (e.g., depression, anxiety) as well as addiction-related behaviors including MA seeking (Shoener et al. 2006; Fernández-Guasti et al. 2012; Nawata et al. 2012). Both emotional and addictive behaviors have been reported to be sexually dimorphic in humans and rodents (Palanza 2001; Brecht et al. 2004; Leach et al. 2008; Reichel et al. 2012) although the contribution of MA-induced activation of the HPA axis is unknown. In rodents, MA exposure in the developing brain has been shown to acutely and chronically activate the HPA axis, which results in increased plasma levels of adrenocorticotropic hormone and corticosterone (Williams et al. 2000; Acevedo et al. 2008). In the developing brain, the effects of MA exposure on HPA axis activation are more pronounced in female than male mice (Acevedo et al. 2008). In contrast to the developing brain, little is known about potential sex differences in HPA axis activation following MA exposure in adulthood which may contribute to sex differences in MA effects on the brain and behavior.
Activation of the HPA axis involves a complex brain circuitry. This circuitry includes areas of the cortex, amygdala, hippocampus, and hypothalamus, which project to the paraventricular nucleus of the hypothalamus (PVN), the central HPA axis regulatory nucleus in the brain. Within the PVN, several cell phenotypes are critical for the regulation of the HPA axis, including arginine vasopressin (AVP), corticotropic-releasing hormone (CRH), and glucocorticoid receptor (GR)-expressing cells. AVP-expressing cells within the PVN are divided into two functionally distinct subpopulations, magnocellular and parvocellular neurons. These neurons increase transcription of AVP following HPA axis activation (Angulo et al. 1991). Magnocellular AVP-expressing cells project to the posterior pituitary and are involved in the regulation of water and electrolyte homeostasis (Skorecki et al. 1992) as well as cardiovascular function (Robertson 1977). AVP synthesized from the parvocellular population acts in conjunction with CRH to stimulate secretion of adrenocorticotropic hormone from the pituitary gland, and subsequently, glucocorticoids (e.g., corticosterone) from the adrenal gland (Plotsky 1987). Following activation of the HPA axis, glucocorticoids activate GRs located throughout the brain and pituitary. Within the brain, activation of GRs in the cortex, hippocampus, amygdala, hypothalamus, and other regions are involved in terminating the HPA axis through negative feedback mechanisms (Herman et al. 2012). GR activation also contributes to stress-related memory formation (de Kloet et al. 2005) and has been linked to addiction-related behaviors (Wang et al. 2008; Ambroggi et al. 2009). In contrast, glucocorticoids primarily act via mineralocorticoid receptors in the brain to regulate glucocorticoid levels during basal conditions (Reul and de Kloet 1985; Dallman et al. 1987).
Activation of the HPA axis and corticosterone release have also been linked to addiction-related behaviors. Chronic elevations in plasma corticosterone levels lead to alterations in HPA axis-associated genes and proteins within the PVN, amygdala, and hippocampus (Kovács and Mezey 1987; Reul et al. 1987; Herman et al. 1989; Sawchenko et al. 1993). Disruptions in these genes, particularly CRH, have been shown to affect MA addiction-related behaviors including MA seeking following withdrawal (Nawata et al. 2012).
In conditions of stress in adulthood, female rodents show greater increases in plasma corticosterone levels and different activation patterns in HPA axis-associated brain regions compared to males (Handa et al. 1994a; Figueiredo et al. 2002; Zuloaga et al. 2011). As little is known about potential sex differences in HPA axis activation following MA exposure in adulthood; in this study, we investigated and compared MA-induced plasma corticosterone responses in male and female mice. MA also increases body temperature for a prolonged period of time (Kuo et al. 2003) which coincides with activation of the HPA axis, although the specific relationship between the two is not well understood. Therefore, we also examined core body temperature responses to MA at the same sampling time points for plasma corticosterone to examine the temporal relationship between these measures. In addition, we explored immediate early gene activation in regions associated with the HPA axis to assess potential MA and sex differences in activation of brain nuclei. Finally, we analyzed the colocalization patterns of the immediate early gene c-Fos with AVP and GR to identify the specific cell phenotypes activated by MA. These phenotypes were chosen based on previous studies that indicate MA can modify their expression (Zuloaga et al., 2013, Lowy 1990). We hypothesized that females would show elevated corticosterone responses to MA and an altered activation of HPA axis-associated brain regions compared to males.
- Top of page
- Materials and methods
- Acknowledgements and conflict of interest disclosure
The results of this study indicate that MA causes sex-dependent activation of the HPA axis with females showing a more prolonged elevation in plasma corticosterone compared to males. In females, the plasma corticosterone response following MA exposure peaked at 70 min post injection and subsequently decreased, but remained elevated above baseline levels at 120 min. In males, MA induced an earlier peak plasma corticosterone response at 30 min with a faster decline by 70 min and levels similar to baseline at 120 min. This sex difference in HPA axis activation with greater plasma corticosterone responses in females than males is similar to that observed in adult rodents following novelty and restraint stress (Handa et al. 1994a, 2009; Zuloaga et al. 2011), as well as administration of pharmacological agents such as the SSRI citalopram (Goel et al. 2011). Furthermore, the pattern of sex difference in HPA axis activation seen in the current study is similar to that seen following chronic exposure to MA from post-natal days 11–20 in the developing mouse brain (Acevedo et al. 2008). Saline injection alone induced a smaller rise in plasma corticosterone levels which peaked in male and female mice at 30 min and declined thereafter. This finding is consistent with previous studies indicating that stress from injection itself increases plasma corticosterone levels in mice (Goel et al. 2011).
The higher plasma corticosterone responses to MA in females than males may be regulated by sex differences in endogenous androgen and estrogen levels and differential activation of androgen and estrogen receptors. Gonadectomy of male rodents increases, while treatment with testosterone or dihydrotestosterone decreases, the plasma corticosterone response to stress (Handa et al. 1994b). Opposite effects are found in females, in which gonadectomy decreases and treatment with estradiol increases, plasma corticosterone responses to stress (Handa et al. 2009). These sex-dependent effects are mediated via estrogen receptor alpha stimulatory, and androgen receptor and estrogen receptor beta suppressive effects on the HPA axis (Zuloaga et al. 2008; Handa et al. 2009; Weiser and Handa 2009). Whether MA acts through similar hormone and receptor pathways to regulate sex differences in corticosterone release has yet to be determined.
Unlike plasma corticosterone responses, no sex differences were found in the MA-induced rise in body temperature. This temperature rise replicated a previous study which reported comparable rises in body temperature in male and female mice following MA administration (Kuo et al. 2003). Saline injection itself also resulted in a rise in body temperature, likely caused by stress-induced hyperthermia (McGivern et al. 2009). These data are comparable to previous reports, in which stress induced by injections resulted in temporally similar elevations and declines in plasma corticosterone levels and body temperature (Veening et al. 2004). However, MA and saline injections resulted in opposite correlations between plasma corticosterone levels and body temperature. There was a negative correlation between plasma corticosterone levels and body temperature in MA-treated mice. This negative correlation seemed driven by the maintained elevation in body temperature occurring in opposition to declining plasma corticosterone levels at later time points. This finding therefore indicates that there are distinct mechanisms that regulate HPA axis activation and hyperthermia following MA exposure. Activation of both independent and overlapping brain circuitry contribute to regulation of these functions following stress (Veening et al. 2004; Busnardo et al. 2010). Specifically, the PVN appears to be critical to the regulation of both functions (Veening et al. 2004; Busnardo et al. 2010). Other regions including the dorsomedial hypothalamus, anterodorsal preoptic area, and periaqueductal gray may be more pertinent to control of body temperature changes that parallel activation of the HPA axis (Veening et al. 2004). However, specific effects of MA on neural control of body temperature and the HPA axis are less understood.
Sex differences in MA activation of the HPA axis may be related to increased MA seeking in females. The corticosterone response to MA and MA seeking are both greater in female than male rodents (Roth and Carroll 2004; Reichel et al. 2012). One possibility is that the prolonged release of corticosterone in females may lead to greater alterations in HPA axis-associated genes and proteins. In turn this may contribute to alterations in MA seeking and reinstatement following withdrawal since disruptions in these genes, particularly CRH, have been linked to MA addiction-related behaviors (Nawata et al. 2012). Further studies are needed to test this hypothesis and determine the molecular pathways involved in the bidirectional relationship between effects of MA on HPA axis-associated proteins and effects of alterations in the levels and/or activational states of these proteins on MA addiction.
This increased HPA axis reactivity to MA in female mice is paralleled by a decrease in c-Fos-positive cells in two distinct brain regions, the CA3 region of the hippocampus and cingulate cortex. A sex difference in c-Fos induction in the CA3 area with greater activation in males than females has been reported following restraint stress (Figueiredo et al. 2002) and following administration of the SSRI citalopram (Goel et al. 2011), which also activates the HPA axis (Jensen et al. 1999). In both studies, decreased plasma corticosterone responses paralleled increased c-Fos expression in the CA3 region. The CA3 hippocampal region plays a key role in negative feedback functions of the HPA axis (Silverman et al. 1981). Therefore, activation of cells in this region may contribute to reducing the HPA axis response following its initial activation. In support of this notion, electrical stimulation of the dorsal CA3 region reduces plasma corticosterone levels (Dunn and Orr 1984). MA also increased c-Fos immunoreactivity within the CA1 hippocampal region, although not in a sex-dependent fashion, and failed to alter expression in the dentate gyrus. This suggests that sex differences in hippocampus-associated negative feedback of the HPA axis may be conferred primarily through the hippocampal CA3 area.
MA exposure also resulted in a sexually dimorphic c-Fos response in the cingulate cortex, again with males showing a greater activation. Sex differences in c-Fos activation in this brain region have previously been reported following stress, in which males also show greater responses than females in a similar division of the cingulate cortex (Figueiredo et al. 2002). Like the CA3 area, the cingulate is part of the neural circuitry that regulates inhibition of HPA axis function via multisynaptic projections to the PVN (Silverman et al. 1981; Diorio et al.1993). Reduced GR/Fos dual labeling in both the CA3 and cingulate cortex further suggest a mechanism for decreased negative feedback since activation of GR in hippocampal and cortical regions is critical to this function (Furay et al. 2008). Sex differences in neural activation within the cingulate and CA3 may also indicate differential effects of MA on memory functions and stress-related behaviors that are linked these brain regions (Handelmann and Olton 1981; Hunsaker et al. 2009; Wang et al. 2010). CA3 has also been linked to withdrawal-like behaviors and may potentially contribute to sex differences in this behavior in rodents (Isaacson and Lanthorn 1981). However, the specific role of CA3 in MA withdrawal is currently unknown.
MA administration also induced c-Fos expression in several other HPA axis-associated brain regions including the PVN, CEA, BNST, and PVT. In the PVN and CEA, two regions central to HPA axis activation, there were large inductions of c-Fos-positive cells following MA which parallel elevations in plasma corticosterone levels. However, no sex differences were found in MA-induced c-Fos-positive cells within these brain regions, although the plasma corticosterone response was greater in females. This finding is similar to reports in rats and mice that also demonstrate that c-Fos expression in the PVN and CEA does not differ although plasma corticosterone levels are elevated in females (Figueiredo et al. 2002; Campbell et al. 2003; Goel et al. 2011). Specifically, sex differences in corticosterone response (females > males) do not correlate with elevated c-Fos expression in females in these studies. It is possible that c-Fos expression in the PVN and CEA may be sexually dimorphic at different time points, either earlier or later, and involve either more or less activation of the same number of specific cells, or recruitment of more or less cells. However, in the current study we assessed only one specific time point at which previous studies have indicated extensive brain c-Fos levels following activation of the HPA axis (Giardino et al. 2011; Zuloaga et al. 2011). Further studies are needed to assess potential sex differences in the temporal dynamics of c-Fos following MA exposure.
The PVT and BNST also showed large inductions of c-Fos following MA injection with no sex differences. The PVT is proposed to be a key component of the brain circuitry that regulates habituation and facilitation of HPA axis function following stress (Bhatnagar and Dallman 1998; Bhatnagar et al. 2002), although this role has been debated (Fernandes et al. 2002). Given the high level of brain activation in the PVT following MA, it is possible that the PVT regulates similar HPA axis functions in the context of MA. The PVT projects to both the PVN and CEA and may therefore mediate HPA axis functions via these connections (Sawchenko and Swanson 1983; Turner and Herkenham 1991). The BNST has been demonstrated to play an inhibitory role on the HPA axis via GABAergic projections to the PVN (Choi et al. 2007). Therefore, activation of cells within this region may confer negative feedback regulation of the HPA axis. Robust neural activation of these brain regions (PVN, CEA, BNST, PVT) may also mediate MA effects on stress-related behaviors associated with these brain regions, particularly those associated with anxiety and depression (Salomé et al. 2004; Ventura-Silva et al. 2013). Long-term consequences of MA on these brain regions and their effects on stress and addiction-associated behaviors remain to be determined.
In an attempt to identify the phenotype of cells activated within the PVN and CEA, we performed dual-label immunohistochemistry to determine colocalization of c-Fos-positive cells with two proteins associated with the HPA axis, AVP and glucocorticoid receptor. Our results indicate that acute MA exposure results in a selective activation of neurons expressing GR with a lesser activation of AVP containing cells. This extensive activation of GR-positive cells in the PVN and CEA following MA exposure is greater than that following injection stress both in terms of total number of c-Fos/GR-positive cells and in the percentage of c-Fos-positive cells coexpressing GR. The percentage of c-Fos/GR-positive cells following acute MA injection reported here is also greater than that reported in rats subjected to restraint (Zavala et al. 2011), indicating that MA activation of GR-positive cells may be even greater than that observed following stress. However, species differences in HPA axis function may also contribute to this difference. Activation of GR has been linked to addiction-related behaviors (Wang et al. 2008; Ambroggi et al. 2009) although little is known about the specific involvement of the PVN and CEA in this process. In this study, we demonstrate that GR-positive cells in the PVN and CEA are specifically responsive to MA and may therefore be targets for MA-induced behavioral alterations. In the current study, we focused on GR rather than MR given their more prominent role in regulating HPA axis negative feedback in response to high levels of corticosterone (Reul and de Kloet 1985). However, the balance of GR/MR activation and expression in the brain is critical to HPA axis regulation (De Kloet and Derijk, 2004). Therefore, it is likely that MR also contributes to the corticosterone response following MA.
A smaller number and percentage of c-Fos-positive cells also expressed AVP, suggesting a lesser role for this neuropeptide in acute MA-induced activation of the HPA axis. It is possible that AVP may play a greater role in chronic activation of the HPA axis following repeated MA exposure. AVP has been reported to be involved in the adaptation of the HPA axis to chronic stress (Volpi et al. 2004). CRH-, rather than AVP-, expressing neurons in the PVN may be more specifically involved in MA activation of the HPA axis. Unfortunately, labeling for CRH in the mouse PVN is difficult to detect using available primary antisera, therefore studies utilizing alternative methods for visualizing CRH (eg. CRH-GFP models) are warranted to explore activation of these cells following MA exposure. C-Fos colocalization patterns with both AVP and GR did not differ between males and females suggesting that differential activation of these phenotypes in the PVN and CEA are not involved in sex differences in corticosterone response to MA. However, it remains possible that these cells may be activated for varying lengths of time in males compared to females thus resulting in a differential corticosterone response. Alternatively, a greater activation of CRH-positive cells in the PVN female mice may also contribute to their elevated corticosterone response. There is also some overlap in the distribution of GR and AVP-ir cells within the PVN (Uht et al., 1988; de Souza and Franci, 2010) so it is possible that activation of cells expressing both phenotypes may differ in response to MA. Further studies are warranted to test these possibilities.
In summary, the results of this study indicate that adult female mice show a prolonged corticosterone response to MA compared to males. Furthermore, we identified several HPA axis-associated regions that are activated following MA exposure. Two of these regions that have been associated with negative feedback regulation of the HPA axis (CA3 and cingulate) display a greater number of c-Fos and c-Fos/GR-positive cells after MA exposure in males than females. This suggests that these regions may contribute to sex differences in MA-induced corticosterone release. Finally, we report that MA injection results in extensive activation of GR-positive cells in the PVN and CEA with a lesser activation of AVP-positive cells in the PVN. Therefore, these cell phenotypes, particularly GR, appear to regulate effects of MA on the HPA axis and may therefore be potential targets for modulating MA effects on HPA axis function. These findings provide insight into anatomical regions and cell phenotypes involved in MA activation of the HPA axis and may also contribute to increased understanding of mechanisms involved in sex differences in MA use and the consequences of MA use.