Activation in vagal afferents and central autonomic pathways: Early responses to intestinal infection with Campylobacter jejuni
Introduction
Peripheral immune activation induces brain-mediated functions keyed to supporting host defense including thermogenesis, neuroendocrine responses, as well as exerting profound influence on sleep, social behavior, and cognition (Kent et al., 1992, Maier, 2003). Abundant evidence now supports the idea that multiple pathways or mechanisms underlie communication from the immune system to the brain (reviewed in Dantzer, 2001, Goehler et al., 2000, Rivest, 2001). These mechanisms include active transport of cytokines (Banks et al., 2002) into the brain parenchyma, interaction of cytokines with endothelial cells of the brain microvasculature (Ericsson et al., 1997, Yamagata et al., 2001), cytokines, as well as pathogens themselves, interacting with immune cells or neurons in circumventricular organs (Lee et al., 1998, Rivest, 2001), or in barrier tissues such as choroid plexus, ventricular ependyma, and meninges. In addition to interacting with brain tissues directly, cytokines activate primary sensory neurons associated with peripheral nerves, notably but not exclusively, the vagus (Ek et al., 1998, Goehler et al., 1998, Romeo et al., 2001, Watkins et al., 1995).
The presence of a variety of mechanisms suggests that they may each contribute something different to immunosensory signaling. Although to date few studies have addressed this issue, a consensus seems to exist regarding the general circumstances in which direct brain-mediated signal transduction occurs compared to the activation peripheral nerves. That is, brain mediated immune signal transduction is dependent upon the presence of circulating mediators (such as cytokines or pathogens/pathogen products), whereas peripheral nerves are more likely to be important perhaps early on in an infection, prior to elevation of circulating cytokines, or in local infections within the terminal fields of these nerves. For example, the vagus nerve innervates most internal organs, including lung and gastrointestinal tract, and thus may signal local infection or inflammation in these tissues.
Whereas the idea that the vagus nerve senses local infection in, for instance, the gut seems plausible, this idea has never been directly tested. Studies investigating the role of vagal sensory nerves in immune-to-brain signaling have utilized bolus injection of immune stimulants, such as bacterial endotoxin, or cytokines. Peripherally injected substances, even low intraperitoneal doses, eventually lead to circulating levels (e.g., Gaykema et al., 2000, Hansen et al., 2001), which may not necessarily reflect the conditions during natural infection or inflammation. Hence, it has been difficult to determine the likely contribution of specific signaling pathways to brain responses to inflammation.
Local infections can lead to subtle changes in affective states and emotional responsiveness (Bradfield et al., 1992, Lyte et al., 1998), as well as the induction of fever and neuroendocrine responses (Campisi et al., 2004). In a model of gastrointestinal infection with Campylobacter jejuni (C. jejuni), mice developed anxiety-like behavior in the absence of systemic immune activation (Lyte et al., 1998). A subsequent study showed that this local gastrointestinal infection leads to c-Fos induction (used as a measure of neuronal activation) in a subset of central autonomic network brain nuclei also evoked by exogenous administration of immune stimulants. This effect was seen, again, in the absence of obvious systemic immune activation (Gaykema et al., 2004). Thus it is possible that a neural route, such as via vagal sensory nerves, may obtain in this paradigm.
To test the hypothesis that local infection in the gut activates vagal sensory neurons, we directly assessed expression of the neuronal activation marker c-Fos in neurons in the vagal sensory ganglia in mice treated orally either with saline or live C. jejuni. We also documented c-Fos patterns in the primary sensory relay nucleus for the vagus, the nucleus of the solitary tract (nTS), as well as induction in higher-order brain regions. C. jejuni colonizes a fairly restricted part of the gastrointestinal tract, the cecum, that is innervated by vagal sensory nerve fibers, making it an appropriate model for assessing a role of the vagus in signaling local infection.
Whereas previous experiments investigating effects of C. jejuni on the nervous system have assessed responses at time points at which the infection was established (18–42 h) in the cecum, the aims of the experiments described here were focused on detecting more rapid immunosensory events. Based on estimates of intestinal transit time in mice and pilot data (not shown), we chose a survival time window of between 4 and 13 h post-inoculation. The wide time window was used to enhance the likelihood of observing c-Fos expression in vagal sensory neurons, as this can be transient, and to determine latency, or time to onset, of vagal sensory neuron activation following inoculation with C. jejuni.
To gain clues to the potential signaling mechanism from the bacteria to vagal sensory fibers, we assessed c-Fos induction in the enteric (local intrinsic) ganglia in the cecum. In addition to innervating the submucosa and subepithelium of the gut, vagal sensory neurons also innervate these enteric ganglia. Neurons in the enteric ganglia express receptors for the cytokine IL-1, and express c-Fos immunoreactivity under conditions of infection or inflammation (Sharkey and Kroese, 2001, Sharkey and Mawe, 2002), and thus may serve as an immunosensory conduit to the vagus. In addition, to assess potential inflammation and cellular infiltration we stained sections of cecum with Cresyl violet. Because our previous study (Gaykema et al., 2004) lacked measures of circulating cytokines at the early time points investigated in this study, we also assessed levels of the cytokines IL-1β, TNFα, and IL-6 between 4 and 8 h post-inoculation, i.e., during the onset of establishment of the bacterial pathogen.
Section snippets
Animals
Male CF-1 mice (n = 66) were purchased from Charles River (Portage, MI), and upon receipt, maintained on a reversed day-night light cycle with lights off at 4:00 AM and on at 4:00 PM. They were housed in groups of two per cage (polypropylene with a bedding of wood shavings) with water and food available ad libitum, and were allowed to acclimate for at least two weeks. Animals received per oral challenge with C. jejuni (5 × 108 colony-forming units; n = 38) or saline (n = 28). All procedures were
Organization of vagal ganglia in the mouse and c-Fos induction in vagal sensory neurons
The vagal ganglia occupy the vagal–glossopharyngeal ganglion complex, which lies outside the caudal medulla bilaterally and is comprised of the nodose (inferior vagal), petrosal (glossopharyngeal), and jugular (superior vagal) ganglia. In contrast to the rat, in which the three ganglia are crudely evident, the mouse vagal–glossopharyngeal ganglion complex has the external appearance of a single flattened bag mostly within the jugular foramen.
A very low basal level of c-Fos expression was seen
Discussion
The objectives of the experiments described herein were focused on exploring a potential role for vagal sensory neurons in signaling the brain following local intestinal infection. Previous studies had demonstrated behavioral (Lyte et al., 1998) and brain c-Fos responses (Gaykema et al., 2004) in this model, providing evidence for sensitivity of the brain to local, low grade (“subclinical”) infections in the periphery. However, the route of intestine-to-brain communication was not directly
Conclusions
The finding that C. jejuni infection, which likely does not induce circulating immune-related mediators, produced widespread robust activation in the characteristic network of immune-responsive brain nuclei, supports the idea that vagal sensory neurons likely contribute significant drive on brain activation and host defense responses. The response of vagal sensory neurons to live bacterial inoculation supports the idea that peripheral nerves monitor local immune or inflammatory conditions, and
Acknowledgments
This study was supported by NIMH Grants MH50431 and MH64648. In addition, the authors sincerely thank the National Alliance for Research of Schizophrenia and Affective Disorders (NARSAD) for their generous support to Ronald Gaykema (2002 Callaghan Investigator).
References (40)
Neural mediation of food aversions and anorexia produced by tumor necrosis factor and tumors
Neurosci. Biobehav. Rev.
(1996)- et al.
The functional and chemical anatomy of the afferent vagal system
Auton. Neurosci.
(2000) - et al.
Pyrogenic signaling via vagal afferents: what stimulates their receptors?
Auton Neurosci.
(2000) Cytokine-induced sickness behavior: where do we stand?
Brain Behav. Immun.
(2001)- et al.
Subdiaphragmatic vagotomy blocks interleukin-1β-induced fever but does not reduce interleukin-1β levels in the circulation
Auton. Neurosci.
(2000) - et al.
Brain response to cecal infection with Camplyobacter jejuni: analysis with Fos immunohistochemistry
Brain Behav. Immun.
(2004) - et al.
Interleukin-1 induces c-Fos immunoreactivity in primary afferent neurons of the vagus nerve
Brain Res.
(1998) - et al.
Vagally mediated fever: a visceral chemoreceptive pathway
Auton. Neurosci.
(2000) - et al.
Sickness behavior: a new target for drug development
Trends Pharmacol. Sci.
(1992) - et al.
Temporal and spatial relationship between lipopolysaccharide-induced expression of Fos, interleukin-1β and inducible nitric oxide synthase in rat brain
Neuroscience
(1999)