Oxytocin attenuates demand for cocaine in female rats

Abstract

There are substantial sex differences in substance use disorders (SUDs), and a key feature of SUD is pathologically high economic demand for drug. The hypothalamic neuropeptide oxytocin (OXT) is heavily implicated in the modern treatment of SUDs. Using a within-session threshold behavioral economics (BE) procedure, we quantified demand elasticity (a, inverse motivation) and free consumption (Q₀) in male and female rats to investigate the effect of OXT on cocaine demand. Results showed that OXT decreased motivation for cocaine; an effect greater during the high-demand phase (diestrus, low progesterone, P₄) vs low demand phases (proestrus, high P₄). We confirmed our prior findings that P₄ attenuates cocaine demand in female rats and that chronic cocaine self-administration disrupts estrus cyclicity. Following each injection, OXT at either 0.1mg/kg or 0.3mg/kg restored estrous cycling in intact females with prior cocaine experience for one week and remained effective with up to 4 weeks of injections. Fos reactivity in OXT+ neurons was greater when rats were in proestrus compared to diestrus and significantly correlated to motivation and circulating levels of P₄. Finally, using ovariectomized females with P₄ replacement, we show that P₄'s demand attenuating effects are reversed by atosiban (1.0 mg/kg, IP), an OXT antagonist. These data show an interaction between oxytocin and progesterone in female rats that may underlie differences in cocaine demand between sexes. Additionally, we show critical periods for using OXT as a treatment to reduce cocaine demand in females. Our results indicate novel therapeutic treatments for SUDs must be tailored to hormonal states.

Keywords: Behavioral economics; Cocaine; Oxytocin; Progesterone; Sex differences.

Fig. 1.

Representative images from frontal brain sections of OXT and Fos expression in the paraventricular nucleus (PVN). In all images, brown staining indicates OXT-positive neurons (brown arrows), and purple staining indicates Fos-positive nuclei (red arrows). In all images, black arrows indicate double-labeled neurons. PVN subregions were identified using landmarks established by Swanson and Sawchenko (1980). A. PVNr. Representative image of OXT/Fos dual labeling in PVNr (AP – 1.60; Paxinos and Watson 2007). B. PVNm. Representative images of OXT/Fos dual labeling in PVNm (AP – 1.88; Paxinos and Watson 2007). C. PVNc. Representative images of OXT/Fos dual labeling in PVNc (AP – 2.12; Paxinos and Watson 2007) D. Representative image of SON OXT/Fos dual labeling (AP – 1.80; Paxinos and Watson 2007).

Fig. 2.

The percentage of Oxytocin/Fos+ neurons in the rostral paraventricular nucleus negatively predicts economic demand for cocaine in intact female rats. A. Self-administration behaviors over the 10 days of FR-1 acquisition prior to starting behavioral economics training. B. Demand elasticity (α) and consumption at null effort (Q₀) in proestrus (P) and diestrus (D) rats. Demand elasticity (α; inverse to motivation for cocaine) was lower when females were in diestrus compared to when they were in proestrus. No significant differences were observed in Q₀. *p < 0.05. C. Female rats in diestrus have significantly less Fos+ OXT neurons in the rostral region of the paraventricuar nucleus (PVN_R). * p < 0.05. D. Female rats in diestrus have significantly less Fos+ OXT neurons in the middle region of the paraventricuar nucleus (PVN_M). * p < 0.05. E. There were no significant differences between cycle phases in Fos+ OXT neurons of the caudal region of the paraventricular nucleus (PVN_C). F. There were no significant differences between cycle phases in Fos+ neuron reactivity in OXT neurons of the supraoptic nucleus (SON). G. Fos+ OXT neurons in the PVN_R positively correlated to α values. Linear regressions are plotted (black lines for collapsed females across cycle, pink lines and symbols for females in proestrus, green lines and symbols for females in diestrus). * p < 0.05. H. Fos+ OXT neurons in the PVN_R positively correlated to circulating progesterone levels. Linear regressions are plotted (black lines for collapsed females across cycle, pink lines and symbols for females in proestrus, green lines and symbols for females in diestrus). * p < 0.05.

Fig. 3.

Oxytocin significantly decreases economic demand for cocaine in female rats and restores estrous cyclicity. A. Demand elasticity (α; inverse to motivation for cocaine) was significantly attenuated by oxytocin at 0.1 and 0.3 mg/kg while females were in diestrus, and 0.3 mg/kg only while in proestrus. * p < 0.05. B. Consumption at null effort (Q₀) is increased by 0.1 mg/kg oxytocin only during proestrus. * p < 0.05. C. FR-1 or BE cocaine self-administration disrupted estrous cyclicity, as measured by the observed number of different cycle events. Oxytocin treatments occurred in the time period indicated by the grey overlay. On the week following oxytocin treatment, cycle events normalized as compared to the prior week on BE self-administration of cocaine. * p < 0.05. D. Color coded bar graph depiciting the percentage of each cycle phase observed. Following OXT, proestrus epochs return to 25 % of observed cycle phases across subjects in each week. Statistics are described in text.

Fig. 4.

Atosiban, an oxytocin receptor antagonist, partially attenuates progesterone’s treatment efficacy on cocaine demand. A. Timeline depicting testing schedule. B. Progesterone administration increases demand elasticity (α; inverse to motivation for cocaine) in ovariectomized rats, an effect which is partially attenuated by co-administration with atosiban 1.0 mg/kg, an oxytocin antagonist. Atosiban did not impact α in the absence of progesterone. * p < 0.05. C. Cocaine consumption at null effort (Q₀) was increased by administration of 4mg/kg progesterone, which was attenuated by co-administration of atosiban 1.0 mg/kg. Atosiban did not impact Q₀ in the absence of progesterone. * p < 0.05.