Orexin (Hypocretin) Signaling in the Basolateral Amygdala Contributes to Individual Differences in Stress Sensitivity

Exposure to stress contributes to the development of psychiatric disorders (1). Notable in this regard is the ongoing global COVID-19 pandemic, which has been linked to an increase in the prevalence of several stress-related disorders, including depression and posttraumatic stress disorder, as well as increased rates of drug and alcohol abuse (2). Some individuals are more prone to stress than others (1,2), and thus efforts to characterize the neural basis of stress responsivity in susceptible versus resilient populations may be informative for developing more effective therapeutics to treat stress-related disorders.

The orexin (Orx; hypocretin) neuropeptide system represents a promising candidate system contributing to individual differences in stress sensitivity (3,4). Hypothalamic neurons that produce the Orx peptides A (Orx_A) and B (Orx_B) are activated after exposure to acute stressors, and increased Orx levels are observed in neuropsychiatric patients with anxiety symptoms (3–5). Orx_A and Orx_B exert their neuromodulatory effects via two G protein–coupled receptors, Orx receptor 1 (Orx₁R) and Orx₂R (6). Manipulations of signaling at Orx₁R versus Orx₂R affect distinct aspects of stress behavior; however, it remains unclear how signaling at either receptor, or their interplay, might contribute to stress susceptibility (7). One interesting candidate region for Orx signaling in stress is the basolateral amygdala (BLA), a region classically associated with behavioral responses to traumatic stressors (7). Indeed, several preclinical studies indicate that Orx acting in the BLA might support adaptive stress responsivity. For example, BLA Orx is enhanced after repeated stress, and intra-BLA injection of Orx_A results in depressive-like behaviors (4). In contrast, intra-BLA inhibition of Orx₁R facilitates extinction of fear memories (4). To date, however, no study has investigated how orexin signaling in the BLA might contribute to individual differences in stress responsivity or how these effects might be mediated by actions at Orx₁R versus Orx₂R.

In the current issue of Biological Psychiatry, Yaeger et al. (8) begin to explore these questions using the rodent Stress Alternatives Model (SAM), an ethologically relevant paradigm of social stress that identifies stable individual differences in stress responsivity (Figure 1) (4,9). In the SAM, animals are exposed to social aggression from a larger conspecific across multiple days. Sessions are conducted in an arena equipped with size-restricted escape routes, enabling the smaller subject to shorten the duration of the stressful encounter (i.e., to escape). SAM results in two distinct and stable phenotypes: 1) Stay animals (those that remain in the arena and continue to be aggressively attacked) and 2) Escape animals (those that use the escape route to avoid the aggressor) (4,9). Previous studies by the authors indicate that Stay animals display heightened stress susceptibility, including enhanced anxiety and depressive-like behavior. In contrast, Escape animals are resilient to stress [reviewed in (4,9)].

Figure 1.

Inhibition of Orx₁R signaling in the BLA promotes distinct behavioral and molecular changes in Stay and Escape mice. The SAM identifies a roughly equal distribution of Stay (left box) and Escape (right box) mice that exhibit stress-susceptible versus stress-resilient behaviors, respectively. Highlighted here are the behavioral and molecular consequences of inhibiting Orx₁R versus Orx₂R signaling in the BLA in each phenotype immediately prior to the third SAM session, which is after the development of stable behavioral phenotypes. In general, blocking BLA Orx₁R signaling in Stay animals promotes a shift toward the Escape phenotype. Green arrow, significantly greater; red arrow, significantly lower. Bdnf, brain-derived neurotrophic factor; BLA, basolateral amygdala; Hcrtr1/Orx₁R, orexin receptor 1; Hcrtr2/Orx₂R, orexin receptor 2; SAM, Stress Alternatives Model.

As in previous studies, the authors found an even distribution of Stay and Escape mice after 2 days of the SAM (all experiments used male mice). Escape mice displayed increased time spent investigating escape routes, which the authors interpret as enhanced motivation for active avoidance (or stress resilience). To test the role of BLA Orx signaling in the expression of Escape and Stay phenotypes, the authors used a combination of pharmacological and genetic knockdown strategies. We focus on the major finding that intra-BLA Orx₁R inhibition resulted in a shift from a Stay (stress-susceptible) to an Escape (stress-resilient) phenotype (summarized in Figure 1). First, infusion of a selective Orx₁R antagonist in the BLA increased attention toward escape routes in Stay mice, which corresponded to an increase in the percentage of these mice escaping the aggressor. In contrast, intra-BLA Orx_A infusion led to a reduction in attention toward the escape routes in Escape mice, pointing to bidirectional effects of Orx signaling on the expression of stress behaviors. Second, intra-BLA Orx₁R inhibition in Stay mice was associated with a decrease in circulating levels of the stress hormone corticosterone after the SAM paradigm. Together, these data indicate that in stress-susceptible mice, inhibition of Orx₁R signaling in the BLA results in a shift toward a stress-resilient state, possibly by reducing BLA-mediated recruitment of the hypothalamic-pituitary-adrenal axis. Remarkably, on the day after intra-BLA Orx₁R blockade, Stay mice spent less time with the social aggressor, pointing to experience-dependent plasticity mediated by BLA Orx₁R receptors. What might be mediating this neuroplasticity? To investigate this, the authors quantified the expression of brain-derived neurotrophic factor (Bdnf) in the BLA after local infusions of the Orx₁R antagonist. As predicted, Bdnf expression increased after BLA Orx₁R inhibition in Stay versus Escape mice. This indicates that the behavioral switch observed in Stay mice after BLA Orx₁R inhibition could be a result of neuroplasticity orchestrated by downstream genetic changes (discussed below).

Previous research indicates that the generalization of fear memories is associated with panic disorders and post-traumatic stress disorder, and that the Orx system is involved in this process (7). Thus, the authors tested whether Orx signaling in the BLA plays a role in cue and contextual fear memory after chronic social stress. Both Stay and Escape mice displayed enhanced freezing to a discrete cue that predicted the social aggressor (Figure 1) (8). This increase in fear response was ameliorated by intra-BLA Orx₁R, but not Orx₂R, antagonist administration in Stay mice, whereas the opposite was true in Escape mice. These results hint at different contributions of Orx₁R versus Orx₂R signaling after stress in stress-resistant versus susceptible individuals.

To further probe this potential phenotypic difference in Orx receptor signaling, Yaeger et al. (8) characterized BLA Orx₁R (Hcrtr1) and Orx₂R (Hcrtr2) gene expression in Stay versus Escape mice after exposure to the SAM procedure. Escape mice express higher levels of Hcrtr2 in the BLA compared with Stay mice; given that these receptors are almost exclusively located on GABAergic (gamma-aminobutyric acidergic) neurons, this finding indicates that adaptive stress behavior may be linked with enhanced orexinergic influence of inhibitory circuits. Consistent with this, BLA Orx₁R inhibition in Stay mice is associated with an increase in Hcrtr2 expression, perhaps representing one mechanism underlying enhanced escape behavior in these animals after intra-BLA Orx₁R infusions (although the phenotype of the cells expressing increased Hcrtr2 after SAM is not known). Moreover, the magnitude with which Hcrtr2 increased in Stay animals tended to be associated with lower freezing in response to the fear cue, further pointing to a proresilience role for Hcrtr2 expression in BLA.

To examine downstream signaling dynamics that might mediate the behavioral effects of inhibiting BLA Orx₁R signaling in Stay mice, the authors quantified expression of ERK genes, Mapk1 and Mapk3, as these genes normally drive downstream changes after Orx receptor activation (8). While Mapk1 expression was unaltered, Mapk3 expression increased in Stay mice compared with Escape mice after BLA Orx₁R antagonism. This increase in Mapk3 expression was associated with reduced responsivity to fear cues after intra-BLA Orx₁R antagonism, perhaps indicating a mechanistic link. Finally, in an attempt to draw together their behavioral and molecular findings, the authors propose a working model to explain the shift from the Stay to Escape phenotype after BLA Orx₁R inhibition. In Stay mice, inhibition of BLA Orx₁R signaling increases the expression of Hcrtr2, Mapk3, and Bdnf in putative GABAergic neurons. Under this model, the Stay → Escape phenotype transition is thus associated with biased Orx₂R activation in the local BLA inhibitory neurons, which promotes downstream molecular changes that are protective against stress. This model provides an excellent starting point for future studies focused on further characterizing how the balance of Orx₁R versus Orx₂R signaling in BLA might contribute to stress sensitivity.

Together, the data presented by Yaeger et al. (8) add considerably to an emerging literature that indicate a critical, albeit complicated, role for Orx signaling in stress behaviors. Most notably, the authors reveal what appears to be conflicting roles for BLA Orx₁R versus Orx₂R signaling in stress-prone animals, whereby blockade of Orx₁R promotes stress resilience by biasing signaling at Orx₂R. This observation indicates that selective Orx₁R antagonists may have some therapeutic utility for the treatment of stress disorders, particularly in susceptible individuals. This aligns with evidence from studies examining Orx₁R in the context of drug abuse, which points to enhanced efficacy of Orx₁R antagonists specifically in high-risk individuals (6,10). In contrast, the efficacy of dual Orx receptor antagonists for promoting stress resilience may be limited. It will be interesting for future studies to test whether the effects observed here generalize to other stress paradigms in which individual differences in behavioral outcomes are observed, including the chronic unpredictable mild stress or social defeat paradigms. Several other questions are important to address. For example, do the behavioral and molecular changes observed here in male mice also occur in female mice? Are the differences in Orx₁R and Orx₂R in Stay versus Escape mice preexisting, or do they arise as a result of SAM experience? If the former, then the relative expression of Orx₁R versus Orx₂R might act as a biomarker for stress sensitivity. Thus, Yaeger et al. (8) not only provide compelling mechanistic insights into how Orx signaling may be linked to individual differences in stress sensitivity, but also highlight several exciting new directions to be explored by future studies focused on understanding the etiology of stress disorders.