Autoimmune diseases (ADs) are characterized by the immune system mistakenly attacking the body’s own tissues, leading to chronic inflammation and tissue damage (1). The burden of ADs on patients is substantial, encompassing a wide array of chronic conditions that significantly impair quality of life, lead to long-term health care needs, and pose considerable challenges to health care systems worldwide (2).

In the context of ADs, the interplay between mitochondrial dynamics and immune responses is an area of intense investigation. Mitochondria, the powerhouses of the cell, are central to energy production and are intimately involved in a myriad of cellular processes, including apoptosis, signaling, and reactive oxygen species production (3). Their role extends beyond mere energy suppliers to being pivotal in maintaining cellular homeostasis and function. The number of mitochondria within a cell, quantified as the mitochondrial DNA (mtDNA) copy number, reflects the cell’s metabolic demand and its ability to respond to physiological stress (4). This metric has emerged as a significant biomarker for various diseases, highlighting the intricate relationship between mitochondrial function and pathological states (5). A number of previous cross-sectional studies, case-control studies reported an increased risk of developing some ADs has been reported to be associated with a reduction in mitochondrial copy number (6–9). Several observational studies have reported that mtDNA is associated with AS (10), CD, UC (11), PSO (12). To date, it is unclear whether there is a causal relationship between mtDNA copy number and these autoimmune diseases. Furthermore, it has yet to be determined whether a bidirectional relationship exists, whereby autoimmune disease might influence mtDNA copy number.

MR is an epidemiological approach based on genome-wide association study (GWAS) data that uses genetic variation as an instrumental variable (IV) to proxy for exposure. The MR approach minimizes outcome bias due to common environmental confounders and reverse causality interference, and in recent years it has been used primarily to infer causality between exposure and outcome (13). The causal relationship between mtDNA copy number and Crohn’s disease has been previously studied by the MR method. However, the relationship between mtDNA copy number and other ADs has not been established by Mendelian randomization. There is also a lack of use of MR designs to explore reverse causality. Therefore, in the present study, we performed a Mendelian randomization study to assess the causal relationship between mtDNA copy number and ADs, including SLE, RA, AS, CD, UC, T1DM, thyrotoxicosis, psoriasis.

To the best of our knowledge, this is the first study using MR to fully explore the relationship between mtDNA copy number and ADs. Our study revealed potential associations between genetically predicted mtDNA copy number and CD, JRA, RA, thyrotoxicosis and T1DM. In addition, we validated this in patients with RA and found that whole blood mtDNA copy number decreased with increasing disease activity.

Previous observational studies have explored the relationships between mtDNA copy number and various ADs including AS (10), CD (11, 24), JRA (25), PSO (12),RA (9), SS (26), SLE (6, 27–29), SSc (7, 30), thyrotoxicosis (31), T1DM (8), UC (11, 32, 33), and vitiligo (34). Studies on SLE have shown that mtDNA copy number correlates with disease activity, response to glucocorticoids, and prognosis (6, 27). Regarding SSc, observational studies have yielded conflicting results, with some showing negative correlations (30)and others showing nonsignificant correlations (7). The same is valid for PSO. Previous clinical articles have found lower mtDNA levels in patients with PSO than controls (12), other research has indicated increased mtDNA copy number in patients with PSO. Basic research has shown that mitochondrial dysfunction contributes significantly to the pathogenesis of PSO by modulating innate immunity via redox-sensitive inflammatory pathways (35, 36). A study from the UK on 60 patients with UC demonstrated that elevated mtDNA levels correlating with C-reactive protein, endoscopic severity, and disease activity. However, our results did not show a bidirectional causal relationship between mtDNA copy number and these diseases. Oxidative stress is known a key factor in the pathogenesis of CD and as a key mechanism leading to tissue and cellular damage (37, 38). A previous meta-analysis of MR studies revealed a robust causal relationship between mtDNA copy number and CD (39), which is consistent with our results. Most current clinical studies on mtDNA copy number and ADs are cross-sectional or retrospective, which making them prone to reverse causality and confounding factors. Our MR study provides ideas for the design of future clinical studies. Interventions targeting mitochondrial function may be a new strategy for treating these ADs. For example, modulating mitochondrial function could help reduce inflammatory responses and alleviate disease symptoms (40). Such investigations might pave the way for innovative clinical tests to enhancing the diagnosis and prognosis of ADs or improve patient stratification to enable more tailored clinical care approaches (41).

In recent years, there has been growing interest in the role of mitochondria in ADs, especially the influence of mitochondrial DNA on the onset and progression of such diseases. Many studies have demonstrated that alterations in mtDNA copy number contribute to developing of ADs. One way mtDNA affects autoimmunity is by alter mitochondrial function, thereby increasing oxidative stress and damaging tissues and cells. Previous basic research on RA has shown that oxidative stress plays an important role in the pathogenesis of RA (42–44). Interestingly, our study revealed stronger relationship between JRA and mtDNA copy number than with RA, providing a new direction for exploring the mechanism of JRA. Moreover, mtDNA mutations can alter cellular energy metabolism, thereby altering intracellular ATP levels and affecting the onset and progression of ADs (45, 46). The purinergic system is involved in a variety of pathological and physiological processes and plays a role in the pathophysiology of RA (47). Under physiological conditions, extracellular ATP is present at approximately 10 nM, which is regulated by nucleic acid exonucleases anchored to the plasma membrane (48). However, during pathological events, the concentration of extracellular ATP can rise significantly, reaching micromolar levels (49).. When stress induces cellular damage, ATP is released, acting as a damage-associated molecular pattern (DAMP) to initiate a pro-inflammatory response. DAMPs are an endogenous factors that specifically recruit and activate innate immune cells to promote healing and tissue homeostasis (50).In autoimmune disorders such as RA, failure to regulate DAMPs exacerbates inflammation (51). DAMPs act as ATPs that play crucial roles, through positive feedback, facilitating the early inflammatory process in RA and prolonging immune activation. Mitochondria can respond to external or endogenous stresses by triggering mitochondrial autophagy, mtDNA release, and apoptosis, and can ultimately regulating the survival or death of stressed cells (52–54). The release of mtDNA into the cytoplasm results in the expression of type I interferon (IFN) and a variety of inflammatory cytokines, including members of the TNF superfamily, interleukins and chemokines (55). Excessive release of mtDNA under sterile conditions triggers an IFN response leading to autoinflammatory conditions such as SLE (56, 57). Moreover, a study on mitochondrial transplantation showed that the in vitro transfer of healthy mitochondria into CD4+ T cells from patients with RA restored aspartate metabolism and reduced TNF production (58, 59). It has also been noted that in patients with JRA, mitochondrial-encoded DNA expression is inhibited, and the specific downregulation of mitochondrial-encoded components suggests a possible disruption of mitochondrial respiratory function, which is critical for cellular energy production and immune response (25). These studies imply that mitochondria are a potential therapeutic target that could provide more precise and effective patient treatment.

This study is the first comprehensive exploration of the reverse causality between ADs and mtDNA copy number. The onset and progression of ADs involve oxidative stress and inflammatory responses, making it mechanistically plausible that ADs may further affect mtDNA copy number. However, in the present study, we did not find any causal relationship between ADs and mtDNA copy number. This may be related to the fact that the effects of ADs on mtDNA copy number involve multiple complex pathways that are difficult to capture adequately in current studies. In reverse Mendelian Randomization studies, the relatively small number of cases in comparison to a larger pool of controls might have obstructed the identification of genetic variations linked to the disease. Such an imbalance could potentially bias the results of the study and obstruct an accurate determination of the relationship between genetic variations and the disease (60).

Our study has several advantages. First, a two-sample MR method was employed using SNPs as instrumental variables, which were less susceptible to confounding factors and reverse causality due to the random assignment during meiosis. Second, this is the first MR study to systematically assess the causal relationship between mtDNA copy number and ADs, and it is the first to report the causal relationship between mtDNA copy number and certain ADs from an MR perspective. A better understanding of the relationship between mtDNA and AD could contribute to a clearer understanding of the pathophysiology of ADs and help in the search for new biomarkers. Nevertheless, our study has several limitations. Firstly, it focused only on GWAS data from a European population, limiting its generalizability to the population. Genetic variations and environmental interactions can vary significantly across populations, which may affect the applicability of our results to non-European populations. Further studies in diverse populations are needed to confirm our findings. This will help in understanding population-specific genetic architectures and improve the precision of personalized medicine approaches. Second, due to the limitations of MR study, the second and third major assumptions are difficult to ensure precisely. Therefore, we performed MR-Egger to detect pleiotropy, which did not show pleiotropy. Third, our GWAS was derived from two samples, and data from different sources may have different measurements or definitions, which may have affected the accuracy of the results.

In conclusion, our study suggests that mtDNA copy number may affect several ADs, such as CD, JRA and T1DM. Further clinical and basic studies are needed to understand the relationship between mtDNA and ADs in the future.