Anatomical Characterization of aversive visceral stimuli analyzed by iDISCO+ method
To characterize brain circuits engaged by anorexigenic signals, we used Fos, an intermediate-early gene with well-characterized activity-dependent expression[34], to identify anatomical structures in which neurons were activated following aversive delivery. We have observed that Fos expression after intraperitoneal (i.p.) injection of anorexigenic signals, such as lithium chloride (LiCl), which causes nausea and visceral malaise, induces a robust activation of Fos in the middle part of the right, but not the left insular cortex[13]. To gain a better understanding of the landscape of Fos expression in the insular cortex and throughout the brain, we used the iDISCO + method, which is a pipeline for high-speed acquisition of brain activity at cellular resolution through profiling immediate early gene expression using whole-tissue immunolabeling and light-sheet 3D imaging, followed by automated mapping and analysis of activity by an open-source software program ClearMap. Results obtained from automated segmentation of the brain regions were confirmed by inspection of both voxelized density maps and raw data (Fig. 1A). Compared with reference annotation, we observed Fos+ neurons along the entire length of the anterior-posterior (A-P) axis of the insular cortex but with a greater number within the middle part of the insular cortex (Bregma between − 0.15 and 0.26 mm) (Fig. 1A) (t12 = 6.34, P < 0.0001). Using total cell counts and peak intensity counts, we extracted digital positions of segmented cells in the insular cortex and plotted their density along the A-P axis. The density of Fos+ neuron was significantly higher in the right side of the insular cortex relative to the left side (Fig. 1B, C), especially in the Bregma of 0.1 mm. Together, our iDISCO + imaging data indicate that aversive visceral stimuli induces a robust neuronal activation in the middle part of the right insular cortex in vivo.
Lateralized activation of the insular cortex in adult female mice
Since the adult male mice were used in the previous experiment, we next examined whether similar phenomenon are observed in the female mice. We therefore injected (i.p.) LiCl into the adult female mice and found that LiCl induced Fos positive (Fos+) neurons were mainly localized within the segment between 1.26 mm before Bregma (Bregma + 1.26) and 0.02 mm after Bregma (Bregma-0.02) of the right insular cortex (Fig. 2A, B) (mIC, t14 = 2.186, P = 0.047). Notably, the activated areas were concentrated in the agranular insular cortex, dorsal part (AID) (AID, t25 = 2.297, P = 0.030) and dysgranular insular cortex (DI), but not in granular insular cortex (GI) (Fig. 2C). Visceral afferents ascending are topographically organized in the granular and dysgranular fields of the insular cortex, whereas the agranular cortex appears to receive highly integrated limbic afferents from the infralimbic cortex and the mediodorsal nucleus of the thalamus[25]. Additionally, LiCl induced substantial Fos expression bilaterally in several brain regions, including parabrachial nucleus (PBN), nucleus of solitary tract (NTS), lateral hypothalamus (LHA), and paraventricular thalamic nucleus (PVT) (Fig. 2D, E). These results indicate that the right-side middle part of the insular cortex of the female mice responds to aversive visceral stimuli, in consistence with the male mice.
No obvious LiCl-induced insular activation in the young or aged male mice
Previous studies in rodents have shown that adolescents exhibit differences from adult rodents on measures of fear-, anxiety- and depression-related behaviors and reactivity to stress[35, 36], and alteration of CREB phosphorylation and spatial memory deficits in aged mice[37]. It’s necessary to assess whether and to what extent the aversive visceral stimuli-induced insular activation is age-dependent. Our immunohistochemical data revealed that LiCl induced substantial Fos+ expression bilaterally in several brain regions, such as the central amygdala (CeA), the PVT, the LHA, the PBN, and the NTS in 3 week-old mice. The density of Fos+ neurons in the CeA and the PVT were obviously higher than that in the LHA, the PBN and NTS (Fig. 3A, C). However, the whole insular cortex was rarely activated (Fig. 3A, B). We next tested the aged mice (15 months, 15M) using similar protocol. As illustrated in Fig. 4A, I.P. injection of LiCl induced no obvious neuronal activation in the whole insular cortex of the 15 M male mice (Fig. 4A, B). An age-related reduction of Fos+ neurons expression also occurred in the CeA, the PVT, the LHA, and the NTS (Fig. 4A, C). Taken together, these results indicate that neuronal activation of the insular cortex induced by aversive visceral stimuli is age-dependent, and obvious acrtivation is only observed in the insular cortex of the adult mice, but not in that of the young and aged mice.
Intravenous injection of LiCl induced no obvious Fos expression in the insular cortex
To identify the possible pathways involved in the LiCl-induced insular activation, we first focused on the blood circulatory system. Lithium has the anti-convulsant effects both in human and in rodents[38, 39], and the intravenous route of delivery is the most efficient means of delivering substances to animals because it bypasses the need for solute absorption[40]. Here, we intravenously injected LiCl (15 mg/kg) via tail vein, and observed that intravenous administration of LiCl resulted in a specific distribution of Fos+ activity throughout the mice brains, such as dense labeling was observed throughout the PVT, moderate numbers of Fos+ neurons were observed in the PBN, whereas, scattered labeled cells were observed within CeA. It’s noteworthy that extremely few Fos+ neurons were observed in the insular cortex from anterior to posterior (Fig. 5B, C). These data suggest that lateralized activation of the insular cortex is not due to the LiCl in the circulatory system.
The Effect of Subdiaphragmatic Vagotomy on Fos Expression
We next asked whether vagal nerve afferent pathway is necessary for the activation of the insular cortex in response to i.p. injection of LiCl. We detected the number of the Fos+ neurons in the insular cortex of the mice with or without subdiaphragmatic vagotomy (Fig. 6A) and observed no obvious LiCl-induced activation of the insular cortex in the mice with subdiaphragmatic vagotamy compared with the sham group (Fig. 6B, C). However, we still observed Fos+ neurons in some brain regions such as the PVT, the CeA, and PBN (Fig. 6B, D). These data indicate that the vagal afferent pathway is required for the activation of the insular cortex induced by aversive visceral malaise.