Fig 1.
QPCR analysis of Adrb2, Chrna7, and Chat expression in wild-type mice.
Samples were obtained from mice treated with saline (grey dots) or LPS (red dots; n = 5 per group). LPS was administered at 1 mg/kg (ip) and mice sacrifice 2 hours post-injection. Gene expression was normalized against GADPH. Data are expressed as mean fold changes values ± S.E.M. Average Cq values are indicated above each bar graph. (A) Adrb2 expression in the brainstem, duodenum, and spleen of saline- and LPS-treated mice. *, p = 0.0276 (unpaired t test). (B) Chrna7 expression in the brainstem, duodenum, and spleen of saline- and LPS-treated mice. (C) Chat expression in the brainstem, duodenum, and spleen of saline- and LPS-treated mice.
Fig 2.
Anatomical distribution of Chrna7 in wild-type mice.
Hybridization signals for Chrna7 (red dots) were found in neurons throughout the parasympathetic nervous system including the dorsal nucleus of the vagus (A), the nodose ganglion (B, C), and the enteric nervous system (D, E). White arrows indicate representative neuronal profiles positive for Chrna7. Note that Chrna7 expression was not seen in non-neuronal areas such as the intestinal mucosa (D) or the Peyer’s patch (F). Hybridization signals for Chrna7 were also abundant in neurons belonging to the dorsal root ganglion (G) and celiac ganglion (H), but not the spleen (I). white arrows indicate representative Chrna7-expressing neurons. Tissues were counterstained with DAPI (grey) and all images were acquired using confocal microscopy. Abbreviations: cc, central canal; DMX, dorsal nucleus of the vagus; ep, epithelium; lp, lamina propria; mp, myenteric plexus; Sol, nucleus of the solitary tract; XII, hypoglossal nucleus. Scale bar in A applies to B, D, F, G, I. Scale bar in C applies to E and the insets in G and H.
Fig 3.
Chrna7 expression is unchanged in the nodose ganglion of LPS-treated mice.
Multiplex RNAScope for Phox2b (green dots) and Chrna7 (red dots) in the nodose ganglion of saline- (A, C) and LPS-treated (B, D) wild-type mice (n = 4 mice/group). White arrows indicate representative neurons co-expressing both transcripts. White asterisks indicate neurons solely positive for Phox2b. A visual inspection of the ganglia did not reveal obvious differences in the intensity of signals or frequency of Chrna7-positive neurons. Tissues were counterstained with DAPI (grey) and all images were acquired using confocal microscopy. (E) Graphs showing the Chrna7 signal strengths after saline or LPS treatments. The number of Chrna7-positive dots per identified Phox2b-positive profile was counted. Each dot represents 1 identified profile and the total number of counted Phox2b profiles is listed below each group. The black horizontal lines represent the mean numbers of dots per profile ± SEM. The percentage of Chrna7-positive Phox2b cell is indicated above each plot. Abbreviations: NG, nodose ganglion.
Fig 4.
Chromogenic RNAScope assay for Adrb2, Chat and Chrna7 in the mouse spleen.
Chromogenic RNAScope (FastRed dots) was performed on the spleen of wild-type mice. Tissues was counterstained with hematoxylin. (A) Adrb2 positive signals were seen across most of the spleen in both the white and red pulps. (B, C) In agreement with our fluorescent assays, the spleen was virtually devoid of any signals for Chat and Chrna7.
Fig 5.
Validation of a new Chrna7 knockout mouse.
(A) Schematic representation of the transgenes used to generate Chrna7 knockout mice. (B) QPCR for Chrna7 expression in the duodenum from wildtype vs Chrna7+/− and Chrna7−/− mice (N = 6 − 9 mice per genotype). **** for p < 0.0001 (Dunnett post-hoc comparison). (C) QPCR for Chrna7 expression in the spleen from wildtype vs Chrna7+/− and Chrna7−/− mice (ns, not significant). (D) QPCR for Chat expression in the duodenum from wildtype vs Chrna7+/− and Chrna7−/− mice. *** for p < 0.001 (Dunnett post-hoc comparison with). (E) QPCR for Chat expression in the spleen from wildtype vs Chrna7+/− and Chrna7−/− mice. Gene expression was undetectable (ND) in all samples. (F) QPCR for Adrb2 expression in the duodenum from wildtype vs Chrna7+/− and Chrna7−/− mice. ** for p < 0.01 (Dunnett post-hoc comparison). (G) QPCR for Adrb2 expression in the spleen from wildtype vs Chrna7+/− and Chrna7−/− mice. ** for p < 0.01 (Dunnett post-hoc comparison). Average Ct values for each group is listed above bar graphs.
Fig 6.
Distribution of YFP in the Chat-Cre-ChR2-YFP mouse.
Confocal microcopy of YFP-stained tissues revealing the distribution of putative cholinergic structures in the ChAT-Cre-ChR2-YFP mouse. (A) In the brainstem, both YFP-positive somas and axons were observed in cholinergic nuclei containing vagal motor neurons. (B) Within the nodose ganglion, only YFP-positive axons were seen, likely originating from vagal motor neurons. (C) The GI tract contained abundant YFP immunoreactivity in enteric neurons, both submucosal and myenteric. The axons of enteric neurons traveled towards the mucosa. (D) Of note, the mucosa also contained round-shape YFP positive cells resembling immune cells (white asterisk), but only occasionally and not consistently across animals. (E) Sparse YFP-positive cells resembling immune cells were also occasionally detected in the Peyer’s patches. (F) A rich network of YFP-positive axons was observed in postganglionic sympathetic ganglia, such as the celiac ganglion, but no cell bodies. (G) The spleen was devoid of YFP-positive axons and cells. Upon visual inspection, the above distribution pattern was strictly identical in saline- and LPS-treated mice. Abbreviations: mp, myenteric plexus; sol, solitary tract nucleus; X, vagus nerve; XII, hypoglossal nerve; Amb, nucleus ambiguous.
Fig 7.
Cholinergic signaling in the gut and spleen of ChAT-Cre-ChR2-YFP mice.
(A-C) Confocal microscopic analysis of RNAscope for Chat (white dots) and Chrna7 (red dots) expression in the duodenum of ChAT-Cre-ChR2-YFP mice, with GFP-positive enteric neurons (green staining) identified using a GFP antibody. YFP-positive enteric neurons commonly co-expressed both transcripts. Tissue was counterstained with DAPI (blue). (D) Pie charts illustrating the percentage of GFP-positive cells in the duodenum that express Chat and/or Chrna7. (E-G) In the spleen of the same mice, neither YFP expression nor the transcripts for Chat and Chrna7 were detected. Scale bar in (A) applies to all images.
Fig 8.
Heat maps compiling publicly available transcriptomics data from various databases.
Common markers of cholinergic and adrenergic signaling including, but not limited to, Chrna7, Chat, and Adrb2, were assessed across different animal species (mouse, human, non-human primates) and cell types (spleen, brain, cortex, specific immune cell types). (A) The Tabula Muris compiles single cell expression profiles from various mouse cell types including splenic immune cells. Original data can be found at <https://tabula-muris.ds.czbiohub.org/ > . (B) The GTEx Portal combines RNAseq data obtained from healthy human donors including the spleen. Original data can be found at <https://www.gtexportal.org/home/ > . (C) The NHPRPR project provides RNAseq data from tissues obtained from humans (HUM) and many non-human primates species. Original data and abbreviations can be found at <https://nhprtr.org/ > . (D) The DICE database is an RNAseq resource for human immune cells. Original data can be found at <https://dice-database.org/ > . (E) The Allen Institute offers a collection of single cells data of cell types including immune cells from young and healthy human donors. Original data can be found at <https://apps.allenimmunology.org/aifi/resources/imm-health-atlas/ > . Raw values are included in each cell. In line with our biochemical and histological findings, splenocytes and immune cells were consistently enriched in adrenergic markers and impoverished in cholinergic markers.
Fig 9.
Proposed alternative model of the anti-inflammatory cholinergic pathway.
We speculate that the unilateral electrical stimulation of the peripheral end of the vagus nerve (VNS) triggers a cascade of nerve reflexes on both sides of the body. Upon the release of acetylcholine from enteric neurons, vagal efferents, and epithelial cholinergic cells, stimulation of α7nAChR-bearing sensory neurons (vagal and spinal) endings occurs. In turn, the stimulation of gut-to-brain pathways mobilizes the sympathetic outflow to the gut and spleen. Ultimately, adrenergic signaling in peripheral immune cells may contribute to suppressing immunity. The expression of β2 receptors in immune cells is otherwise well-established. For the sake of simplicity, the adrenergic supply to the gut is not indicated. In this model, and in agreement with others [51,105], cholinergic signaling in peripheral immune cells themselves plays little or no role in mediating the anti-inflammatory effects of vagus nerve stimulation. Here, we found little evidence of α7nAChR expression in the mouse spleen. Further studies are warranted to validate our speculative model including, most notably, studies with mice that lack α7nAChR only in neurons. If our hypothesis is correct, we predict such animals to be unresponsive to the anti-inflammatory actions of vagus nerve stimulation and α7nAChRs agonists. Black arrows indicate the direction of electrophysiological impulses. Abbreviations: ACh, acetylcholine; CMG, celiac-mesenteric ganglion; cX, contralateral vagus nerve; DMV, dorsal motor of the vagus; DRG, dorsal root ganglion; ENS, enteric nervous system; NG, nodose ganglion; NTS, nucleus of solitary tract; Nor, norepinephrine; iX, ipsilateral vagus nerve; sn, spinal nerves; VNS, vagus nerve stimulation; GI, gastrointestinal; α7, alpha 7 nicotinic acetylcholine receptors.