Sex differences in stress-related disorders such as anxiety and depression are robust in humans with women showing approximately double the incidence of these disorders throughout the lifespan when compared to men (Gater et al., 1998; Kessler et al., 1994; Weissman et al., 1997) Sex differences in subclinical symptoms of anxiety and depression are also reported with women again showing a greater prevalence (Gater et al., 1998; Hankin, 2009; Kessler et al., 1994; Nolen-Hoeksema et al., 1999; Weissman et al., 1997). For major depression diagnoses and symptoms, the sex difference appears and peaks during mid-adolescence which is a period that coincides with puberty and increasing levels of sex hormones (Salk et al., 2017). Furthermore, a peak for sex differences in depression occurs during menopause and coincides with a decline in estrogens (Kundakovic and Rocks, 2022). These reports suggest a potential role for sex hormones in the regulation of this disorder. These sex differences in stress-related disorders may, in part, be regulated by sex differences in regulation of the hypothalamic pituitary adrenal (HPA) axis which have also been reported in humans (Bangasser and Valentino, 2014; Gomez et al., 2004; Handa et al., 2022; Kirschbaum et al., 1999; Kudielka and Kirschbaum, 2005; Seeman et al., 2001; Uhart et al., 2006). The HPA axis is a system that involves release of corticotropin releasing factor from the hypothalamus, which stimulates secretion of adrenocorticotropin releasing hormone (ACTH) from the pituitary, which then subsequently stimulates secretion of glucocorticoids (e.g. cortisol, corticosterone) from the adrenal cortex (Herman et al., 2016). Upon release, glucocorticoids travel throughout the body and brain to produce an array of effects including modifications in behavior (Herman et al., 2008; Papadimitriou and Priftis, 2009). The HPA axis is potently activated in response to stress and various studies have linked stress and subsequent hypersecretion of glucocorticoids to pathological behavioral and mood changes including increases in depression and anxiety (Qin et al., 2019; Shishkina and Dygalo, 2017). Therefore, regulation of the HPA axis by sex hormones is reviewed here as it is known to impact behavioral responses to stress.

In rodents, sex differences in stress-related behaviors are inconsistently reported, contrasting with literature in humans. In rats, various studies have reported that females show decreased indices of stress-related behaviors in assays that have traditionally been used to assess anxiety-like behaviors such as the elevated plus maze (Imhof et al., 1993; Steenbergen et al., 1990; Zimmerberg and Farley, 1993). Furthermore, female rats have also been reported to show decreased despair-like behavior in the forced swim test and decreased anhedonia-like behavior in the sucrose preference test relative to male rats (Barros and Ferigolo, 1998; Kokras and Dalla, 2014). Of note, other studies in rats report no sex differences (e.g. Lu et al., 2017) or sex differences in the opposite direction (Paré and Redei, 1993). In mice, sex differences in stress-related behaviors in traditional rodent assays of anxiety- and depressive-like behaviors vary by study, and assay used, and are sometimes reported in opposing directions (Bielajew et al., 2003; Burke et al., 2016; Võikar et al., 2001; Yau et al., 2019). Factors that contribute to this variability in reporting are genetic background (strain) and housing conditions (Bielajew et al., 2003; Burke et al., 2016; Hong et al., 2012; Palanza et al., 2001; Võikar et al., 2001). Furthermore, studies that involve assessing mice for stress-related behaviors following a period of prior stress exposure also yield different patterns of behavioral sex differences. For example, female mice exposed to a 6-day sub-chronic variable stress exposure show elevated indices of anxiety-like behavior in the novelty suppressed feeding test and increased anhedonia-like behavior in the sucrose preference test, while these behaviors were unaffected by sub-chronic variable stress in male mice (Hodes et al., 2015; Williams et al., 2020). It is also important to note that behavioral motivations may differ between the sexes in rodents which can affect interpretation of behavioral output. For example, factor analysis studies revealed that mouse and rat behavior in the elevated plus maze is primarily motivated by anxiety in males, but on the contrary is primarily motivated by activity in females (Fernandes et al., 1999; File, 2001). Sex differences in stress-related behaviors can also be influenced by estrous phase. For example, female mice tested in the open field, light dark box, and elevated plus maze show increased anxiety-like behavior during diestrus relative to proestrus (Jaric et al., 2019). This may also contribute variability in sex difference reporting of stress-related behaviors in studies that do not account for estrus status, which accounts for the majority of studies. However, it should be noted that recent meta-analyses have shown that stress-related behaviors in females tested without consideration of estrus status are not more variable than those of males (Kaluve et al., 2022). So, while estrus status is an important factor, it should not preclude the inclusion of females in studies.

In contrast to wide variability in reporting of sex differences in stress-related behaviors in rodents, sex differences in stress-induced activity of the HPA axis are consistently reported and reveal that females show greater activation. Specifically, female mice and rats show elevated levels of ACTH and corticosterone relative to males in response to a variety of psychogenic stressors and pharmacological agents (Babb et al., 2013; Handa et al., 1994a, Handa et al., 1994b; Iwasaki-Sekino et al., 2009; Jacobskind et al., 2017; Rosinger et al., 2020; Viau et al., 2005). Female rats are also often reported to show higher levels of baseline corticosterone compared to males (Babb et al., 2013; Iwasaki-Sekino et al., 2009). Furthermore, the temporal dynamics of the HPA axis response to stress differs between male and female rodents with females showing a prolonged ACTH and corticosterone response to stressors (Babb et al., 2013; Handa et al., 1994a, Handa et al., 1994b; Iwasaki-Sekino et al., 2009; Viau et al., 2005). This prolonged response, as well as more robust response in females is due to sex differences in negative feedback on the HPA axis (Babb et al., 2013; Handa et al., 1994a, Handa et al., 1994b; Iwasaki-Sekino et al., 2009; Viau et al., 2005). Mechanisms that underlie the elevated peak and sustained rise in glucocorticoids in female rodents include higher levels of stress induced CRF mRNA in the paraventricular hypothalamus and proopiomelanocortin (POMC; precursor to ACTH) in the anterior pituitary (Babb et al., 2013; Iwasaki-Sekino et al., 2009; Seale et al., 2004; Viau et al., 2005). Furthermore, female rats show lower levels of glucocorticoid binding in the hypothalamus and fewer corticosteroid receptors in the pituitary gland which likely contribute to sex differences in HPA axis negative feedback (Turner, 1990; Turner and Debra, 1985).

Unlike the robust and generally consistent sex differences in HPA axis responses to stress reported in rodents, reports of sex differences in human HPA axis reactivity to stress are quite variable (Handa et al., 2022). One factor that appears to contribute to somewhat inconsistent results in humans is the differential nature of the stressor (generally psychosocial) that is applied. For example, some studies report greater HPA axis reactivity in women when they are assessed for social performance, while men show elevated HPA axis responses when evaluated for achievement (Kirschbaum et al., 1999; Kudielka and Kirschbaum, 2005; Stroud et al., 2002; Uhart et al., 2006). Other studies have reported no sex differences in human HPA axis responses to stress (Kudielka and Kirschbaum, 2005). It is important to note that this stressor modality is also not consistent across species where the majority of studies in rodents utilize a psychological stressor that induces a strong physical experience (e.g. restraint, swim, shock) while in humans the stressors are predominantly psychosocial (Altemus, 2006; Handa et al., 2022). This factor affects interpretation of sex differences across species. Furthermore, variability in reporting of sex differences in HPA axis responses within humans appears to be influenced by age, health, menstrual cycle phase, use of oral contraceptives, and potentially gender identity (Bangasser and Valentino, 2014; Handa et al., 2022; Kajantie and Phillips, 2006; Kirschbaum et al., 1999; Seeman et al., 2001). In this review we will primarily focus on the role of androgens in regulating sex differences in stress-related behaviors and will also review androgen regulation of the HPA axis, a system that can ultimately impact behaviors related to stress.