Discussion
Psychological disorders have long been related to gastrointestinal dysfunction. For example, patients with IBD have more opportunity to suffer from anxiety and depression, in contrast, children with autism spectrum disorders have a higher prevalence of CD and UC7,34,35. Evidence have now been published to support the hypothesis that intestinal disorders can lead to psychological disorders, including depression and anxiety, and vice versa . This mutual effect is mediated by bidirectional communications involving reciprocal signaling between the brain and gut. In our experiments, DSS-induced colitis invoked an abnormal stress response in mice. Compared with mice in the control group, the IBD mice were immobile for long periods and were less motivated to perform self-care activities. While immobilization reflects a reduced willingness to explore, mobilization and grooming reflect motivation and beneficial self-care. The cause of stress-related behavior in mice with colitis appears to be the activation of nNOS/ NO pathway. In a previous report, an elevation in the nNOS/NO activity in the hippocampus was shown to account for stress-induced anxiety behavior28. As a consequence of the increased nNOS/ NO activity, increase in OTR, ERK and BDNF expression may be expected in order to alleviate the stress response. In addition, previous research has demonstrated that BDNF mRNA and protein expression in the hippocampus increased after chronic restraint stress29. The increased NO level leads to neurodegeneration, the induction of acidosis, and inflammation. Morris et al . reported that the increased NO impaired mitochondrial function, contributing to depression36. In the central nervous system, nNOS neurons co-exist with OT in the hypothalamus37. Therefore, a decrease in the NO level after intranasal OT administration may facilitate recovery from intestinal inflammation. In addition to the above mechanisms, other brain regions, and other genes, may be involved in the behavioral changes observed after intranasal OT administration. However, the importance of additional mechanisms has not been explored.
In the present study, we confirmed that DSS-induced colitis could activate the HPA axis. Moreover, we show that subsequent intranasal OT administration can inhibit this HPA axis activation. Considering the anti-inflammatory properties of glucocorticoids, the observed elevation in plasma cortisol levels is likely a manifestation of the restoration of homeostasis. However, superabundant cortisol levels have been shown to increase vulnerability to psychological disorders38. Indeed, hypercortisolism is a common finding in patients with depression and anxiety 3940. Thus, the inflammatory response caused an increase of cortisol. Elevated cortisol crossed the blood-brain barrier to enter the brain, which leads to increase of negative factor (nNOS/NO) and compensatory protective factors (BDNF). Intranasal OT administration also increased the concentration of OT level in the hypothalamus. This externally administered OT may activate the hypothalamus neurons to release more OT through an autofeedback mechanism41. OT can also prevent transcription of CRH in the hypothalamus by interfering with promotor activity42. Thus, changes in the HPA axis and alleviation of intestinal inflammation may be attributed to intranasal OT. Increased serum OT caused by DSS-induced colitis may be a compensatory response of the body considering the protective role of OT in intestinal inflammation. Importantly, our results showed that increased OT expression in the hypothalamus did not result in increased OT levels in the periphery. Since we can also exclude the possibility that intranasally administered OT was released into the blood, OT must exert its influences through another pathway.
The induction and development of IBD disturbs the balance between the sympathetic and vagal nervous systems which are responsible for the maintenance of homeostasis43,44. In IBD patients, this autonomic dysfunction manifests itself in higher concentrations of serum catecholamine and low heart rate variability4,45,46. As a component of the sympathetic nervous system, catecholamine plays a crucial role in maintaining homeostasis in the gut47. Our results show a fluctuation in catecholamine levels consistent with the changes in HPA activity. We hypothesize that that intranasal OT activates neurons in the dorsal vagal complex (DVC), the primary center of the CAP that regulates gastrointestinal immunity. The DVC is known to express OTR and microinjection of OT into the DVC can enhance gut motility48-50. In future studies, we aim to investigate whether intranasal OT can enhance the activity of DVC and if so, to determine the underlying mechanism.
An elevation in the M1 and M2 macrophages in the spleen of the IBD mice and a decrease in the M1/M2 ratio have both been reported in earlier research51. The M2 macrophages, which possess anti-inflammatory activity, are known to reside in the gut and to counteract intestinal inflammation52. However, not much is known about the macrophage activity in lymphoid organs outside the gut (e.g. spleen). For the first time, we reported here that intranasal OT administration could shift the M1/M2 equilibrium in the spleen of DSS-treated mice toward the M2 type. The classic mechanism in the vago-splenic pathway involved the release of ACh from T lymphocytes residing in the spleen, Ach activation α7nAChR channels on macrophages, and inhibition of release of TNF-α53. In the gut, ACh promotes macrophage polarization to the M2 phenotype in the gut54,55. The observed decrease in TNF-α level and increase in Ach level in the spleen after intranasal OT administration are consistent with the intranasal OT activation of the vago-splenic pathway (part of the CAP). NKp46(+) NK cells have been reported to be increased in the intestinal mucosa of patients with CD compared with controls56. However, the role of splenic NKp46+NK cells in the regulation of gut inflammation is unknown. For the first time, our results indicate that DSS-induced colitis did not change the expression of peripheral NKp46+NK cells. Moreover, we specifically demonstrated that intranasal OT administration down-regulated NKP46+NK cells in the spleen. We speculate that the decrease in NKp46+NK cells in the spleen was related to activation of the vago-splenic pathway. DSS-induced colitis did reduce the percentage of Treg cells in the spleen. Similarly, patients with IBD exhibited reduced numbers of peripheral Treg cells57. Peripheral Treg cells have been reported to serve as a protective mechanism in IBD58. In a model of viral myocarditis, activation of the cholinergic anti-inflammatory pathway increased the portion of Treg cell in the spleen59. The observed increase in Treg cells in DSS-treated mice after intranasal OT administration may be related to the decrease of TNF-α and increase of ACh2760.
Splenectomy did not affect DSS-induced colitis, which means colitis is independent from spleen involvement. This observation was in accord with the hypothesis that intestinal immune tissue, but not the spleen, contributes to the development of colitis61. In the DSS-treated mice, who had undergone a splenectomy, attenuation of colitis severity or of abnormal behaviors was not observed after intranasal OT administration. Furthermore, while inflammation is initiated from the gut, regulation of circulating immune cells and the release of cytokine are mediated by the spleen. The beneficial effects of intranasal OT administration on DSS-induced colitis mice were lost following splenectomy in our mice model. Thus, intranasal OT required spleen mediation. Elimination of cholinergic anti-inflammatory efficacy in mice following splenectomy has previously been demonstrated in both DSS- and DNBS- induced intestinal inflammation, lethal endotoxemia, polymicrobial sepsis and kidney ischemia-reperfusion injury62636465. It should also be acknowledged that the vagus nerve innervates the proximal colon in addition to the spleen. Therefore, it is theoretically possible that intranasal OT administration altered enteric nervous system activity in the remainder of the colon66, and that this, in turn, may result in alterations in immune cell activities, cytokines release, and suppression of inflammation. This could also account for the effects of vagus nerve signaling on colitis that were observed in this and other studies6768.
However, the speculation about colon participation should be proven. In future studies, we aim to investigate the effects of intranasal OT administration on the enteric nervous system.
Further, our study has several limitations. Firstly, a direct correlation between changes in OT and neuron activity in the DVC has not been shown. Moreover, we did not investigate the relationship between intranasal OT administration and the enteric nervous system. Secondly, the route, timing, dosage and side effects of intranasal OT treatment should be elucidated. Finally, our study only reports data from an IBD mouse model. Few studies have investigated intranasal OT administration in IBD patients. Thus, additional research in IBD patients is required. In conclusion, our study demonstrates that intranasal OT administration significantly attenuates DSS-induced abnormal stress-related behavior and intestinal inflammation. We conclude that intranasal OT administration may provide a basis for the treatment of both exaggerated stress responses and intestinal diseases in the future.
Abbreviations
IBD: inflammatory bowel disease
CD: Crohn’s disease
UC: ulcerative colitis
HPA axis: hypothalamic-pituitary-adrenal axis
OT: oxytocin
OTR: oxytocin receptor
Treg: regulatory T
DSS: dextran sodium sulfate
WAST: water avoidance stress test
SAM axis: sympathetic-adrenal medulla axis
CRH: corticotropin-releasing hormone
ACTH: adreno-cortico-tropic-hormone
ELISA: enzyme-linked immunosorbent assay
Ach: acetylcholine
CAP: cholinergic anti-inflammation pathway
DVC: dorsal vagal complex