Lateral parabrachial nucleus astrocytes control food intake

Abstract

Food intake behavior is under the tight control of the central nervous system. Most studies to date focus on the contribution of neurons to this behavior. However, although previously overlooked, astrocytes have recently been implicated to play a key role in feeding control. Most of the recent literature has focused on astrocytic contribution in the hypothalamus or the dorsal vagal complex. The contribution of astrocytes located in the lateral parabrachial nucleus (lPBN) to feeding behavior control remains poorly understood. Thus, here, we first investigated whether activation of lPBN astrocytes affects feeding behavior in male and female rats using chemogenetic activation. Astrocytic activation in the lPBN led to profound anorexia in both sexes, under both ad-libitum feeding schedule and after a fasting challenge. Astrocytes have a key contribution to glutamate homeostasis and can themselves release glutamate. Moreover, lPBN glutamate signaling is a key contributor to potent anorexia, which can be induced by lPBN activation. Thus, here, we determined whether glutamate signaling is necessary for lPBN astrocyte activation-induced anorexia, and found that pharmacological N-methyl D-aspartate (NMDA) receptor blockade attenuated the food intake reduction resulting from lPBN astrocyte activation. Since astrocytes have been shown to contribute to feeding control by modulating the feeding effect of peripheral feeding signals, we further investigated whether lPBN astrocyte activation is capable of modulating the anorexic effect of the gut/brain hormone, glucagon like peptide -1, as well as the orexigenic effect of the stomach hormone - ghrelin, and found that the feeding effect of both signals is modulated by lPBN astrocytic activation. Lastly, we found that lPBN astrocyte activation-induced anorexia is affected by a diet-induced obesity challenge, in a sex-divergent manner. Collectively, current findings uncover a novel role for lPBN astrocytes in feeding behavior control.

Keywords: GLP-1; NMDA; astrocytes; ghrelin; glia; high-fat diet; hindbrain; lateral parabrachial nucleus.

Figure 1

Activation of lateral parabrachial nucleus (lPBN) astrocytes leads to hypophagia in male and female rats. Activation of lPBN astrocytes by IP CNO injection in male rats expressing activational DREADD receptors on lPBN astrocytes leads to reduced chow intake in overnight-fasted rats (A). Activation of lPBN astrocytes also reduces chow intake during the natural feeding period, dark cycle, in ad-libitum-fed male rats (B). Similarly, a robust food intake reduction results from lPBN astrocyte activation in overnight-fasted female rats (C). Activation of lPBN astrocytes in female rats during the dark cycle also leads to lower chow intake (D). mCherry expression introduced by the DREADD-carrying AAV indicates that the virus spread throughout the lPBN but not into the medial PBN (E, F). Expression of AAV-introduced label was present exclusively on astrocytes as indicated by the complete overlap of cell bodies expressing Gfap, as measured by RNAscope (green), and AAV-introduced mCherry (red) (G–J). DAPI, a nuclear stain is labeled in blue. Fasted study females: n = 12 per group, males: n = 8–19. Dark cycle study: females: n = 6, males: n = 8–9. Data are expressed as mean ± SEM. Post-hoc tests under line graphs refer to Controls to CNO comparisons (all DREADD positive). #p = 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. DREADD -ve: CNO control rats, DREADD-negative rats.

Figure 2

The orexigenic effects of ghrelin are attenuated, and the anorexic effects of exendin-4 are potentiated by lPBN astrocytic activation. Activation of GHSRs in lPBN results in a potent orexigenic response, which is attenuated by chemogenetic activation of lPBN astrocytes in ad-libitum-fed male rats at 1 h (A) and 3 h (B) post injections. At 5 h post-lPBN CNO and ghrelin injections, ghrelin remains orexigenic, but astrocyte-induced attenuation is no longer significant (C). Body weight was not altered by any treatment at the 24-h measurement point (D). At 1 h post-lPBN exendin-4 and CNO injection in male rats, none of the treatments alter chow intake (E). Activation of GLP-1R in lPBN by lPBN-directed exendin-4 infusion results in an anorexic response, which is potentiated by chemogenetic activation of lPBN astrocytes in overnight-fasted male rats at 3 and 24 h time points (F, G). Body weight was reduced by exendin-4 at the 24-h measurement point, but the effect was not potentiated by chemogenetic lPBN astrocytic activation (H). n = 9–13. Data are expressed as mean ± SEM. #p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3

Glutamatergic signaling is necessary for lPBN astrocyte-induced feeding suppression. Feeding suppression induced by lPBN astrocyte activation is attenuated by pharmacological blockade of NMDA receptors in male rats at 1 (A) and 5 h post-lPBN injection (B). Body weight was not altered by any treatment 24 h after injection (C). n = 8–10 Data are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ****p < 0.0001.

Figure 4

Sex-divergent effects of lPBN astrocyte activation-induced anorexia under high-fat high-sugar diet challenge. In females challenged by a choice of high-fat and high-sugar diet and chow, lPBN astrocytic activation robustly reduced palatable lard–sugar mix intake at all time points measured (A). Females in this diet context also significantly, albeit less robustly, reduced their chow intake, significant only at 12 h post-injection (B). Given the significant reductions in both sources of calories, also total caloric intake was reduced in female rats after lPBN astrocyte activation at all time points measured (C). Astrocyte activation also led to body weight gain suppression in females (D). In contrast to the results obtained in females, males challenged by a choice of high-fat high-sugar diet and chow did not reduce their palatable food intake at any time points measured in response to lPBN astrocytic activation (E). However, males did reduce their chow intake similar to females (F). Total caloric intake was also reduced in male rats after lPBN astrocyte activation, albeit less robustly than in females and only at 12 h post-injection (G). Astrocyte activation resulted in significant body weight gain suppression also in males (H). Females: n = 12 per group, males: n = 10 per group. Post-hoc tests under the line graphs refer to Controls to CNO comparisons (all DREADD-positive). Data are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. DREADD -ve: CNO control rats, DREADD-negative rats injected with CNO.

Figure 5

Modulation of anorexia induced with GLP-1R activation by lPBN astrocytes under diet-induced obesity challenge. In rats challenged by a high-fat high-sugar choice diet, GLP-1R activation in the lPBN by exendin-4 was not effective at reducing chow intake and neither was the deliberately selected subthreshold dose of CNO at 1 h (A), 3 h (B), or 24 h (C) post-injection. However, at the 24-h time point, rats treated with both exendin-4 and CNO significantly reduced their chow intake, suggesting a synergistic effect of the two treatments (C). Exendin-4 alone was also not effective at reducing lard intake at any time points measured (D–F); however, at 24 h, concomitant activation of lPBN astrocytes via CNO injection and GLP-1R with exendin-4 resulted in a potent suppression of lard intake to less than 40% of the amount consumed by control rats (F). Sucrose intake was not altered by any of the treatments at 1 h (G); at 3 h, only the combination of exendin-4 and CNO resulted in reduced intake (H). At 24 h, both exendin-4 alone and its combination with CNO resulted in a comparable suppression of sucrose intake (I). Chow, 30% sucrose solution, and lard were provided separately in order to distinguish the preference for each. None of the treatments affected water intake at any time points measured (J–L). Total caloric intake was reduced only by the combined application of exendin-4 and CNO at 1 and 3 h (M, N), while at 24 h, both exendin-4 alone and the combination reduced intake although the combination was still significantly more effective than GLP-1R activation alone (O). Body weight was also reduced by exendin-4 alone and the combination of CNO and exendin-4 to a similar extent (P). Males: n = 10. V, Vehicle; E, Exendin-4; C, CNO. Data are expressed as mean ± SEM. #p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 6

High-fat high-sugar maintenance alters the interaction of ghrelin and lPBN astrocytic activation on feeding behavior. Under obesogenic diet maintenance, ghrelin lPBN infusions alone did not affect chow intake (A–C), lard intake (D–F), sucrose intake (G–I), or total caloric intake (M–O) at any of the time points measured. Surprisingly, however, and in contrast to the attenuating effect of lPBN astrocyte activation on ghrelin’s hyperphagia in chow-maintained rats, lPBN astrocyte activation enhanced ghrelin hyperphagia at 1 and 3 h post-injections (M, N). Chow, 30% sucrose solution, and lard were provided separately in order to distinguish the preference for each. Water intake (J–L) and body weight (P) were not affected by any of the treatments. n = 9. V, Vehicle; G, Ghrelin; C, CNO. Data are expressed as mean ± SEM. *p < 0.05, **p < 0.01.