Earlier reports had indicated that exposure to a series of unpredictable mild (Monleon et al., 1995
; Willner et al., 1987
), relatively intense (Katz, 1982
), or uncontrollable stressors (Griffiths et al., 1992
) results in persistent reductions in the consumption of sweet solutions. Likewise, we find that restraint stress causes a persistent decrease in sucrose preference that does not vary across days of restraint, suggesting that behavioral adaptation to repeated exposure to restraint stress did not occur within 10 days of restraint. Moreover, in agreement with previous reports (Ayensu et al., 1995
; Willner et al., 1987
), caloric content seems to be unimportant since similar effects of restraint stress are observed in animals consuming a calorie-free, saccharin solution.
Contrary to a previous report (Forbes et al., 1996
), in our study, the decrease in sucrose preference in food-deprived and stressed mice is not due to a decrease in body weight. The use of the two-bottle choice paradigm allows for the determination of sucrose preference rather than the absolute amount of sucrose consumed; sucrose preference was unaffected by changes in body weight (compare Fig. and ). Both sucrose consumption normalized to body weight and sucrose preference were significantly reduced by an acute exposure to restraint stress prior to the fluid consumption test. Thus, restraint stress causes a decreased sensitivity to natural reward.
receptor agonists enhance voluntary sucrose consumption (Kirkham and Williams, 2001
) and CB1
receptor antagonists decrease sucrose consumption (Higgs et al., 2003
), doses of CP55940, rimonabant, and URB597, that had no effect on sucrose preference in stress-naïve mice were identified. This allowed for the determination of the effects of CB1
receptor activation and blockade on stress-induced decreased sensitivity to natural reward rather than on baseline consumption. Pretreatment with these doses of CP55940 or URB597 blocked restraint stress-induced decreased sucrose preference on restraint days 1, 4, and 7, but not 10. By contrast, pretreatment with a dose of rimonabant that had no effect on sucrose preference in unstressed mice accentuated restraint stress-induced decreased sucrose preference on restraint days 1, 4, 7, and 10. This finding is in agreement with the report that CB1
receptor deletion decreased sucrose solution intake in mice chronically exposed to unpredictable mild stressors (Martin et al., 2002
). The finding that CP55940, URB597, and rimonabant had qualitatively similar effects on stress-induced decreases in sucrose and saccharin preference suggests that the caloric content of the solution was not an important factor. Thus, while CB1
receptor activation blocks, CB1
receptor blockade exacerbates stress-induced decreased sensitivity to natural reward.
We hypothesize that endocannabinoids modulate decreased sensitivity to natural reward induced by acute restraint stress by regulating the central stress response. In support of this hypothesis, it has been demonstrated that endocannabinoids modulate hypothalamus-pituitary-adrenal axis function (Haller et al., 2004
; Manzanares et al., 1999
; Patel et al., 2004
). In particular, earlier studies from our laboratory (Patel et al., 2004
) demonstrated that CB1
receptor blockade potentiates restraint stress-induced hypothalamus-pituitary-adrenal axis activation and neuronal activity in the paraventricular nucleus of the hypothalamus, and CB1
receptor activation blocks restraint stress-induced hypothalamus-pituitary-adrenal axis activation. These data are consistent with the effects of CB1
receptor agonists and antagonists to affect stress-induced sucrose preference reported herein.
We hypothesize that a global function for endocannabinoid/CB1
receptor signaling is to dampen stress-induced behavioral changes, including the expression of active escape behaviors during acute and repeated restraint stress episodes (Patel et al., 2005b
), immobility time in a tail suspension test and swimming time in a forced swim test (Gobbi et al., 2005
), and decreased sensitivity to natural reward. Our data is in accord with previous research that demonstrated the protective role of the endocannabinoids becomes more pronounced as the stress is repeated. While a single exposure to restraint stress had no effect on 2-AG content in the limbic forebrain and amygdala, that same restraint exposure produced a large increase in 2-AG content in the limbic forebrain and amygdala following the fifth restraint episode (Patel et al., 2005b
). Thus, 2-AG content in the limbic forebrain and amygdala exhibit sensitization to repeated restraint exposure. While CP55940 and URB597 both blocked the effects of restraint to reduce sucrose preference on day 1, these drugs were ineffective on day 10. There are two possible explanations for this change: first, the CB1
receptor no longer plays a role in stress-induced effects on reward or, second, the CB1
receptor is maximally activated such that further agonist treatment produces no effect. The finding that CB1
receptor blockade on day 10 results in a reduction in sucrose preference that is significantly greater than its effect on day 1 supports the second scenario, i.e. CB1
receptors maximally activated and required for the maintenance of sucrose preference in the face of 10 days of restraint stress. The finding that after 10 days of restraint, rimonabant, which had no effect on sucrose preference in the absence of restraint, significantly reduced sucrose preference in the absence of acute stress suggest that a global increase in endocannabinoid tone induced by subchronic restraint stress persists for at least 24 hours after the termination of the stressor.