Hepatic vagal afferents convey clock-dependent signals to regulate circadian food intake

Abstract

Circadian desynchrony induced by shiftwork or jet lag is detrimental to metabolic health, but how synchronous or desynchronous signals are transmitted among tissues is unknown. We report that liver molecular clock dysfunction is signaled to the brain through the hepatic vagal afferent nerve (HVAN), leading to altered food intake patterns that are corrected by ablation of the HVAN. Hepatic branch vagotomy also prevents food intake disruptions induced by high-fat diet feeding and reduces body weight gain. Our findings reveal a homeostatic feedback signal that relies on communication between the liver and the brain to control circadian food intake patterns. This identifies the hepatic vagus nerve as a potential therapeutic target for obesity in the setting of chronodisruption.

Figure 1.. Internal desynchrony induced by loss of hepatocyte REV-ERBs (HepDKO) disrupts food intake patterns.

A) Hepatocyte specific deletion of REV-ERBs was achieved by tail vein injection of male Nr1d1^fl/fl/Nr1d2 ^fl/fl mice with AAV8-TBG-Cre. The control group was created by tail vein injection of Nr1d1^fl/fl/Nr1d2^fl/fl littermates with AAV8-TBG-EGFP. B-C) Confirmation of REV-ERBs deletion in the liver at Zeitgeber time (ZT) 4, 10, 16, and 22 (n = 3–5, mean ± SEM). D) Food intake every hour over 24hrs measured in control and HepDKO animals in light:dark conditions (n = 4, mean ± SEM). E-F) Total 24h food intake and percentage of 24h food intake in the light phase were calculated in control and HepDKO animals (n = 4, mean ± SEM). B-D) Results were compared by two-way ANOVA with Šidák’s multiple comparisons test. E-F) Results were compared by Mann-Whitney U test. *p<0.05, ***p<0.001, ****p<0.0001.

Figure 2.. HepDKO alters the transcriptome of the arcuate nucleus (Arc) and nodose ganglia.

A) Heatmap representing 105 rhythm-disrupted genes in the Arc of HepDKO animals compared to control. B) The molecular clock of the Arc in control and HepDKO animals (n = 3 pool of 3–4 Arc per group per time point). C) Expression of genes for orexigenic neuropeptides in control and HepDKO animals. D) Expression of genes for anorectic neuropeptides in control and HepDKO animals. E) Heatmap representing 522 rhythm-disrupted genes in the nodose ganglia of HepDKO animals compared to control. F) The molecular clock of the nodose in control and HepDKO animals (n = 3 pool of 4 left nodose per group per time point). To identify the rhythmic genes for each tissue and condition, raw counts obtained from STAR were processed with dryR with an amplitude cut-off of greater than or equal to 1.5-fold change. Genes with chosen_model 3 or 5 were assigned as rhythm-disrupted by HepDKO.

Figure 3.. Hepatic vagotomy prevents disrupted food intake patterns from internal desynchrony induced by HepDKO.

A) Male Nr1d1^fl/fl/Nr1d2^fl/fl littermates received i.v. injections of either AAV8-TBG-EGFP to create control groups or AAV8-TBG-Cre to create HepDKO groups and at the same time underwent surgery to create Sham and HVx groups. B-C) Confirmation of REV-ERBs (Nr1d1 and Nr1d2) deletion in the liver by RT-qPCR (n = 3–5, mean ± SEM). D) Food intake patterns every hour over 24h in light:dark conditions (n = 4–6 mean ± SEM). E-F) Total 24h food intake and percentage of 24h food intake in the light phase were calculated in Sham control, HVx control, Sham HepDKO, and HVx HepDKO groups (n = 4–6, mean ± SEM). B-D) Results were compared by two-way ANOVA with Tukey’s multiple comparisons test. E-F) Results were compared by one-way ANOVA with Tukey’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Figure 4.. Hepatic vagotomy prevents disrupted food intake patterns from internal desynchrony induced by HepBKO.

A) Male Arntl^fl/fl littermates received i.v. injections of either AAV8-TBG-EGFP to create control groups or AAV8-TBG-Cre to create HepBKO groups at the same time as surgery to create Sham and HVx groups. B-C) Confirmation of BMAL1 (Arntl) deletion in the liver and its effect on REV-ERBa (Nr1d1) by RT-qPCR (n = 3–4, mean ± SEM). D) Food intake patterns every hour over 24h in light:dark conditions (n = 4 mean ± SEM). E-F) Total 24h food intake and percentage of 24h food intake in the light phase were calculated in Sham control, HVx control, Sham HepBKO, and HVx HepBKO groups (n = 4, mean ± SEM). B-D) Results were compared by two-way ANOVA with Tukey’s multiple comparisons test. E-F) Results were compared by one-way ANOVA with Tukey’s multiple comparisons test. D) Results were compared with repeated measures ANOVA. *p<0.05, **p<0.01, ****p<0.0001.

Figure 5.. Disruption of vagal afferent signaling prevents aberrant food intake patterns from internal desynchrony induced by HepDKO.

A) HepDKO animals received hepatic portal vein injections of rgAAV-hSyn-Cre two weeks prior to receiving bilateral injections to the nodose ganglia of either AAV5-flex-mCherry (control) or AAV5-flex-taCasp3 (Caspase). B) Food intake patterns every hour over 24hrs in control and Caspase groups (n = 5–8 mean ± SEM). C-D) Total 24h food intake and percentage of 24h food intake in the light phase were calculated in control and Caspase groups (n = 5–8, mean ± SEM). B) Results were compared by two-way ANOVA with Šidàk’s multiple comparisons test. C-D) Results were compared by Mann-Whitney U test. *p<0.05, **p<0.01, ***p<0.001 ****p<0.0001.

Figure 6.. HVx protects against diet-induced disruptions in food intake patterns and body weight gain.

A) Food intake every hour over 24hrs measured in Sham and HVx animals maintained on a 60% HFD in light:dark conditions (n = 4–5, mean ± SEM). B) Percentage of 24h food intake in the light and dark cycles calculated in Sham and HVx HFD animals (n = 4–5, mean ± SEM). C) Body weight gain over 12 weeks of HFD treatment in Sham and HVx animals (n = 4, mean ± SEM). A&C) Results were compared by two-way ANOVA with Tukey’s multiple comparisons test. B) Results were compared by Mann-Whitney U test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.