Diabetic neuropathy (DN) is a common and burdensome complication of diabetes that is significant but often over-looked. It can markedly impair psychological functions and quality of life (1). Diabetic Peripheral Neuropathy (DPN) is the most frequently observed type of DN that affects the feet and legs. It presents a range of symptoms, including pain, numbness, and severe discomfort. autonomic dysfunction, however, affects the autonomic nervous system that regulates involuntary bodily functions. This dysfunction contributes to various problems, such as cardiovascular dysfunction characterized by blood pressure and heart rate changes, gastrointestinal dysfunction leading to gastroparesis, and urogenital dysfunction affecting bladder control and sexual function (2). DPN alone contributes to more than $10 billion in annual healthcare expenses, exceeding one-fourth of the total direct medical costs of diabetes (3). Managing DN requires a comprehensive approach that includes strict glycemic control to slow neuropathy progression, pain management and treatment of autonomic symptoms to enhance quality of life (4). Early diagnosis and thorough management are key to prevent complications and improve patient outcomes. Due to the limited treatment options for DN, it is crucial to investigate and identify new therapeutic targets (5).

Gut microbiota (GM) generally refers to the bacteria residing in the human gut. It plays an important role in regulating a wide array of physiological functions in the host and providing protection against pathogenic bacteria (6). GM is involved in processes such as digestion, immune system modulation, also influencing mood and behavior through the gut–brain axis (7, 8). The pathogenesis of microbiota dysbiosis significantly contributes to the onset and advancement of diabetes mellitus and its complications, such as cardiovascular disease, nephropathy, and DN, by promoting systemic inflammation and disrupting metabolic functions (9). However, these observational studies did not show a causal relationship between GM with DN, and it is still uncertain whether reverse causality weakens this correlation.

Mendelian randomization (MR) is a method in genetic epidemiology that uses genetic variants as instrumental variables (IVs) to assess the causal relationship between an exposure and outcome (10). Genome-wide association studies (GWASs) offer extensive datasets featuring many single nucleotide polymorphisms (SNPs) and significant sample sizes. MR leverages Mendelian inheritance laws using one or more genetic polymorphisms as the exposure variable. This makes GWAS-based MR a compelling method for determining causality (11, 12).

The two-sample MR technique offers increased statistical power to identify the causal effects between exposure factors and outcomes using published summary estimates from various large-scale GWASs (13). Also, large-scale summary statistics enable the analysis of the relationships between GM and DN by enhancing the statistical power of two-sample MR analysis. Hence, a bidirectional MR methodology was employed to investigate the potential causal association between GM and DN by combining data from the MiBioGen and FinnGen consortiums’ GWASs on genetic variations. The adoption of a bidirectional MR strategy enhanced the robustness of our findings against potential confounding variables and reverse causation. Finally, we conducted a gene ontology (GO) analysis using lead SNPs to investigate the GM’s biological impact on DN. This GO analysis offered deeper understanding of the physiological mechanisms involved. Our research opens new avenues and provides fresh insights for future DN studies.

In recent times, there has been significant focus on the gut microbiota in metabolic illnesses, specifically type 2 diabetes. This topic has been extensively explored in scientific literature (28). In addition, previous studies have used MR analysis to elucidate potential causal relationship between various biomarkers from different sources and the risk of various diseases (29–31). In this study, we utilized GM data derived from a GWAS meta-analysis conducted by the MiBioGen consortium and DN data from the R8 release of the FinnGen consortium. The causal effects of GM taxa (from phylum to genus level) on DN were investigated. We found increased levels of Christensenellaceae R-7, Ruminococcaceae UCG013 and Eggerthella groups, which may be associated with a higher risk of DN, while increased levels of Peptococcaceae and Eubacterium coprostanoligenes groups could be linked to a lower risk. Additionally, we performed a reverse MR analysis to demonstrate the causal relationships between DN and GM, and found that the risk of DN may be potentially linked with elevated levels of Anaerofilum, Dorea, Lachnospiraceae UCG-010, Ruminococcus_2, and order NB1n. The GO enrichment analysis showed a considerable enrichment of the genes involved in glycosyltransferase and hexosyltransferase activities.

There is an increasing interest in studying the harmful effects of the microbiome on various human diseases. Dysbiosis of GM may disrupt normal gut microbial activity, leading to various neurological defects (32). Similarly, previous studies in vivo have shown that transplanting dysbiotic GM from individuals with distal symmetric polyneuropathy, a prevalent neuropathy in people with diabetes mellitus, to db/db mice had accelerated the development of peripheral neuropathy (33). Ruminococcus has been reclassified as Blautia, a genus of anaerobic bacteria that play specific roles in metabolic disorders, inflammatory diseases, and biotransformation (34). Recent investigations have shown that Ruminococcus torques level was significantly elevated in the clinically diagnosed DPN group, relative to the normal or disease controls (35). Similarly, Ruminococcaceae_UCG013 may be a causative agent of DN in our study. A prospective cohort analysis showed that Eggerthella is an important risk factor for diabetic foot ulcers (36), whereas peripheral neuropathy was identified as one of the most prominent variables linked with diabetic foot ulcers (37). implying that Eggerthella may be a potential risk factor for DN. These results were consistent with our study. Previous studies have confirmed that Peptococcaceae is a protective factor for diabetic retinopathy (38). However, the role of Peptococcaceae in DN has not been previously investigated. Besides, the reverse MR analysis suggested that DN may have a causal association with the elevated levels of Ruminococcus_2. There may be a two-way causal relationship between different genera of GM and the same disease. Therefore, further investigations are needed to clarify the functional significance of specific genera of GM and explore targeted therapies for gut bacterial flora. Previous research has shown a potential genetic relationship between the Christensenellaceae R-7 group and frailty, highlighting the significance of GM in human physiology (39). Our research confirms that Christensenellaceae R-7group has been associated with an increased incidence of DN, offering a novel avenue to explore the impact of GM.

There is less research on the effect of DN on GM. GM is dynamic and mostly stable in healthy people, but it can be influenced by many disorders (40). From the perspective of metabolic and immune processes, the distribution of GM significantly differed between the patients with and without diabetes (41). In our study, increased levels of Anaerofilum, Dorea, Lachnospiraceae UCG-010, Ruminococcus_2, and order NB1n may potentially be associated with the risk of DN. Previous studies have shown that delayed neurocognitive recovery was enriched by Anaerofilum compared to the non-delayed neurocognitive recovery group (42). The study of the relationship between Dorea and insomnia (43), Lachnospiraceae UCG-010 and chronic kidney disease (44), Ruminococcus_2 and rheumatoid arthritis (45), order NB1n and gastroduodenal ulcers (46) highlight the potential for GM-focused treatments.

The precise processes through which gut bacteria influence the likelihood of developing metabolic diseases are still unknown. Prior research has shown specific factors contributing to the progression of diabetic issues are elevated levels of reactive oxygen species, chronic hyperglycemia, reduced antioxidant capacity (47) and the anti-inflammatory effects of certain bacteria (such as, Faecalibacterium in patients with DN) (48). Moreover, numerous recent studies have recognized the gut–brain axis as a pivotal mechanism for investigating the advancement of diseases. However, there is a lack of research on the connection between the microbiota and various types of pain that lack a clearly identifiable localized cause, such as DN (49, 50). In our study, we performed a GO analysis for the 13 genes to find potential mechanisms of disease pathogenesis. The analysis showed an increase in the apical plasma membrane, glycosyltransferase activity, hexosyltransferase activity, and membrane raft.

This research has several strengths. Causal inference between GM and DN was determined using an MR analysis to exclude the confounding variables and reverse causation. The genetic variations of GM were derived from the most extensive GWAS meta-analysis to ensure that the robustness of the instruments used in the MR analysis. Horizontal pleiotropy was identified and ruled out by the MR-PRESSO and MR-Egger regression intercept term analyses. The leave-one-out analysis confirmed the robustness of the results. A two-sample MR test was used, utilizing non-overlapping exposure and result summary-level data to prevent bias (51). Nevertheless, our research has some drawbacks. We chose SNPs with p< 1 × 10⁻⁵ as IVs due to the limited number of SNPs with p< 5 × 10⁻⁸. We conducted multiple IV screening processes to ensure the reliability of IVs. This included removing SNPs with an F-value< 10 to prevent bias from weak IVs and scanning all SNPs in PhenoScanner V2 to eliminate any confounding effects. This study examines the relationship between GM and DN without studying the underlying mechanism. This MR analysis can be affected by potential pleiotropy. Each exposure in our study had a minimum of three IVs, which could potentially reduce the impact of pleiotropy given that distinct IVs are unlikely to exhibit the same correlation due to pleiotropy. The genetic IVs showed a slight effect on the variances of certain microbial taxa, possibly limiting the statistical power of the association findings.

The research participants were mostly of European descent, with limited GM data collected from other ethnic groups, who were less influenced by ethnic bias. Hence, this prevents the generalizability of the findings to other groups. Therefore, future research should examine the complex interactions and communications between the host and gut bacteria to enhance our understanding of the relation between GM and illness.

In conclusion, by carrying out a two-sample MR analysis using publicly available GWAS summary-level data, we investigated the causal influence of GM on DN neuropathy and found potential flora for DN development. This work may be relevant for screening gut microbial-derived metabolites and indicators for early diagnosis of DN, which could serve as non-invasive diagnostic or therapeutic targets.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author/s.

LX: Conceptualization, Data curation, Formal analysis, Project administration, Writing – original draft, Writing – review & editing. WG: Data curation, Formal analysis, Methodology, Project administration, Supervision, Validation, Writing – review & editing. GC: Formal analysis, Funding acquisition, Project administration, Resources, Validation, Visualization, Writing – review & editing.