SLE is a polysystemic autoimmune disease characterized by the formation of autoantibodies and inflammation in multiple organs (Zhu et al. 2016). Inflammation is mainly mediated by cytokines and chemokines, which could cause a persistent response in SLE. Although the relationship between SLE and cytokines has been extensively recognized, it remains a lack of systemically investigation.

In this study, we combined MR analysis and observational study to explore the relationship between the cytokines levels and SLE, and some of the results were consistent with previous study (Xiang et al. 2023). It is worth noting that we used cellular transcriptomic data in an observational study and cannot fully represent the levels of circulating cytokines. However, the two have a certain degree of relationship, and the trend similar to the MR analysis results also provides some evidence for our results. The expression of IL18 exhibited the most dramatic difference in both MR and observational analysis. IL-18 is a proinflammatory cytokine and can induce the production of INFγ in mature T helper 1 cells, thus affecting the development of SLE (Favilli et al. 2009). In MR analysis, we found a positive association between genetically determined circulating level of IL-18 and an increased risk of SLE. This was consistent with the previous studies. The serum level of IL-18 was significantly higher in MRL/lpr mice compared to normal mice (p < 0.028) (Esfandiari et al. 2001). A cross-sectional study of 113 SLE patients and 50 healthy controls identified a higher mean level of serum IL-18 among the cases than controls (p < 0.001) (Jafari-Nakhjavani et al. 2016). Besides, we found that the expression of IL18 was highly positively correlated with the severity of SLE, and it should also have a high predictive value for LN. A clinical study has indicated that patients with active renal disease had statistically significantly elevated serum IL-18 levels, and the association between serum IL-18 and active renal disease was confirmed in multivariable analysis after adjusting for ethnicity and organ damage (Mende et al. 2018). And a research included 133 lupus patients and 44 healthy subjects through renal biopsy also showed that the expression level of IL-18 in LN was increased, especially in LN-IV, and its expression site was mainly in the glomerulus (Calvani et al. 2004). However, further work is warranted to determine whether targeting IL18 can treat SLE. At present, the main treatment options include glucocorticoids and nonselective immunosuppressants. The long-term administration of these drugs is associated with numerous side effects. Thus, what we need to do is find safe, effective drugs, especially those that target IL18. Studies indicated that obtaining lower toxic and effective compounds from natural products as pro-drugs is viable. Therefore, we identified four potential compounds, which may have good binding activity with IL18. Among them, luteolin and procyanidin have been proven in animal experiments to alleviate inflammation of lupus nephritis (Ding et al. 2023; He et al. 2018). Although these studies did not explicitly point to IL8, current research has shown that some natural products may have multi-target effects; therefore, they still have important research value in the future. Meanwhile, current studies showed that the four compounds all have been demonstrated to exert anti-inflammatory effects (Yi 2018; Zhang et al. 2022; Vilahur et al. 2019; Martinez-Micaelo et al. 2012). Therefore, further exploring the function of them in SLE treatment may be a good direction.

Besides IL18, we also identified circulating levels of two cytokines (CTACK and GRO-a) to be significantly associated with the risk of SLE and found suggestive associations of circulating eotaxin, MIG, and HGF with SLE risk. CTACK is a member of the CC chemokines family which regulates T cell–mediated skin inflammation by the CTACK-CCR10 interactions (Homey et al. 2002). However, the current evidence was inconsistent with our findings. A case–control study of 32 SLE patients and 31 healthy controls reported no significant difference in serum levels of CTACK between the two groups (Hayakawa et al. 2005). Due to the relatively small size in the observational study, the lack of accuracy may be related to insufficient statistical power. Thus, the association between CTACK and SLE should be deeply researched. Furthermore, GRO-a can induce neutrophilic inflammation and thus may play a vital role in the development of SLE (Kuo et al. 2012). Epidemiological evidence for this association is inconclusive. While a case–control study of 15 SLE patients and 35 healthy controls found no difference on serum level of GRO-a in SLE patients compared with controls (Furuse et al. 2003), another case–control study of 40 SLE patients and 40 controls found a statistically significantly higher plasma concentration of GRO-a in patients than in control (Lit et al. 2006). Additionally, the directionality of the suggestive association between the three cytokines (i.e., eotaxin, MIG, and HGF) and the risk of SLE are consistent with the results from previous observational studies (Capuano et al. 2006; Idborg et al. 2018; Bauer et al. 2006).

In this study, we also found that the circulating levels of ten cytokines changed significantly in SLE patients compared to healthy. Among them, IL13 can regulate several subtypes of Th cells and affect their transformation, which play a vital role in the pathological processes of SLE (Yi et al. 2017). Our result showed that the serum IL13 concentration correlates with the severity of SLE, and the current study supports this finding (Mao et al. 2019). HGF is a multifunctional cytokine. There is clinical a study shown that HGF serum concentration is associated with the SLE activity score (Robak et al. 2001). Meanwhile, literature reported that IFN-α-induced chemokines, such as MCP1 and IP10, can attract monocytes, macrophages, T cells, and NK cells, leading to exacerbation of SLE (Geneva-Popova et al. 2022). In previous studies, MIP1β was increased in SLE patients (Ghafouri-Fard et al. 2021). However, this study presented the different trends in mild or moderate activity SLE.

At present, renal disease still is one of the most common causes of death in SLE patients. Early diagnosis, early intervention, and timely protective treatment of the kidneys are essential to improve the prognosis of SLE. Besides IL18, several cytokines may play a role. A previous study showed that upregulated CXCR4 expression on circulating B cells in active SLE may enhance their chemotactic response toward SDF1α, which may promote infiltration of these cells into inflamed renal tissue (Hanaoka et al. 2015). Another study found that the levels of IL4 were significantly lower in patients with LN when compared to SLE patients with no renal involvement (Calvani et al. 2004). Our results supported these to some extent. As for IL8 and TNF, the current results are inconsistent (Jakiela et al. 2018; Ruchakorn et al. 2019; Park et al. 2020). In current clinical studies, anti-TNFα therapies had limited efficacy in LN and better efficacy in lupus arthritis (Lorenzo-Vizcaya and Isenberg 2021). This may to some extent indicate that the relationship between TNF and LN is worthy of further investigation.

There exist a couple of limitations in the research. First, the participants included in MR and bioinformatics were restricted to European; the results may not be generalized to the population with other ancestries. Second, the protein quantification may be unpredictably altered due to qualitative protein changes. Third, bioinformatics analysis may require a larger sample size to reduce possible bias in the results.