Prodynorphin-derived peptides are critical modulators of anxiety and regulate neurochemistry and corticosterone

Abstract

Stress and anxiety are mainly regulated by amygdala and hypothalamic circuitries involving several neurotransmitter systems and providing physiological responses to peripheral organs via the hypothalamic-pituitary-adrenal axis and other pathways. The role of endogenous opioid peptides in this process is largely unknown. Here we show for the first time that anxiolytic parameters of explorative behavior in mice lacking prodynorphin were increased 2-4-fold in the open field, the elevated plus maze and the light-dark test. Consistent with this, treatment of wild-type mice with selective kappa-opioid receptor antagonists GNTI or norbinaltorphimine showed the same effects. Furthermore, treatment of prodynorphin knockout animals with U-50488H, a selective kappa-opioid receptor agonist, fully reversed their anxiolytic phenotype. These behavioral data are supported by an approximal 30% reduction in corticotropin-releasing hormone (CRH) mRNA expression in the hypothalamic paraventricular nucleus and central amygdala and an accompanying 30-40% decrease in corticosterone serum levels in prodynorphin knockout mice. Although stress-induced increases in corticosterone levels were attenuated in prodynorphin knockout mice, they were associated with minor increases in depression-like behavior in the tail suspension and forced swim tests. Taken together, our data suggest a pronounced impact of endogenous prodynorphin-derived peptides on anxiety, but not stress coping ability and that these effects are mediated via kappa-opioid receptors. The delay in the behavioral response to kappa-opioid receptor agonists and antagonist treatment suggests an indirect control level for the action of dynorphin, probably by modulating the expression of CRH or neuropeptide Y, and subsequently influencing behavior.

Figure 1

Ambulation in the open field was assessed over 10 min for dyn^(+/+) and dyn^(−/−) mice. Dyn^(−/−) mice more often entered the intermediate and central zone and spent significantly more time and traveled longer distances in these compartments. In contrast, dyn^(+/+) mice preferentially remained in the border zone (a). Intraperitoneal injection of the KOR-specific agonist U-50488H did not affect the behavior of mice 30 min after treatment but reversed the anxiolytic phenotype of dyn^(−/−) mice after 2 days (b). In line with this, treatment of dyn^(+/+) mice with specific KOR antagonists norBNI (10 mg/kg, i.p.) and GNTI (3 nmol, i.c.) mimicked the phenotype of dyn^(−/−) mice after 48 and 20 h, respectively (c). *p<0.05; **p<0.01; ***p<0.001 vs respective control.

Figure 2

Explorative behavior was tested in group- and single-housed mice (a). Independent of housing conditions, dyn^(−/−) mice more often entered the open arms and spent significantly more time on the open arms than dyn^(+/+) mice. No differences were observed in the number of closed arm entries, suggesting unchanged motor activity (a). Intraperitoneal injection of the KOR-specific agonist U-50488H did not affect the behavior of mice 30 min after treatment but reversed the anxiolytic phenotype of dyn^(−/−) mice after 2 days (b). In line with this, treatment of dyn^(+/+) mice with specific KOR antagonists norBNI (10 mg/kg, i.p.) and GNTI (3 nmol, i.c.) mimicked the phenotype of dyn^(−/−) mice after 48 and 20 h, respectively (c). *p<0.05; **p<0.01; ***p<0.001 vs respective control.

Figure 3

At 150 and 400 lux illumination, dyn^(+/+) mice displayed less explorative behavior in the lit area, as reflected by lower time (a) and lower distance (b) measured in the open compartment. Interestingly, open area entries differed only at the higher aversive 400 lux (c). *p<0.05; ***p<0.01 vs respective control.

Figure 4

Time spent immobile did not differ between dyn^(+/+) and dyn^(−/−) mice in the first and last trial of the repeated forced swim test. However, immobility times increased earlier in dyn^(−/−) mice, reaching statistical significance at trial 3 (a). Non prestressed dyn^(+/+) mice displayed less immobility in the tail suspension test than dyn^(−/−) mice. Interestingly, in prestressed animals, no difference was observed, which was due to increased immobility in dyn^(+/+) mice (b). Immobility was reduced in dyn^(−/−), but not in dyn^(+/+) mice under meloxicam anesthesia (b). This is in line with reports on mild hyperalgesia and lack of stress-induced analgesia in dyn^(−/−) mice. No differences between the genotypes were observed in stress-induced hyperthermia (c). Basal corticosterone serum levels were reduced in both dyn^(+/+) mice pretreated with 10 mg/kg norBNI and dyn^(−/−) animals. Stress-induced corticosterone levels were also augmented in dyn^(−/−) mice (d). *p<0.05; **p<0.01; ***p<0.001 vs respective control.

Figure 5

High-resolution in situ hybridization for CRH mRNA in the central amygdala (upper panel) and the paraventricular nucleus (lower panel). Photomicrographs obtained from 20 μm sections are depicted for wild-type mice 48 h after saline (a, d) or 10 mg/kg norBNI (b, e) injection and saline-injected dyn^(−/−) mice (c, f). Note the marked drop in signal after norBNI treatment, which is comparable to the labeling seen in dyn^(−/−) animals.