AS is a chronic inflammatory disorder, and the main symptoms of this disease are inflammatory spinal pain and stiffness 1. Although the pathogenesis of AS remains unclear, studies have demonstrated that the severity and susceptibility to AS are largely determined by gene; notably, the HLA-B27 gene determines ~ 20% of the genetic risk 18. As an autoimmune disease, a previous study explored the associations between AS and mental disorders10,19; however, most of these studies are retrospective, so MR was used to explore the associations genetically. To the best of our knowledge, this study is the first MR analysis to explore the causal associations between AS and 36 common mental disorders, and we found that AS was associated with the risk for ASD, insomnia, depression, BD, delirium, anankastic personality disorder, generalized anxiety disorder, anxious personality disorder, mental and behavioral disorders due to opioids, mental and behavioral disorders due to hallucinogens, mental and behavioral disorders due to sedatives or hypnotics, SCZ, persistent delusional disorder, and sexual dysfunction.
Large population-based epidemiological studies have identified an association between ASD and a family history of autoimmune disease, including rheumatoid arthritis 20 and Crohn's disease 21. In our study, we further found that there was a positive effect of AS on ASD. While the pathogenesis of ASD is still unknown, numerous lines of evidence suggest that the microbiota gut-brain axis plays a role in the generation of autistic behaviors22. Defects in the epithelial barrier may lead to ASD and AS23. Avolio E et al. (2022) 24 found that fecal microbiota transplantation (FMT) into mice from autistic child donors colonized mice with microbiota associated with ASD and autistic behaviors, and Dan Z et al. (2020) found that the abundance of Prevotellaceae had a positive effect on ASD25. AS can also be considered a kind of microbiome-driven disease, and the best-known link is the ability of Chlamydia trachomatis and Klebsiella to cause HLA-B27-related diseases26. A study showed that compared with healthy controls, AS patients have a lower gut microbiome abundance but a higher abundance of Prevotellaceae27. Therefore, we speculate that the gut microbiota (especially the abundance of Prevotellaceae) and epithelial barrier defects may explain the association between AS and ASD. In addition, as a kind of autoimmune disease, innate immune cells such as group 3 innate lymphocytes and neutrophils are involved in the pathogenesis of AS 28. Proinflammatory cytokines are also increased in patients with AS, including interleukin-4 (IL-4), IL-6, IL-17, IL-23, tumor necrosis factor (TNF)-α, and C-reactive protein (CRP)29,30. Simultaneously, studies confirmed that administration of IL-17 to the fetal brain promotes abnormal cortical development and ASD-like behavioral phenotypes31, and some studies also regarded IL-17 as a kind of social cytokine and regarded IL-6 as an important regulator of autism-like behavior32,33. This evidence demonstrates that AS-induced immune activation interacts with the nervous system, leading to autism-like behaviors.
The inverse relationship between SCZ and rheumatoid arthritis has been confirmed by many studies34, and recently, some articles further reported that AS was associated with a lower risk of SCZ8. This study supported this negative association by the MR method and excluded the uncertainty of confounding factors. The mechanisms underlying the association between AS and SCZ are unknown, and we speculated that the high levels of CRP and IL-6 caused by AS may be an indispensable factor. An MR study conducted by Hartwig, Fernando Pires et al. (2017) showed that blockade of IL-6 canonical signaling and reduction of CRP levels were associated with increased risk of schizophrenia35. Observational studies have shown that a low level of acute phase proteins in newborns is associated with a higher risk of SCZ 36, and other prospective studies have shown that high levels of CRP in newborns are associated with maternal exposure to Toxoplasma gondii or cytomegalovirus37. Chronic maternal infection with these pathogens may ultimately lead to an increased incidence of fetal SCZ. This study also predicted that AS was associated with a lower risk of persistent delusional disorders. The symptoms of AS and persistent delusional disorders overlap in many ways; both are characterized by delusions of existence, and some studies believe that delusional disorder is currently thought to be on the same spectrum or dimension as SCZ38. Therefore, we speculated that the proinflammatory cytokines caused by AS may also explain the association between AS and persistent delusional disorders. Delirium is a fatal acute attention and cognitive impairment in the elderly (i.e., those 65 years or older). This study predicted that AS was positively associated with delirium. A previous study showed that chronic inflammation and the release of cytokines, including TNF-α, IL-1β, and IL-6, may lead to disruptive and nondisruptive blood‒brain barrier changes39. Disruption of the blood‒brain barrier, proinflammatory cytokine action on the brain, and changes in the microenvironment of the central nervous system eventually lead to the occurrence of delirium39.
Depression, BD and insomnia are three common mental disorders, and strong evidence exists for a polygenic contribution to the risk for these three mental diseases. The heritability of BD is approximately 70%, the heritability of depression is approximately 30–50%, and the heritability of insomnia is approximately 40%40–42. Previous studies have confirmed that AS is an independent risk factor for these three mental disorders. Meesters JJ et al. (2014) confirmed that during 13 years, 10% of an AS cohort compared to the expected 6% had doctor-diagnosed depression43. Leverment S et al. (2017) found that 35–90% of patients with AS suffer from poor sleep44. Sundquist K et al. (2008) confirmed that individuals with AS are at higher risk of BD45. Our study is consistent with these results, and we speculated that the following aspects may explain these associations. First, AS is a chronic inflammatory disease of unknown cause. It is now clear that proinflammatory cytokines such as IL-6, IL-1β, and TNF-α are responsible for mood changes and psychological abnormalities such as anxiety, inattention, depression and sleep disorders46, and IL-4, IL-6 and IL-10 cytokine levels are also reduced when symptoms are reduced after antidepressant treatment for depression47. In a study, as an anti-TNF-α drug, infliximab prescribed to AS patients improved the symptoms of AS as well as the depressive symptoms associated with AS48. The microbiota-brain-gut axis also plays an indispensable role in the pathogenesis of depression, BD and insomnia. AS leads to dysbiosis of the intestinal flora, and the gut microbiota influences hypothalamic‒pituitary‒adrenal (HPA) axis programming and stress responses across the lifespan49. Clinically, depression and BD are associated with dysregulation of the HPA axis50, and the hyperactivation of the HPA axis by the microbial microbiota-brain-gut axis can trigger insomnia and contribute to the development of depression and BD51. However, the ORs of the associations between AS and these three mental disorders in this study were low, which is a huge gap from the previous study. Therefore, we speculated that the individual background factors related to AS, not AS itself, are also responsible for the development of depression, BD and insomnia. By reviewing the literature, we found that a sense of loss of control is the key factor for the development of depression in AS patients 52. The physical disability and pain caused by AS make the patient unable to play a corresponding social role, and the reduction in social income leads to a decrease in the patient's sense of control, which is related to the occurrence of depression, BD and insomnia 21. This study also confirmed that AS was associated with mental disorders due to opioids, hallucinogens, and sedatives or hypnotics, which may be associated with some drug dependence.
This is the first MR study that focuses on the associations between AS and common mental disorders. The MR study minimizes the possibilities of reverse causation and potential confounding factors, which always exist in observational studies, and this study is the closest to an RCT because of its random and natural assortment of genotypes. The MR‒Egger intercept test and MR-PRESSO were used to test whether the observed causal association was influenced by horizontal pleiotropy. Cochran’s Q test was performed to estimate the heterogeneity. Multiple sensitivity analyses were conducted in this study to test the influence of horizontal pleiotropy on our causal estimates, which makes our results robust.
Our study still has several limitations. First, all the data used in this study are from the FinnGen GWAS meta-analysis, so more studies should be performed among populations other than Europeans. Second, due to the limited GWAS data used in this study, patients with AS were not further classified, which limits the exploration of the activity and severity of AS in relation to different types of mental disorders. Third, due to the lack of research on some mental disorders, this article could not explain the mechanism of all the significant associations. More research and exploration of these associations are needed.