Search

Abstract Background Vagus nerve stimulation has been suggested to affect immune responses, partly through a neuronal circuit requiring sympathetic innervation of the splenic nerve bundle and norepinephrine (NE) release. Molecular and cellular mechanisms of action remain elusive. Here, we investigated the therapeutic value of this neuromodulation in inflammatory bowel disease (IBD) by applying electrical splenic nerve bundle stimulation (SpNS) in mice with dextran sulfate sodium (DSS)-induced colitis. Methods Cuff electrodes were implanted around the splenic nerve bundle in mice, whereupon mice received SpNS or sham stimulation. Stimulation was applied 6 times daily for 12 days during DSS-induced colitis. Colonic and splenic tissues were collected for transcriptional analyses by qPCR and RNA-sequencing (RNA-seq). In addition, murine and human splenocytes were stimulated with lipopolysaccharide (LPS) in the absence or presence of NE. Single-cell RNA-seq data from publicly available data sets were analyzed for expression of β-adrenergic receptors (β-ARs). Results Colitic mice undergoing SpNS displayed reduced colon weight/length ratios and showed improved Disease Activity Index scores with reduced Tumor Necrosis Factor α mRNA expression in the colon compared with sham stimulated mice. Analyses of splenocytes from SpNS mice using RNA-seq demonstrated specific immune metabolism transcriptome profile changes in myeloid cells. Splenocytes showed expression of β-ARs in myeloid and T cells. Cytokine production was reduced by NE in mouse and human LPS-stimulated splenocytes. Conclusions Together, our results demonstrate that SpNS reduces clinical features of colonic inflammation in mice with DSS-induced colitis possibly by inhibiting splenic myeloid cell activation. Our data further support exploration of the clinical use of SpNS for patients with IBD.

The magnitude of innate inflammatory immune responses is dependent on interactions between peripheral neural and immune cells. In particular, a cholinergic anti-inflammatory pathway (CAP) has been identified in the spleen whereby noradrenaline (NA) released by splenic nerves binds to ß2-adrenergic receptors (β2-AR) on CD4+ T cells which, in turn, release acetylcholine (ACh). The binding of ACh to α7 acetylcholine receptors (α7-AChR) expressed by splenic macrophages inhibits the production of inflammatory cytokines, including tumor necrosis factor (TNF). However, the role of ACh-secreting CD4+ T-cells in the CAP is still controversial and largely based on the absence of this anti-inflammatory pathway in mice lacking T-cells (nude, FoxN1−/−). Using four conscious, non-lymphopenic transgenic mouse models, we found that, rather than acting on CD4+ T-cells, NA released by splenic nerve terminals acts directly onto β2-AR on splenic myeloid cells to exert this anti-inflammatory effect. We also show that, while larger doses of LPS are needed to trigger CAP in nude mouse strain compared to other strains, TNF production can be inhibited in these animals lacking CD4+ T-cell by stimulating either the vagus or the splenic nerve. We demonstrate that CD4+ T-cells are dispensable for the CAP after antibody-mediated CD4+ T-cell depletion in wild type mice. Furthermore, we found that NA-mediated inhibition of in vitro LPS-induced TNF secretion by human or porcine splenocytes does not require α7-AChR signaling. Altogether our data demonstrate that activation of the CAP by stimulation of vagus or splenic nerves in mice is mainly mediated by direct binding of NA to β2-AR on splenic macrophages, and suggest that the same mechanism is at play in larger species.