Overnight food deprivation markedly attenuates hindbrain noradrenergic, glucagon-like peptide-1, and hypothalamic neural responses to exogenous cholecystokinin in male rats

Abstract

Systemic administration of sulfated cholecystokinin-8 (CCK) activates neurons within the hindbrain nucleus of the solitary tract (NTS) that project directly to the paraventricular nucleus of the hypothalamus (PVN), and these projections underlie the ability of exogenous CCK to activate the hypothalamic-pituitary-adrenal (HPA) stress axis. CCK inhibits food intake, increases NTS neuronal cFos expression, and activates the HPA axis in a dose-dependent manner. While the hypophagic effects of exogenous CCK are attenuated in food-deprived rats, CCK dose-response relationships for NTS and hypothalamic activation in fed and fasted rats are unknown. Within the NTS, noradrenergic A2 and glucagon-like peptide-1 (GLP-1) neurons express cFos after high doses of CCK, and both neuronal populations project directly to the medial parvocellular (mp)PVN. We hypothesized that increasing and correlated proportions of A2, GLP-1, and mpPVN neurons would express cFos in rats after increasing doses of CCK, and that food deprivation would attenuate both hindbrain and hypothalamic neural activation. To test these hypotheses, ad libitum-fed (ad lib) and overnight food-deprived (DEP) rats were anesthetized and perfused with fixative 90min after i.p. injection of 1.0ml saline vehicle containing CCK at doses of 0, 3, or 10μg/kg BW. Additional ad lib and DEP rats served as non-handled (NH) controls. Brain tissue sections were processed for dual immunocytochemical localization of cFos and dopamine-β-hydroxylase to identify A2 neurons, or cFos and GLP-1. Compared to negligible A2 cFos activation in NH control rats, i.p. vehicle and CCK dose-dependently increased A2 activation, and this was significantly attenuated by DEP. DEP also attenuated mpPVN cFos expression across all treatment groups, and A2 activation was strongly correlated with mpPVN activation in both ad lib and DEP rats. In ad lib rats, large and similar numbers of GLP-1 neurons expressed cFos across all i.p. treatment groups, regardless of CCK dose. Surprisingly, DEP nearly abolished baseline GLP-1 cFos expression in NH controls, and also in rats after i.p. injection of vehicle or CCK. We conclude that CCK-induced hypothalamic cFos activation is strongly associated with A2 activation, whereas the relationship between mpPVN and GLP-1 activation is less clear. Furthermore, activation of A2, GLP-1, and mpPVN neurons is significantly modulated by feeding status, suggesting a mechanism through which food intake and metabolic state might impact hypothalamic neuroendocrine responses to homeostatic challenge.

Keywords: CCK; Hypothalamic–pituitary–adrenal axis; Nucleus of the solitary tract; Paraventricular nucleus of the hypothalamus; cFos.

Figure 1

Representative color images and summary data for neuronal cFos expression (black nuclear label) within the A2 region of the caudal NTS. A, NTS cFos activation at the rostrocaudal level of the area postrema, approximately 14.16 mm caudal to bregma. In the larger image, robust cFos expression is present in a rat after CCK (10 μg/kg BW), including activation of DβH-positive (brown) neurons comprising the A2 cell group. Inset, little or no cFos activation is present in an ad lib-fed rat after i.p. vehicle (0 μg/kg BW CCK). B, bar graph illustrating the number of double labeled (i.e., both cFos- and DβH-positive) NTS neurons in ad lib-fed rats (solid bars) or food-deprived rats (DEP; open bars) after no i.p. injection (nonhandled) or after injection of CCK at doses of 0 (vehicle), 3, or 10 μg/kg BW. See Table 1 for two-way ANOVA results. In both feeding status groups, CCK dose-dependently increased cFos expression in DβH-positive A2 neurons, but cFos activation was significantly attenuated in food-deprived rats after vehicle or CCK injections. Asterisks over DEP bars indicate significant differences (p < .05) in the number of double-labeled NTS neurons compared to ad lib-fed rats in the same i.p. treatment group. Within the same feeding status group (i.e., ad lib or DEP), bars with different letters are significantly different (p < .05). Scale bars in A = 100 μm.

Figure 2

Representative color images and summary data for neuronal cFos expression (red fluorescence) within GLP-1-positive neurons (green fluorescence) of the caudal NTS. A, in an ad lib-fed rat, many GLP-1 neurons express cFos after i.p. injection of CCK (3 μg/kg BW). Inset, higher magnification view of several double-labeled neurons. B, in a food-deprived rat, GLP-1 neurons are not activated to express cFos after CCK (3 μg/kg BW), despite activation of other caudal NTS neurons. C, bar graph illustrating the number of double labeled (i.e., both cFos- and GLP-1-positive) neurons within the NTS and adjacent reticular formation in ad lib-fed rats (solid bars) or food-deprived rats (DEP; open bars) after no i.p. injection (nonhandled) or after injection of CCK at doses of 0 (vehicle), 3, or 10 μg/kg BW. See Table 1 for two-way ANOVA results. In ad lib-fed rats, a moderate number of GLP-1 neurons express cFos in the nonhandled control condition, and this number more than doubled after i.p. injection of vehicle or CCK, regardless of dose. Food deprivation markedly suppressed cFos expression by GLP-1 neurons, with a non-significant trend towards increased activation seen after the highest CCK dose (10 μg/kg BW). Asterisks indicate significantly reduced (p < .05) GLP-1 cFos expression in DEP rats compared to GLP-1 cFos expression in ad lib-fed rats in the same i.p. treatment group. Within the same feeding status group (i.e., ad lib or DEP), bar values with different letters are significantly different (p < .05). cc, central canal. Scale bars in A and B = 100 μm.

Figure 3

Representative color images and summary data for neuronal cFos expression (red fluorescence) within the GLP-1 (green fluorescence) terminal-rich region of the mpPVN. A, little cFos is present in the mpPVN of an ad lib-fed, nonhandled control rat. B, robust mpPVN cFos activation is present in an ad lib-fed rat injected with CCK (10 μg/kg BW). mpPVN images were captured and cFos quantified in sections approximately 1.78 mm caudal to bregma, as schematized in the inset in panel A [77]. C, bar graph illustrating the number of cFos-positive mpPVN neurons in ad lib-fed rats (solid bars) or food-deprived rats (DEP; open bars) after no i.p. injection (nonhandled) or after injection of CCK at doses of 0 (vehicle), 3, or 10 μg/kg BW. See Table 1 for two-way ANOVA results. In both feeding status groups, low levels of mpPVN cFos activation were present in nonhandled or vehicle-injected rats, and CCK delivered at doses of 3 and 10 μg/kg produced marked activation of mpPVN cFos. Food-deprived rats displayed less mpPVN cFos labeling overall compared to labeling in ad lib-fed rats. Within the same feeding status group (i.e., ad lib or DEP), bars with different letters indicate significant i.p. treatment-induced differences (p < .05) in mpPVN cFos activation. mp = medial parvocellular subdivision of the PVN; ml = lateral magnocellular subdivision of the PVN; 3V = third ventricle. Scale bars in A and B = 100 μm.

Figure 4

Scatter plots depicting significant positive correlations between treatment-induced activation of (A) mpPVN and NTS DβH neurons, (B) mpPVN and GLP-1 neurons, and (C) GLP-1 and DβH neurons. In each plot, individual symbols represent data from a single animal. Different symbol styles represent different i.p. treatment groups, as indicated in the key. See Table 2 for a breakdown of correlation R values and significance by i.p. treatment and feeding status groups.