Gut microbiota composition modulates inflammation and structure of the vagal afferent pathway

Highlights

• Both depletion models (antibiotic depletion and germ free) led to receiver rats displaying microbiota profile characteristic of their respective donors.

• Colonization with a dysbiotic microbiota led to an increase in body weight gain and food intake.

• Microbiota composition modulates Iba1⁺ immune cell presence along the gut-brain axis

• Colonization with a dysbiotic microbiota led to a reduction on vagal innervation to the NTS associated with a loss of CCK-induced satiety.

Abstract

Vagal afferent neurons (VAN), located in the nodose ganglion (NG) innervate the gut and terminate in the nucleus of solitary tract (NTS) in the brainstem. They are the primary sensory neurons integrating gut-derived signals to regulate meal size. Chronic high-fat diet (HFD) consumption impairs vagally mediated satiety, resulting in overfeeding. There is evidence that HFD consumption leads to alterations in both vagal nerve function and structural integrity. HFD also leads to marked gut microbiota dysbiosis; in rodent models, dysbiosis is sufficient to induce weight gain. In this study, we investigated the effect of microbiota dysbiosis on gut-brain vagal innervation independently of diet. To do so, we recolonized microbiota-depleted rats with gastrointestinal (GI) contents isolated from donor animals fed either a HFD (45 or 60% fat) or a low fat diet (LFD, 13% fat). We used two different depletion models while maintaining the animals on LFD: 1) conventionally raised Fischer and Wistar rats that underwent a depletion paradigm using an antibiotic cocktail and 2) germ free (GF) raised Fischer rats. Following recolonization, receiver animals were designated as ConvLF and ConvHF. Fecal samples were collected throughout these studies and analyzed via 16S Illumina sequencing. In both models, bacteria that were identified as characteristic of HFD were successfully transferred to recipient animals. Three weeks post-colonization, ConvHF rats showed significant increases in ionized calcium-binding adapter molecule-1 (Iba1) positive immune cells in the NG compared to ConvLF animals. Additionally, using isolectin B4 (IB4) staining to identify c-fibers, we found that, compared to ConvLF animals, ConvHF rats displayed decreased innervation at the level of the medial NTS; c-fibers at this level are believed to be primarily of vagal origin. This alteration in vagal structure was associated with a loss in satiety induced by the gut peptide cholecystokinin (CCK). Increased presence of immunocompetent Iba1⁺ cells along the gut-brain axis and alterations in NTS innervation were still evident in ConvHF rats compared to ConvLF animals 12 weeks post-colonization and were associated with increases in food intake and body weight (BW). We conclude from these data that microbiota dysbiosis can alter gut-brain vagal innervation, potentially via recruitment and/or activation of immune cells.

Introduction

Obesity is a major health issue in Western countries. In the U.S., about 40% of the adult population is obese and an additional 32% is overweight [1]. Obesity raises the risk for both metabolic and psychological disorders [2,3], and the economic burden from health care costs related to obesity is estimated at $147 billion per year [4]. Underlying the development of obesity are genetic predispositions and consumption of energy-dense food such as high fat diets (HFDs) [5,6].

In rodent models, HFD feeding leads to an increase in meal size and overall intake [7]. Gut-originating satiety peptides, such as cholecystokinin (CCK), signal via the vagus nerve to trigger meal termination [8,9]. Vagal afferent neurons (VAN) are the primary sensory neurons integrating these gut-derived signals and relaying them to the nucleus of solitary tract (NTS) in the brainstem; postprandial activation of vagal fibers increases neuronal activity in the NTS [10]. HFD consumption in rodent models leads to decreased sensitivity of VAN to satiety peptides, such as CCK [11], [12], [13], [14], and intestinal nutrients [15], [16], [17], [18], as well as reduced postprandial NTS neuronal activation [12,18]. Disrupting the vagal afferent pathway is sufficient to cause hyperphagia and weight gain, identifying aberrant gut-brain signaling as a triggering factor for obesity [19].

The mechanisms by which HFD affects vagal signaling are not fully understood, however, there is evidence that inflammation may play a role in diet-driven vagal alterations. We have previously shown that leptin enhances CCK-induced satiety [20] and HFD leads to vagal leptin resistance and reduced CCK sensitivity, potentially via a suppressor of cytokine signaling 3 (SOCS3) dependent pathway [16]. SOCS3 is a negative regulator of cytokine signaling and its expression is upregulated during chronic inflammation [21,22]. Interestingly, HFD consumption leads to a rapid increase in ionized calcium binding adaptor molecule 1 (Iba1) positive cells in the nodose ganglia (NG) [23,24], where VAN cell bodies are located, and in the NTS [25]. Iba1 is a pan-marker for monocytes and microglia. Microglia are the resident parenchymal myeloid cells of the central nervous system (CNS) [26]. Microglia activation and microglia-induced inflammation are observed in several brain pathologies, and there is evidence that chronically activated microglia exert direct deleterious effects on neurons [27], [28], [29]. HFD consumption leads to microglia-mediated inflammation in the CNS [30], [31], [32] that is associated with neuronal death [32].

Increases in Iba1⁺ cells along the gut-brain axis are accompanied by alterations in vagal structure. HFD consumption leads to a decrease in isolectin B4 (IB4) staining at the level of the medial NTS. IB4 staining has been used as a proxy for vagal projections as it binds to unmyelinated c-fibers, which at the level of the NTS, are predominantly of vagal origin [33]. Additionally, caspase deletion of vagal c-fibers abolishes IB4 labeling in the medial NTS [34]. The timing of NTS vagal withdrawal coincides with onset of weight gain and hyperphagia in HFD fed rats [20]. Interestingly, in HFD fed rats, a decrease in Iba1⁺ cells presence along the gut-brain axis is associated with normalization of NTS IB4 staining and reduced food intake [25,35]. These data suggest that HFD-associated inflammation may induce structural remodeling of the vagal afferent pathway, potentially altering vagal signaling and regulation of intake.

One potent source of inflammation under direct dietary influence is the gut microbiota. The gut contains more than 10¹⁴ microorganisms, the majority of them bacteria, which can impact host physiology and behavior, including energy homeostasis [36]. There is accumulating evidence showing that HFD consumption leads to rapid unfavorable changes in microbiota composition (dysbiosis) [37,38]. HFD-driven dysbiosis is characterized by a decrease in bacterial diversity, a higher ratio between the dominant phyla, Firmicutes and Bacteroides [39], and an increase in the inflammatory potential of the microbiota [40]. Colonization of GF or microbiota-depleted animals with microbiota from obese donors leads to excessive weight gain [41], highlighting a triggering role for the microbiota in obesity. The gut microbiota may affect the vagal afferent pathway. Vagal nerve terminals located in the gastrointestinal (GI) lamina propria are ideally positioned to respond to changes in microbiota composition and/or the release of bacterial byproducts [42], and VAN express receptors for bacterial byproducts [43], [44], [45]. There is evidence that the GI bacterial population communicates with the CNS via a vagal pathway as the ability of the probiotic Lactobacillus rhamonus to protect against stress-induced anxiety is abolished by vagotomy [46]. Neurons and microglia alike can sense bacterial products. In the CNS, microglia alter synaptic function and axonal growth in response to pro-inflammatory bacterial compounds [47], [48], [49]. We have previously found that chronic administration of an obesity-associated pro-inflammatory bacterial factor, lipopolysaccharide (LPS), induces vagal leptin resistance, impairs CCK-induced satiety, and promotes intake and excessive weight gain in rats [45]. Conversely, preventing HFD-associated dysbiosis by antibiotic treatment [25] or prebiotic supplementation [35] prevents diet-associated changes in IB4 staining of the NTS [25]. Prebiotic administration in HFD fed rats reduces intake and preserves CCK-induced satiety [35].

While there is some evidence that dysbiosis is necessary for HFD-driven alteration in vagal structure and function, in this study we aimed to determine if dysbiosis was sufficient to induce vagal inflammation and alter vagal innervation to the NTS. To isolate the microbiota-driven effects from diet and establish a causal relationship between an obesity-associated dysbiotic microbiota and vagal remodeling, we used two different models of microbiota manipulation: 1) conventionally raised, immuno-competent, male Fischer and Wistar rats with microbiota depleted using antibiotic treatment and 2) immuno-deficient and microbiota-deficient GF male Fischer rats. Microbiota-depleted or GF rats were recolonized or conventionalized, respectively, with microbiota from low fat diet (LFD) fed (ConvLF group) or HFD fed donor rats (ConvHF group). We found that bacteria characteristic of each dietary treatment were successfully transferred to receiver animals. Additionally, colonization with a dysbiotic microbiota led to a significant increase in Iba1⁺ cells in both the NG and NTS and was associated with a decrease in NTS IB4 staining compared to colonization with a normal microbiota. These changes were associated with a loss in exogenous CCK-induced satiety. Finally, ConvHF animals displayed significant increases in food intake and BW compared to ConvLF rats.