Peripheral serotonergic neurons regulate gut motility and anxiety-like behavior

eTOC blurb

H. Zhang. et al. created mice lacking the serotonin-producing enzyme Tph2 in the peripheral nervous system (PNS) but with intact Tph2 in the CNS, to study the roles of serotonin produced by neurons outside the CNS. Their findings reveal that serotonin produced by peripheral serotonergic neurons impacts gut motility and anxiety-like behavior.

Serotonergic circuits in the central nervous system play important roles in regulating mood and behavior, yet the functions of peripheral serotonergic neurons are less understood. Here, we engineered mice lacking the serotonin-producing enzyme Tph2 in peripheral neurons but with intact Tph2 in central neurons. In contrast to mice lacking Tph2 in all neurons, mice lacking Tph2 in peripheral serotonergic neurons did not exhibit increased territorial aggression. However, similar to the total body Tph2 KO mice, the conditional KO animals, exhibited reduced gut motility and decreased anxiety-like behavior. These observations reveal that peripheral serotonergic neurons contribute to control of intestinal motility and anxiety-like behavior and suggest that therapeutics targeting this subset of peripheral neurons could be beneficial.

Only a small subset of neurons in the central nervous system (CNS) produce 5-hydroxytryptamine (5-HT), the neurotransmitter commonly known as serotonin. In the CNS, serotonergic neurons originate in the brainstem and ramify throughout the brain, regulating a variety of brain functions and behavior through a set of serotonin receptors¹. Although serotonergic neurons in the CNS have been the focus of much research, the production and function of serotonin by neurons in the peripheral nervous system (PNS) has received relatively little attention. Recently, single-cell RNA-seq analyses have shown that peripheral serotonergic neurons represent ~1% of sympathetic and enteric neurons^2,3. Neuronal serotonin production relies on the enzyme tryptophan hydroxylase 2 (Tph2). Mice lacking Tph2 in all neurons exhibit several behavioral disorders, including decreased anxiety-like behavior and increased territorial aggression as well as reduced gut motility⁴. The existence of enteric serotonergic neurons was initially described almost 6 decades ago⁵, and they have been implicated to play roles in gut motility⁴ and potentially depression. However, the contribution of serotonergic neurons in the PNS to social behaviors is not well-established. To answer this question, we generated a PNS-conditional knockout mouse (Tph2^fl/fl; Hand2-Cre), leveraging the Hand2 promoter, which is active in neural crest-derived cells including enteric neurons and sympathetic neurons in the PNS⁶, but not in the brain⁷, to selectively investigate the function of serotoninergic neurons in the PNS.

To corroborate the importance of peripheral neuronal serotonin in intestinal motility, we analyzed total GI transit time in whole-body and PNS-conditional KO mice. Both Tph2^−/− and Tph2^fl/fl; Hand2-Cre mice had diminished total gut motility compared to co-housed littermate controls (Figure 1A). Furthermore, the propulsion of FITC-dextran down the small intestine was significantly slower in Tph2^fl/fl; Hand2-Cre mice than in their WT littermates (Figure 1B). The results of these two independent assays demonstrate that peripheral serotonergic neurons contribute to intestinal motility. While it is clear that the PNS contributes to gut motility, the impact of CNS serotonin signaling on intestinal motility cannot be determined from these data because of differences in microbiota between the whole-body and PNS-conditional KO animals.

Figure 1. Peripheral serotonergic neurons regulate gut motility and specific aspects of behavior.

(A) Total GI transit time was measured using a carmine red assay in mice of the indicated genotypes. Each dot represents an individual animal.

(B) FITC-dextran assay of small intestinal motility in mice of the indicated genotypes.

(C) Open field test measures of total distance traveled, and time spent in the center area of the open field in mice of the indicated genotypes.

(D) Resident-intruder assay measures of the latency to the first attack, number of attack bites and cumulative attack duration in mice of the indicated genotypes.

(E) Elevated plus maze assay measures of the total distance traveled, and time spent in the open arms in mice of the indicated genotypes.

(F) Elevated zero maze (EZM) measures of time spent in the open arms in mice of the indicated genotypes.

Data shown are means ± SD. Statistical analysis was performed by two-tailed Mann-Whitney test in A-F.

Our PNS-conditional KO mice enabled us to begin to separate the function of serotonergic neurons in the CNS and PNS in the control of behavior. We assessed behaviors of Tph2^fl/fl and Tph2^fl/fl; Hand2-Cre mice to investigate whether peripheral serotonergic neurons contribute to the behavioral disorders that have been observed in the whole-body Tph2 KO mice⁸. The absence of neuronal serotonin production in mice and rats lacking the Tph2 gene is associated with heightened territorial aggression, measured in resident-intruder assays, as well as decreased anxiety-like behavior, measured in elevated plus and elevated zero maze tests. In these experiments, we first replicated reported behavioral defects in Tph2^−/− mice (Figure 1C–1F and movie S1). Co-housed littermates were used for these comparisons in order to control for potential microbiota-related effects.

As observed in whole-body Tph2 KO animals⁸, the deficiency of Tph2 in the PNS did not impact overall locomotor activity (Figure 1C), suggesting that central and peripheral serotonergic neurons have little influence on general movement. In contrast to the Tph2^−/− animals, the Tph2^fl/fl; Hand2-Cre mice did not exhibit heightened territorial aggression (Figure 1D and movie S1), consistent with the finding that aggression originates from a subset of Tph2 neurons in the brain⁹. Notably, the PNS-conditional Tph2 KO mice displayed similar increased exploration in the open arm of the elevated plus and zero maze as observed in the whole-body Tph2 KO animals (Figure 1E and 1F), suggesting that Tph2 activity in peripheral neurons plays an important role in inhibiting anxiety-like behaviors. Together, these observations suggest that peripheral serotonergic neurons control important and specific aspects of behavior and gut motility.

Overall, it is possible that the diverse phenotypes observed in the conditional KO mice, including gut motility and behavior observed here, along with the intestinal immune deficit reported in², are linked within the gut-brain axis. The observed behavioral phenotypes we found in Tph2^fl/fl; Hand2-Cre animals are unlikely to be attributable to Hand2-Cre expression in the CNS. Our analyses of gene expression databases¹⁰ did not detect Hand2 expression in CNS neurons, either during embryonic or post-natal development, arguing against the possibility of Hand2 expression in the CNS during development. In contrast, these analyses revealed that markers like Nes, Syn1, Phox2b, and Sox10, which are commonly used to identify enteric neurons, are indeed present in the CNS (Figure S1).

The overlapping and distinct behavioral patterns in whole-body and PNS-conditional Tph2 KO animals suggest that, at least in part, serotonergic neurons in the PNS control specific aspects of behavior. Territorial aggression appears to be mediated by CNS serotonergic circuits, since the PNS-conditional KO animals did not exhibit elevated aggression. Surprisingly, the Tph2^−/− and Tph2^fl/fl; Hand2-Cre animals both showed elevated exploratory behavior that is thought to reflect decreased anxiety. The PNS is heterogeneous and Tph2 is expressed in a small subset of neural crest derivatives in the enteric nervous system, as well as dorsal root ganglia and autonomic neurons. Dissecting the specific roles of subsets of serotonergic neurons in the PNS (e.g., in enteric neurons) as well as potential interactions of peripheral and central serotonergic circuits in regulating anxiety-like behaviors is of interest. This research may shed light on a potential mechanism underpinning the gut-brain axis. Nevertheless, both whole-body and conditional Tph2 KO animals exhibit reduced anxiety-like behaviors, raising the possibility that peripheral serotonergic neurons contribute to anxiety-like behaviors along with CNS serotonergic circuits. Thus, drugs that block serotonin reuptake (SSRIs) may act, at least in part, by targeting serotonin signaling in the PNS. Engineering SSRIs so that they are unable to penetrate the blood-brain barrier could prevent untoward central nervous system-related side effects of new anxiolytic agents targeting the PNS. Furthermore, defining the cellular networks and mechanisms through which peripheral serotonergic neurons regulate behavior may reveal new targets for pharmacological interventions for anxiety as well as disorders of gut motility.