Research Article

Mendelian Randomization Analysis of NETs-Associated Inflammatory Traits and Type 2 Diabetes and its Complications

DOI:

10.3791/71230

⸱

August 21st, 2026

 ,  ,  ,  ,  ,  ,  ,  , 
1Department of Laboratory Medicine, Zhongshan People's Hospital, 2Department of Advanced Diagnostic and Clinical Medicine, Zhongshan People's Hospital

In This Article

Summary

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

This two-sample MR study identified potential associations of genetically predicted NETs-associated inflammatory traits with T2DM complications. Genetically predicted IL-6 was inversely associated with diabetic CAD, while NETs were associated with renal and peripheral circulatory complications in T2DM. Sensitivity analyses generally supported the stability of these significant associations.

Abstract

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Neutrophil extracellular traps (NETs) -associated inflammatory traits play a significant role in type 2 diabetes mellitus (T2DM) and its complications. Notably, IL-6, a key inflammatory cytokine, is intricately linked to the formation of NETs and the pathogenesis of T2DM and its complications. This study aimed to explore the causal association between NETs-associated inflammatory traits and T2DM, as well as its complications, using a Mendelian Randomization (MR) approach. This study utilized a two-sample MR design with data from Genome-Wide Association Studies (GWAS), comprising a large European population-based meta-analysis for T2DM and its complications. The primary method of analysis was the inverse variance weighted (IVW) approach, complemented by MR-Egger regression, weighted median, and weighted mode methods. Sensitivity analyses included MR-Egger, MR-PRESSO, Cochran’s Q, and leave-one-out methods to assess the robustness of the findings. The study indicated that genetically predicted levels of interleukin-6 (IL-6) were inversely associated with diabetic coronary artery disease (CAD) (OR = 0.8997, 95% CI: 0.8257–0.9803, P = 0.0158). Additionally, NETs showed significant associations with T2DM with renal complications (OR=0.97, 95% CI 0.9428–0.998, P = 0.0358) and T2DM with peripheral circulatory complications(OR = 1.0342, 95% CI 1.002–1.0673, P = 0.037). The significant IVW associations showed no evidence of heterogeneity or horizontal pleiotropy. This study suggests that genetically predicted NETs-associated inflammatory traits are associated with specific T2DM complications. Genetically predicted IL-6 was inversely associated with diabetic CAD, whereas NETs were associated with renal and peripheral circulatory complications in T2DM.

Introduction

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Type 2 diabetes mellitus (T2DM), characterized by insulin resistance and often accompanied by progressive beta-cell dysfunction, has reached epidemic proportions worldwide1,2. Including cardiovascular disease, renal failure, neuropathy, and retinopathy, these complications of T2DM impose a substantial burden not only on individuals in terms of their health and quality of life but also on healthcare systems worldwide, straining resources and increasing costs, and on economies3,4. Considering the multifaceted nature of T2DM, it necessitates a nuanced exploration into its etiological factors. Accumulating evidence has implicated neutrophil extracellular traps (NETs) in the pathophysiology of various metabolic disorders, especially associated with T2DM.

Neutrophil extracellular traps (NETs) are extracellular networks of decondensed DNA and granule proteins that possess potent microbicidal properties5,6. Markers such as cell-free DNA and citrullinated histone H3 serve as key indicators of NETosis, reflecting inflammatory activity in various diseases7,8. Accumulating evidence suggests that diabetes promotes the formation of NETs, thereby propagating inflammation and tissue injury9,10. Clinically, elevated NETs markers correlate with hyperglycemia and adverse outcomes across multiple diabetic complications. For instance, increased dsDNA levels are linked to glucose dysregulation in STEMI patients11, while NETs components are associated with the progression of diabetic retinopathy12 and predict amputation risk in diabetic foot ulcers13. Despite these associations, the precise causal relationship between NETs and the development of T2DM and its complications remains to be fully elucidated. Accumulating evidence highlights a bidirectional interplay between IL-6 and NETs in metabolic inflammation. Hyperglycemia primes neutrophils for NETosis, with IL-6 serving as a potent inducer of this energy-dependent process14. Clinically, IL-6 independently predicts elevated NETs markers (e.g., H3Cit and cell-free DNA) in type 2 diabetes, which correlate strongly with prothrombotic states and microvascular complications like diabetic retinopathy15,16. Furthermore, infection-induced NETosis can enhance IL-6 trans-signaling, establishing a reciprocal regulatory loop17. Collectively, these findings underscore the tightly coupled pathophysiology of IL-6 and NETs, particularly within the diabetic milieu.

Mendelian Randomization (MR) is a powerful statistical method that leverages naturally occurring genetic variants as instrumental variables to estimate causal effects of modifiable exposures on health outcomes, offering insights into complex disease pathways18. This approach, rooted in Mendel’s laws of inheritance, allows researchers to infer causality in observational studies by exploiting the random assortment of alleles during meiosis, thereby circumventing some of the limitations of traditional epidemiological methods in establishing cause-and-effect relationships19.

To date, there have been limited reports on MR studies investigating the causal relationship between genetically predicted NETs-associated inflammatory traits and T2DM and its complications. Therefore, the present study aims to explore the potential causal associations between NETs-associated inflammatory traits and T2DM, as well as its complications, using a two-sample MR analysis method.

Protocol

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Study Design

This research utilized a two-sample MR approach to explore the causal relationship between NETs-associated inflammatory traits and T2DM and its complications. The MR framework relies on three key assumptions20: 1) the genetic variant must be robustly associated with the exposure, 2) the genetic variant must only influence the outcome through the exposure (exclusion of pleiotropy), and 3) the genetic variant must not be associated with any confounding factors. For a detailed overview of this study’s process, please refer to Figure 1. The manuscript adheres to the MR-STROBE guidelines for reporting MR studies, ensuring rigorous standards of transparency and reproducibility21.

Data Sources

The Genome-Wide Association Studies (GWAS) data utilized in this MR analysis were sourced from the GWAS public databases. In terms of the summary-level GWASs concerning T2DM, incorporating the largest meta-analysis of a European population, comprising 74,124 cases and 824,006 controls, alongside a prospective nested case-cohort study conducted in Europe, featuring 9,978 cases and 12,348 controls22. For diabetic complications, including T2DM with renal complications and T2DM with peripheral circulatory complications, this study retrieved GWAS summary statistics from studies in which cases were T2DM patients with the specific complication and controls were individuals without T2DM. Detailed information on data sources can be found in Supplementary Table 1.

GWAS data for NETs-associated inflammatory traits were obtained from the GWAS catalog (see Supplementary Table 2). To address the heterogeneity in biological specificity, this study categorized the included exposures into two distinct groups based on their functional roles in neutrophil biology: (1) Core NETosis-related Factors: This category includes markers directly involved in the structural formation of NETs or the enzymatic process of chromatin decondensation. Specifically, this study included NETs themselves, Myeloperoxidase (MPO), Neutrophil Elastase (NE), and the MPO-DNA complex. MPO and NE are essential enzymes for histone degradation and chromatin decondensation, while the MPO-DNA complex is a specific surrogate marker for NETs23. (2) NETs-associated Inflammatory Mediators: This category comprises cytokines and mediators that act as upstream regulators or downstream effectors closely linked to NETosis but are also involved in broader inflammatory pathways. This group includes Interleukin-6 (IL-6)24, Tumor Necrosis Factor-alpha (TNF-α)25, Neutrophil gelatinase-associated lipocalin levels (NGAL)26, and Cellular Communication Network Factor 1 (CCN1)27.

The data used in this study were sourced from open-access databases or GWAS published in previous studies; therefore, ethical approval was not required for this research.

Selection of NETs-Related Genetic Instruments

Given that NETs formation represents a dynamic biological process not directly measured in conventional GWAS, this study employed a gene-based approach to identify genetic instruments for NETs-associated inflammatory traits. A comprehensive set of 257 genes known to be critically involved in NETs formation and regulation, such as MPO, along with genes encoding histones and neutrophil granule proteins, was curated from established mechanistic studies and published literature28.

Instrumental Variable Selection

To identify robust genetic instruments for NETs-associated inflammatory traits, this study implemented a sequential filtering workflow using GWAS summary statistics. The specific operational steps were as follows:

Initial Screening: Genetic variants significantly associated with the exposure traits were extracted29 based on a significance threshold of P < 5 × 10⁻6.

Minor allele frequency (MAF) Filtering: To ensure statistical power, Single Nucleotide Polymorphisms (SNPs) with a MAF ≤ 0.01 were excluded30.

Linkage Disequilibrium (LD) Clumping: To mitigate linkage disequilibrium (LD) confounding, independent SNPs were selected using the clumping function with parameters set to r2 < 0.001 within a 10,000 kb window31.

Instrument Strength Assessment: The strength of each remaining IV was quantified using the F-statistic, calculated as F = R2 × (N-2) / (1-R2). Only SNPs with an F-statistic > 10 were retained to minimize weak instrument bias32.

Proxy SNP Identification and Data Harmonization

To address missing SNPs in the outcome dataset, this study performed a proxy search and data harmonization process:

Proxy Substitution: When a target SNP from the exposure GWAS was unavailable in the outcome GWAS, this study utilized the LDproxy() function based on the 1,000 Genomes Project European reference panel. A candidate proxy SNP was selected only if it exhibited high LD with the original SNP (r2> 0.8). If no qualified proxy was found, the SNP was excluded.

Harmonization: Exposure and outcome datasets were aligned using the harmonise_data() function from the TwoSampleMR package (R version 4.0.5). This study set the parameter action = 2 to automatically align all SNPs to the forward strand and remove palindromic SNPs with ambiguous strand orientation.

Verification: Post-harmonization, this study manually inspected the harmonized dataset to confirm that the effect allele frequencies were consistent between the exposure and outcome data.

MR Analysis and Sensitivity Tests

Causal inference was conducted using the TwoSampleMR package (R version 4.0.5).

Primary and Secondary Analyses: This study applied the Inverse Variance Weighted (IVW) method as the primary approach33. Complementary analyses were performed using MR-Egger34, weighted median, and weighted mode methods to ensure robustness35.

Sensitivity Assessments: Heterogeneity among IVs was evaluated using Cochran’s Q test36 via the mr_heterogeneity() function. Horizontal pleiotropy was assessed using the MR-Egger34 intercept test (mr_pleiotropy_test()).

Outlier Detection: This study employed the MR-PRESSO package to detect potential outliers37. The mr_presso() function was executed with 1,000 simulations. Identified outlier SNPs (P < 0.05) were removed, and the causal estimates were recalculated to verify result stability. Additionally, a leave-one-out analysis was performed to ensure that the causal association was not driven by a single SNP38.

Statistical Correction

To account for multiple testing, P-values obtained from the MR analyses were corrected using the False Discovery Rate (FDR) method. This was implemented using the P.adjust() function in R with the parameter method = "fdr". Associations with a corrected P-value (PFDR) < 0.05 were considered statistically significant.

Results

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Selected Instrumental Variables

This study conducted MR analysis using NETs-associated inflammatory traits as the exposure. A total of 106 IVs were identified for this purpose, with details provided in Supplementary Table 3. The F-statistics of these IVs had an average value of 26.46, ranging from a minimum of 20.6 to a maximum of 250.4.

When adopting T2DM and its various complications (diabetic ketoacidosis, renal complications, ophthalmic complications, neurological complications, and peripheral circulatory complications) as the outcomes, information for three SNPs could not be matched in the summary data. Proxy SNPs were utilized in this study (Supplementary Table 4).

Causal association of NETs-associated inflammatory traits on T2DM and its complications

The primary findings from the IVW method are presented in Table 1, while Supplementary Table 5 details the complementary MR estimates derived via the MR-Egger, weighted median, and weighted mode approaches. Notably, the IVW results revealed statistically significant associations between interleukin-6 (IL-6) levels and diabetic CAD (OR = 0.8997, 95% CI 0.8257–0.9803, P = 0.0158), consistent with the result of the weighted median method. These findings suggested that IL-6 was genetically associated with cardiovascular complications of diabetes. In addition, the results of the IVW analysis indicated that NETs were significantly associated with T2DM with renal complications (OR = 0.97, 95% CI 0.9428–0.998, P=0.0358) and T2DM with peripheral circulatory complications(OR = 1.0342, 95% CI 1.002–1.0673, P = 0.037). The scatter plot and forest plot for the positive results mentioned above are shown in Figure 2.

The MR Egger regression results indicated no evidence of horizontal pleiotropy in the analysis, as detailed in Table 2. Upon the heterogeneity results presented in Cochran’s Q test (Table 2) and Funnel plots (Figure 3A–C), it is observed that none of the P values for the Q statistic are below the conventional threshold of 0.05, indicating no significant heterogeneity among the IVs for the main outcomes of interest in this study.

Furthermore, the MR-PRESSO analysis (Table 3) identified an outlier (rs139237904) when CCN1 was considered as the exposure and T2DM as the outcome; the results remained non-significant after its exclusion. Similarly, when MPO was the exposure and T2DM the outcome, two outliers (rs10916508 and rs6544660) were detected, and the non-significant results persisted post-removal. For NGAL as the exposure with T2DM as the outcome, one outlier (rs10497219) was found, and the negative results were upheld after its exclusion. In the case of TNF-α as the exposure and T2DM with ketoacidosis as the outcome, a single outlier (rs10103048) was present, and the exclusion did not alter the non-significant outcome. However, when NGAL was associated with T2DM and neurological complications, two outliers (rs10497219 and rs1287048) were identified, and after their removal, the results turned positive, indicating a significant association. The ‘leave-one-out’ analysis (Figure 3D–F) provides further evidence of the robustness of the findings. In summary, this MR analysis suggests a potential role of IL-6 and NETs in the etiology of diabetic complications, warranting further investigation.

DATA AVAILABILITY:

All data generated or analyzed during this study are included in this article and Supplementary Table 1, Supplementary Table 2, Supplementary Table 3, Supplementary Table 4, and Supplementary Table 5.

figure-results-1
Figure 1: Study design in this MR study. (A) MR analyses depend on three core assumptions. (B) Sketch of the study design. NETs: neutrophil extracellular traps; T2DM: type 2 diabetes mellitus; SNP: single nucleotide polymorphism; MR: Mendelian randomization; IVW: inverse variance weighted; MR-PRESSO: Mendelian Randomization Pleiotropy RESidual Sum and Outlier Please click here to view a larger version of this figure.

figure-results-2
Figure 2: Causal association of NETs-associated inflammatory traits on T2DM and its complications. (A) Scatter plot showing the association between IL-6 levels and diabetic CAD. (B) Scatter plot showing the association between NETs and T2D with renal complications. (C) Scatter plot showing the association between NETs and T2D with peripheral circulatory complications. (D) Forest plot of IL-6 levels in diabetic CAD. (E) Forest plot of NETs on T2D with renal complications. F: Forest plot of NETs on T2D with peripheral circulatory complications. Please click here to view a larger version of this figure.

figure-results-3
Figure 3: Results of the sensitivity analysis. (A) Funnel plot of IL-6 levels in diabetic CAD. (B) Funnel plot of NETs on T2D with renal complications. (C) Funnel plot of NETs on T2D with peripheral circulatory complications. (D) Leave-one-out sensitivity analysis of IL-6 levels on diabetic CAD. (E) Leave-one-out sensitivity analysis of NETs on T2D with renal complications. (F) Leave-one-out sensitivity analysis of NETs on T2D with peripheral circulatory complications. Please click here to view a larger version of this figure.

Table 1: Primary causal association of NETs-associated inflammatory traits on T2DM and its complications Please click here to download this Table.

Table 2: Results of heterogeneity and pleiotropy of instrumental variables Please click here to download this Table.

Table 3: Results of MR-PRESSO Please click here to download this Table.

Supplementary Table 1: Data source of T2DM and its complications Please click here to download this file.

Supplementary Table 2: Data source of NETs-associated inflammatory traits Please click here to download this file.

Supplementary Table 3: Selection of instrumental variables (IVs) Please click here to download this file.

Supplementary Table 4: List of proxy SNPs used in the MR analysis Please click here to download this file.

Supplementary Table 5: Complementary MR estimates for the causal associations between NETs-associated inflammatory traits and T2DM complications Please click here to download this file.

Discussion

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

This study employs a two-sample MR approach to investigate the associations between NETs-associated inflammatory traits and T2DM, as well as its complications. The primary findings suggest a statistically significant inverse association between genetically predicted IL-6 levels and diabetic CAD. This finding may reflect the complex genetic architecture of IL-6 signaling pathways rather than a direct protective effect of circulating IL-6 on cardiovascular complications of diabetes. Additionally, this study reveals associations between NETs and various diabetic complications, including renal and peripheral circulatory complications in T2DM patients. These associations highlight the importance of NETs in the pathogenesis of diabetes-related complications and may inform future therapeutic strategies targeting these inflammatory pathways.

The association between IL-6 and diabetic CAD has been a subject of considerable debate, with studies presenting varying perspectives on the role of IL-6. This study suggests that IL-6 may have a complex, negative association with diabetes-related cardiovascular complications. Cupido et al.39 corroborate these findings, linking IL-6 signaling perturbation to reduced cardiometabolic disease risk. Conversely, Indumathi et al.40 observed IL-6 hypomethylation and overexpression in diabetic subjects, suggesting a pathogenic role in CAD. This aligns with Shrivastav et al.41 and Chittaragi et al.42, who identified elevated IL-6 levels as pro-inflammatory drivers and prognostic markers in diabetic CAD. However, Liu et al.43 highlight the dual nature of adipokines like IL-6 in atherosclerosis; this context-dependent duality likely explains the conflicting evidence across studies. The discordance in the findings could be attributed to several factors. First, the heterogeneity in study designs, patient populations, and the methods used to measure IL-6 levels might lead to varying conclusions44. Second, the temporal relationship between IL-6 and CAD development is complex and could involve both causal and reactive components, which might not be fully captured in cross-sectional or even some longitudinal studies45.

While the present study and several others suggest a significant association between IL-6 and diabetic CAD, the exact nature of this relationship remains complex and may be influenced by various biological and methodological factors. It is worth noting that this MR analysis indicates a statistically significant negative association between genetically predicted IL-6 levels and diabetes-related CAD (OR=0.8997), which contrasts with the conventional view that IL-6 promotes cardiovascular pathology. This discrepancy may be attributed to the complex dual role of IL-6 in inflammation and tissue repair46, or to the possibility that the genetic instruments used in the MR analysis reflect a specific biological pathway of IL-6 signaling that differs from its overall systemic effect. The negative association observed in this study should not be simply interpreted as indicating that IL-6 has a cardioprotective effect; rather, it likely reflects the complex genetic mechanisms underlying IL-6 receptor signaling. Alternatively, the negative association observed might be influenced by compensatory mechanisms or negative feedback loops within the inflammatory network that are not fully captured in conventional observational studies. Future research should aim to elucidate these complexities through well-designed longitudinal studies and by considering the multifactorial nature of CAD pathogenesis.

The role of NETs in diabetic complications, particularly in the context of renal disease, has been increasingly recognized as a significant mediator of inflammation and tissue damage. Magaña-Guerrero et al.12 link NETs to hyperglycemia and renal failure in diabetic retinopathy, positioning them as markers of disease progression. In the context of diabetic kidney disease (DKD), Liu et al.47 demonstrate that midkine upregulation promotes kidney injury by enhancing the formation of NETs, while Zheng et al.48 show that NETs induce glomerular endothelial dysfunction and pyroptosis, effects attenuated by DNase I treatment. Beyond these experimental findings, recent MR and transcriptomic analyses have identified specific NETs-related genes (e.g., CLIC3, GBP2) as causal risk factors for proliferative diabetic retinopathy28. Interestingly, this genetic evidence stands in stark contrast to these findings of a negative association with renal complications, highlighting the tissue-specific mechanisms of NETs. Collectively, the study complements this landscape by demonstrating a significant causal association between NETs and peripheral circulatory complications in T2DM. Further research is essential to unravel the intricate interplay of factors that could influence the role of NETs in diabetic nephropathy.

Although this study identified a potential association between IL-6 and diabetic CAD, as well as a significant correlation between NETs and other diabetic complications, the interaction between these two inflammatory mediators warrants further investigation. Although this study did not conduct a dedicated MR analysis to test the causal pathway from IL-6 to NETs, a substantial body of experimental evidence suggests that IL-6 may act as an upstream driver of NETs-associated inflammatory traits.

Biologically, IL-6 is a pivotal cytokine in the acute phase response49 and has been shown to prime neutrophils for formation of NETs50. Elevated IL-6 levels, often observed in the hyperglycemic microenvironment of diabetes51, can activate the JAK/STAT3 signaling pathway within neutrophils52. This activation is a critical step in the transcriptional upregulation of peptidylarginine deiminase 4 (PAD4)53, the enzyme responsible for histone citrullination and chromatin decondensation54, hallmarks of NETosis54. Therefore, it is biologically plausible to hypothesize a sequential pathological axis: systemic upregulation of IL-6 promotes excessive formation of NETs, which in turn exacerbates endothelial damage and thrombosis, leading to diverse diabetic complications, including diabetic CAD and renal lesions. Future studies utilizing multivariable MR or mediation analysis are needed to formally quantify the extent to which the deleterious effects of IL-6 are mediated through NETs.

The analytical framework presented here holds significant potential for future applications. Methodologically, this workflow can be extended to Multivariable MR (MVMR) to dissect the independent causal effects of correlated exposures, thereby addressing complex biological pathways more comprehensively. Furthermore, integrating colocalization analyses in future iterations would help verify that the exposure and outcome share the same causal variant, further reducing the risk of linkage disequilibrium bias. Clinically, this approach provides a robust blueprint for drug target validation. By systematically filtering for horizontal pleiotropy and ensuring instrument validity, this framework can prioritize therapeutic targets with a higher probability of clinical success. Ultimately, translating these genetic insights into risk prediction models could facilitate precision medicine strategies, identifying high-risk subgroups that would benefit most from early intervention.

This study utilizes a two-sample MR approach, which leverages genetic variants as IVs to establish causality, thereby providing a robust framework to explore the relationship between NETs-associated inflammatory traits and T2DM and its complications. This approach improves upon observational studies by using MR to mitigate confounding and reverse causation. Unlike standard MR implementations relying solely on IVW estimates, this study integrated robust sensitivity analyses (weighted median, MR-Egger) and pleiotropy checks. This multi-method framework ensures causal estimates are not driven by invalid instruments, offering greater validity than single-method analyses. Furthermore, the reproducibility of this approach is ensured through a standardized, transparent protocol for data harmonization and quality control, addressing the “replication crisis” often seen in complex genetic analyses. In terms of efficiency and applicability, this streamlined pipeline reduces the computational burden associated with manual data curation, allowing for the rapid screening of multiple phenotypes. This scalability makes the framework broadly applicable to analyses but also to high-throughput phenome-wide association scans, offering a more versatile tool for exploring complex biological networks than rigid, single-step observational approaches.

However, the study is not without limitations. The use of genetic instruments relies on the assumption that the selected variants are associated with the exposure of interest and the outcome through a single pathway, which may not fully capture the complexity of biological systems. Second, since this study primarily relied on data from European populations for its MR analysis, the generalizability of the findings to other ethnic or geographic populations may be limited. Third, although this study identified potential associations between NETs-associated inflammatory traits and T2DM and its complications, the specific molecular regulatory networks and underlying biological mechanisms require further clarification through subsequent experimental studies. It is important to note that significant causal associations were limited to specific complications, such as diabetic CAD, renal, and peripheral circulatory complications, and were not uniformly observed across all tested diabetic outcomes. Additionally, while MR can reduce confounding, it cannot account for all potential sources of bias, such as unmeasured environmental factors or population stratification.

Furthermore, the generalizability of the findings may be limited by the specific populations from which the genetic data are derived, and replication in diverse populations is necessary to confirm these results. While this study leverages large-scale GWAS summary statistics, the absence of supplementary multi-omics sequencing data from clinical cohorts or experimental models limits the ability to validate the molecular mechanisms and transcriptional changes underlying the observed causal associations. This MR workflow required specific adjustments to address analytical challenges, particularly regarding allele harmonization and pleiotropy. To mitigate strand ambiguity with palindromic SNPs, this study modified the pipeline to strictly exclude variants with intermediate allele frequencies (>0.42). Furthermore, to troubleshoot potential horizontal pleiotropy, this study integrated the MR-PRESSO outlier test into this workflow. This allowed for the iterative removal of invalid instrumental variables, ensuring that the final causal estimates were robust and not driven by pleiotropic outliers.

The validity of the MR findings relies heavily on several critical analytical steps. First, the rigorous selection of IVs is paramount. By adhering to strict genome-wide significance thresholds (P<5×10-6) and clumping for linkage disequilibrium (r2<0.001), this study minimized the risk of weak instrument bias, ensuring that the selected SNPs are strongly associated with the exposure. Second, the robustness of the causal estimates is contingent upon addressing horizontal pleiotropy. This study relied on multiple complementary methods, specifically MR-Egger, weighted median, and MR-PRESSO, which provided a necessary safeguard; the consistency across these methods, alongside non-significant MR-Egger intercepts, supports the assumption that the IVs influence the outcome solely through the exposure. Although genetic variations can serve as proxies for NET-associated inflammatory traits, they may not fully reflect the dynamic, real-time fluctuations of NETosis in vivo, as the regulatory pathways from the germline genotype to transcriptional activation and protein release involve complex biological processes that are not strictly linear. Regarding reproducibility, the use of large-scale, publicly available GWAS summary statistics ensures that this analysis can be independently verified. The transparency of the analytical pipeline, including the specific parameters for harmonization and outlier removal, allows for the exact replication of these results. These steps collectively strengthen the internal validity of this study and provide a reproducible framework for future investigations.

In conclusion, this study employs a two-sample MR approach to explore the association between NET-associated inflammatory traits and T2DM and its complications. The findings suggest a significant association between IL-6 levels and diabetic CAD, highlighting a potential role for IL-6 in the cardiovascular complications of diabetes. Additionally, this study reveals associations between NETs and various diabetic complications, including renal and peripheral circulatory complications, underscoring the importance of NET-associated inflammatory traits in the pathogenesis of diabetes-related conditions. These insights contribute to understanding the complex interplay among inflammation, glucose metabolism, and vascular health in T2DM. While the study provides valuable evidence for the role of NETs in diabetic complications, it also acknowledges the complexity of the biological mechanisms involved and the need for further research to elucidate the precise pathways and potential therapeutic targets.

Disclosures

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

The authors declare that they have no competing interests.

Acknowledgements

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

The study was supported by the Social Science and Technology Research Major Project of Zhongshan (Grant number: 2021B3013) and the Social Science and Technology Research Project of Zhongshan (Grant number: 2022B1080). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Materials

List of materials used in this article
NameCompanyCatalog NumberComments
CCN1GWAS CatalogGCST90277805http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90277001-GCST90278000/GCST90277805/harmonised/GCST90277805.h.tsv.gz
Diabetic CADGWAS CatalogGCST006405http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST006001-GCST007000/GCST006405/Fall_gwas_results_any_chd_tab.txt.gz
IL-6GWAS CatalogGCST90010146http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90010001-GCST90011000/GCST90010146/harmonised/33303764-GCST90010146-EFO_0007937.h.tsv.gz
MPOGWAS CatalogGCST90087967http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90087001-GCST90088000/GCST90087967/harmonised/35078996-GCST90087967-EFO_0011039.h.tsv.gz
MPO-DNAGWAS CatalogGCST90013658http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90013001-GCST90014000/GCST90013658/harmonised/33717105-GCST90013658-EFO_0005243.h.tsv.gz
MR-PRESSO PackageGitHub (rondolab)Version 1.0R package used to detect and pleiotropy-corrected causal estimates.
NEGWAS CatalogGCST90162366http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90162001-GCST90163000/GCST90162366/harmonised/GCST90162366.h.tsv.gz
NETsGWAS CatalogGCST90137414http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90137001-GCST90138000/GCST90137414/harmonised/GCST90137414.h.tsv.gz
NGALGWAS CatalogGCST90161504http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90161001-GCST90162000/GCST90161504/harmonised/GCST90161504.h.tsv.gz
R SoftwareR Foundation for Statistical ComputingVersion 4.0.5Main statistical computing environment used for all analyses.
T2DFINN R9T2Dhttps://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_T2D.gz
T2D with ketoacidosisFINN R9E4_DM2KETOhttps://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_E4_DM2KETO.gz
T2D with neurological complicationsFINN R9E4_DM2NEUhttps://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_E4_DM2NEU.gz
T2D with ophthalmic complicationFINN R9E4_DM2OPTHhttps://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_E4_DM2OPTH.gz
T2D with peripheral circulatoryFINN R9E4_DM2PERIPHhttps://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_E4_DM2PERIPH.gz
T2D with renal complicationsFINN R9E4_DM2RENhttps://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_E4_DM2REN.gz
TNF-aGWAS CatalogGCST004426http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST004001-GCST005000/GCST004426/harmonised/27989323-GCST004426-EFO_0004684.h.tsv.gz
TwoSampleMR PackageGitHub (MRCIEU)Version 0.6.22R package used for performing the Mendelian Randomization analysis.

References

$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,
  1. Tanase DM, et al. The intricate relationship between type 2 diabetes mellitus (T2DM), insulin resistance (IR), and nonalcoholic fatty liver disease (NAFLD). J Diabetes Res. 2020;2020:3920196.
  2. Yan Y, et al. Prevalence, awareness and control of type 2 diabetes mellitus and risk factors in Chinese elderly population. BMC Public Health. 2022;22(1):1382.
  3. Dugani SB, Mielke MM, Vella A. Burden and management of type 2 diabetes in rural United States. Diabetes Metab Res Rev. 2021;37(5):e3410.
  4. Rodriquez IM, O'Sullivan KL. Youth-onset type 2 diabetes: Burden of complications and socioeconomic cost. Curr Diab Rep. 2023;23(5):59-67.
  5. Zhang H, et al. Neutrophil, neutrophil extracellular traps and endothelial cell dysfunction in sepsis. Clin Transl Med. 2023;13(1):e1170.
  6. Hidalgo A, et al. Neutrophil extracellular traps: From physiology to pathology. Cardiovasc Res. 2022;118(13):2737-53.
  7. Goggs R, Jeffery U, Levine DN, Li RHL. Neutrophil-extracellular traps, cell-free DNA, and immunothrombosis in companion animals: A review. Vet Pathol. 2020;57(1):6-23.
  8. Tilley DO, et al. Histone H3 clipping is a novel signature of human neutrophil extracellular traps. Elife. 2022;11.
  9. Shafqat A, et al. Emerging role of neutrophil extracellular traps in the complications of diabetes mellitus. Front Med (Lausanne). 2022;9:995993.
  10. Njeim R, et al. NETosis contributes to the pathogenesis of diabetes and its complications. J Mol Endocrinol. 2020;65(4):R65-76.
  11. Helseth R, et al. Glucose associated NETosis in patients with ST-elevation myocardial infarction: An observational study. BMC Cardiovasc Disord. 2019;19(1):221.
  12. Magana-Guerrero FS, et al. Spontaneous neutrophil extracellular traps release are inflammatory markers associated with hyperglycemia and renal failure on diabetic retinopathy. Biomedicines. 2023;11(7).
  13. Ibrahim I, et al. Neutrophil extracellular traps (NETs) are associated with type 2 diabetes and diabetic foot ulcer related amputation: A prospective cohort study. Diabetes Ther. 2024;15(6):1333-48.
  14. Joshi MB, et al. High glucose modulates IL-6 mediated immune homeostasis through impeding neutrophil extracellular trap formation. FEBS Lett. 2013;587(14):2241-6.
  15. Bryk AH, et al. Predictors of neutrophil extracellular traps markers in type 2 diabetes mellitus: Associations with a prothrombotic state and hypofibrinolysis. Cardiovasc Diabetol. 2019;18(1):49.
  16. Park JH, et al. Evaluation of circulating markers of neutrophil extracellular trap (NET) formation as risk factors for diabetic retinopathy in a case-control association study. Exp Clin Endocrinol Diabetes. 2016;124(9):557-61.
  17. Winslow S, et al. Multi-omics links IL-6 trans-signalling with neutrophil extracellular trap formation and Haemophilus infection in COPD. Eur Respir J. 2021;58(4).
  18. Bowden J, Holmes MV. Meta-analysis and Mendelian randomization: A review. Res Synth Methods. 2019;10(4):486-96.
  19. Birney E. Mendelian randomization. Cold Spring Harb Perspect Med. 2022;12(4).
  20. Sanderson E, et al. Mendelian randomization. Nat Rev Methods Primers. 2022;2.
  21. Skrivankova VW, et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: The STROBE-MR statement. JAMA. 2021;326(16):1614-21.
  22. Liang YC, et al. Association of circulating inflammatory proteins with type 2 diabetes mellitus and its complications: A bidirectional Mendelian randomization study. Front Endocrinol (Lausanne). 2024;15:1358311.
  23. Papayannopoulos V, Metzler KD, Hakkim A, Zychlinsky A. Neutrophil elastase and myeloperoxidase regulate the formation of neutrophil extracellular traps. J Cell Biol. 2010;191(3):677-91.
  24. Kindberg KM, et al. IL-6R signaling is associated with PAD4 and neutrophil extracellular trap formation in patients with STEMI. Int J Mol Sci. 2025;26(11).
  25. Mao CL, et al. Neutrophil extracellular traps in rheumatoid arthritis: Pathogenic mechanisms and therapeutic potential. Front Immunol. 2025;16:1717671.
  26. Li H, et al. Hepatocytes and neutrophils cooperatively suppress bacterial infection by differentially regulating lipocalin-2 and neutrophil extracellular traps. Hepatology. 2018;68(4):1604-20.
  27. Li T, et al. Cellular communication network factor 1 promotes retinal leakage in diabetic retinopathy via inducing neutrophil stasis and neutrophil extracellular traps extrusion. Cell Commun Signal. 2024;22(1):275.
  28. Hao L, et al. Transcriptome sequencing and Mendelian randomization analysis identified biomarkers related to neutrophil extracellular traps in diabetic retinopathy. Front Immunol. 2024;15:1408974.
  29. Wang S, et al. Association between periodontitis and temporomandibular joint disorders. Arthritis Res Ther. 2023;25(1):143.
  30. Yin KJ, et al. No genetic causal association between periodontitis and arthritis: A bidirectional two-sample Mendelian randomization analysis. Front Immunol. 2022;13:808832.
  31. Genomes Project C, et al. A map of human genome variation from population-scale sequencing. Nature. 2010;467(7319):1061-73.
  32. Burgess S, Thompson SG, Collaboration CCG. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol. 2011;40(3):755-64.
  33. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658-65.
  34. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32(5):377-89.
  35. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40(4):304-14.
  36. Greco MF, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in Mendelian randomisation studies with summary data and a continuous outcome. Stat Med. 2015;34(21):2926-40.
  37. Verbanck M, Chen CY, Neale B, Do R. Publisher correction: Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(8):1196.
  38. Cheng H, Garrick DJ, Fernando RL. Efficient strategies for leave-one-out cross validation for genomic best linear unbiased prediction. J Anim Sci Biotechnol. 2017;8:38.
  39. Cupido AJ, et al. Dissecting the IL-6 pathway in cardiometabolic disease: A Mendelian randomization study on both IL6 and IL6R. Br J Clin Pharmacol. 2022;88(6):2875-84.
  40. Indumathi B, Oruganti SS, Sreenu B, Kutala VK. Association of promoter methylation and expression of inflammatory genes IL-6 and TNF-alpha with the risk of coronary artery disease in diabetic and obese subjects among Asian Indians. Indian J Clin Biochem. 2022;37(1):29-39.
  41. Shrivastav D, et al. Comparative analysis of Nε-carboxymethyl-lysine and inflammatory markers in diabetic and non-diabetic coronary artery disease patients. World J Diabetes. 2023;14(12):1754-65.
  42. Chittaragi GH, Ambali AP, Honnutagi R, Ganesh V. Interleukin-6 and uric acid among type 2 diabetes mellitus patients with coronary artery disease. Bioinformation. 2023;19(12):1134-8.
  43. Liu L, et al. Adipokines, adiposity, and atherosclerosis. Cell Mol Life Sci. 2022;79(5):272.
  44. Hua MJ, Butera G, Akinyemi O, Porterfield D. Biases and limitations in observational studies of long COVID prevalence and risk factors: A rapid systematic umbrella review. PLoS One. 2024;19(5):e0302408.
  45. Savitz DA, Wellenius GA. Can cross-sectional studies contribute to causal inference? It depends. Am J Epidemiol. 2023;192(4):514-6.
  46. Hirano T. IL-6 in inflammation, autoimmunity and cancer. Int Immunol. 2021;33(3):127-48.
  47. Liu G, Ren X, Li Y, Li H. Midkine promotes kidney injury in diabetic kidney disease by increasing neutrophil extracellular traps formation. Ann Transl Med. 2022;10(12):693.
  48. Zheng F, et al. Neutrophil extracellular traps induce glomerular endothelial cell dysfunction and pyroptosis in diabetic kidney disease. Diabetes. 2022;71(12):2739-50.
  49. Schmidt-Arras D, Rose-John S. IL-6 pathway in the liver: From physiopathology to therapy. J Hepatol. 2016;64(6):1403-15.
  50. Xiao Y, et al. Cathepsin C promotes breast cancer lung metastasis by modulating neutrophil infiltration and neutrophil extracellular trap formation. Cancer Cell. 2021;39(3):423-37.e427.
  51. Sater MS, et al. Plasma IL-6, TREM1, uPAR, and IL6/IL8 biomarkers increment further witnessing the chronic inflammation in type 2 diabetes. Horm Mol Biol Clin Investig. 2023;44(3):259-69.
  52. Martino N, et al. Endothelial SOCS3 maintains homeostasis and promotes survival in endotoxemic mice. JCI Insight. 2021;6(14).
  53. Akboua H, Eghbalzadeh K, Keser U, Wahlers T, et al. Impaired non-canonical transforming growth factor-β signalling prevents profibrotic phenotypes in cultured peptidylarginine deiminase 4-deficient murine cardiac fibroblasts. J Cell Mol Med. 2021;25(20):9674-84.
  54. Thiam HR, et al. NETosis proceeds by cytoskeleton and endomembrane disassembly and PAD4-mediated chromatin decondensation and nuclear envelope rupture. Proc Natl Acad Sci U S A. 2020;117(13):7326-37.

Reprints and Permissions

Request permission to reuse the text or figures of this JoVE article

Request Permission

Tags

Neutrophil Extracellular TrapsInterleukin 6Diabetic ComplicationsCoronary Artery DiseaseRenal ComplicationsPeripheral Circulatory ComplicationsGenome Wide Association

Related Articles