The diverse roles of inflammatory factors in SLE pathogenesis have been extensively documented. For instance, IL-17 concentrations were found to be elevated in patients with SLE and positively correlated with disease activity(18–20). Consequently, IL-17 blockade represents a promising therapeutic strategy for SLE(20). Moreover, the abnormal expression of IL-10 and its receptor were shown to induce B cell hyperactivity, which contributed to SLE development(21). However, due to the limitation of traditional epidemiological studies and the complex pathophysiology of SLE, a direct causal link between inflammatory factors and SLE has not been established. Confounding factors and reverse causality are the main barriers to establishing the causal association between inflammatory factors and SLE through traditional observational studies(14, 15). Multiple factors, including the infection status, type of treatment response, rate of disease progression, and the presence of comorbidities, can cause the levels of inflammatory factors to fluctuate(6). Moreover, SLE is characterized by inflammatory reactions in multiple organs(22).
In this two-sample MR study, we found that the levels of many inflammatory factors were significant altered in patients with SLE versus healthy subjects. For instance, caspase 8 levels were positively correlated with SLE, suggesting that caspase 8 may contribute to SLE onset. Caspase 8 is a cysteine/aspartate-specific protease, which triggers the extrinsic apoptotic pathway, and is also implicated in necroptosis and pyroptosis(23, 24). To date, the role of caspase 8 in SLE has not been well documented. For instance, while one study found that caspase 8 levels were increased in SLE patients versus healthy controls(25), another study showed the opposite(26). In the present study, the MR analysis identified no causal relationship between caspase 8 level and SLE in the presence of immune cells expressing that factor; this suggests that caspase 8 contributes to SLE development through other, yet unidentified pathways, potentially independently of the immune response. Meanwhile, our findings implied that fractalkine levels were protective against SLE. Fractalkine mediates the chemotaxis and adhesion of inflammatory cells by binding to its receptor CX3CR1 on immune cells such as monocytes and lymphocytes(27). In addition, it plays important roles in inflammatory cascade initiation, which exacerbates tissue injury through the inflammatory response(28, 29). However, the role of the fractalkine-CX3CR1 axis in SLE remains controversial(30). In the present study, we found that fractalkine levels exhibited a reverse causal interference with SLE, implying a complex relationship where changes in fractalkine levels may be a consequence rather than a cause of altered immune responses in SLE. Thus, it is possible that the interaction between fractalkine and SLE is bidirectional.
IL-2 is known for its role in the development, survival, maintenance, and function of Tregs(31). In accordance, CD4+ Th cell abnormalities, which lead to insufficient IL-2 production and perturbed Treg function, are implicated in the pathophysiology of SLE(32, 33). Irrespective of the mechanism, an IL-2 shortage is considered a hallmark of SLE(34). In accordance, findings from phase II trials suggest that low-dose IL-2 therapy may be effective for treating active SLE(33, 35). Despite the association between low IL-2 levels and SLE, an MR analysis has not previously been performed to establish a direct causal link between these entities. In the present study, we used the IVW MR method to show that there was indeed a significant positive causal association between IL-2 level and SLE (P = 0.0290, β = 0.41, OR = 1.508, 95% CI = 1.040–2.185). In the subsequent mediation analysis, we examined the direct and indirect effects of IL-2 on SLE. We identified an indirect effect of 0.024, which was mediated through CD25hi CD45RA− CD4+ Tregs, implying that a IL-2 may mediate SLE pathogenesis via this new regulatory mechanism.
Our MR study identified various inflammatory factors and immunophenotypes with a causal relationship to SLE, which could potentially be targeted to treat the disease. Using this approach, we found that different markers on the same T cells could be positively or negatively associated with SLE, which further highlights the complexity of immune response in this context. For instance, TNFRSF9, caspase 8, and SLAM exhibited no causal relationship when expressed on the relevant immune cells in the context of SLE, suggesting that these inflammatory molecules contributed to SLE development through pathways other than those implicated in immune regulation. The pathogenesis of SLE is complex and multifactorial, implicating both adaptive (e.g., T cells and B cells) and innate (e.g., mitochondria, apoptosis, interferon) branches of the immune response(36). Our study included a somewhat modest panel of immune mediators (91 inflammatory factors and 731 immunophenotypes). Thus, the discovery of more SLE-associated inflammatory factors in future studies will provide new diagnostic and therapeutic targets for SLE.
In this study, we used a two-sample MR analysis to assess the causal relationship between inflammatory factors and SLE, with immune phenotypes representing a mediating factor in this relationship. A strength of this approach is that it uses genetic variation as IVs to provide a relatively clear and non-confounding framework for causal inference between exposure and outcome. In addition, we adopted robust statistical analysis techniques, such as including a sensitivity analysis and employing multiple MR analysis methods (e.g., IVW), to enhance the credibility of our results.
Despite the importance of our findings, our study had a number of limitations. First, although a two-sample MR analysis can reduce interference from confounders, it relies on strong assumptions, such as the validity of the selected IVs. Causal inference may be biased if these IVs are associated with potential confounders or influence the results through pathways other than the exposure. Second, although we performed a comprehensive GWAS data analysis of a large number of inflammatory factors to determine their relationship with SLE, the results may be limited by small sample size and low genetic diversity. For instance, the fact that our data were degenerated from a cohort of individuals of European descent may limit the generalizability of our findings. Furthermore, any inaccuracies in the selection and use of genetic IVs could affect our estimation of causality. More studies are needed validate our results and apply them in clinical practice.