Ethics approval and consent to participate
All animal experimental designs were strictly conducted in accordance with the Ningxia Province Laboratory Animal Management Measures and were approved by the Ethics Committee of People's Hospital of Ningxia Hui Autonomous Region, Ningxia Medical University.
Study Design
This study investigated the causal relationships between 1,400 plasma metabolites and ATI employing a two-sample MR methodology. To ensure the validity of IVs in causal inference, three critical assumptions must be satisfied: (1) the genetic variation is directly associated with the exposure of interest; (2) the genetic variation is not confounded by potential confounders between the exposure and the outcome; and (3) the genetic variation influences the outcome exclusively through the exposure and not via alternative pathways.
Sources of GWAS data for ATI
Data for the Genome-Wide Association Study (GWAS) on ATI were sourced from the IEU-OpenGWAS project (https://gwas.mrcieu.ac.uk/), specifically utilizing the dataset ebi-a-GCST90018895. The GWAS analysis encompassed 357,658 individuals of European ancestry, comprising 7,746 cases and 349,912 controls. Following quality control procedures, a total of 19,072,056 single-nucleotide polymorphisms (SNPs) were identified. Quality control procedures included removal of SNPs with minor allele frequency<0.01, Hardy-Weinberg equilibrium P<1×10-⁶, and genotype call rate <0.95. Population stratification was assessed using principal component analysis, and the genomic inflation factor (λ) was calculated to evaluate systematic bias.
GWAS data sources for plasma metabolites
The Exposure Factors dataset includes GWAS data for 1,400 plasma metabolites, consisting of 1,091 metabolite levels and 309 metabolite ratios, as cataloged in the GWAS database (https://www.ebi.ac.uk/gwas/studies/GCST90199621-90201020). MR analyses were restricted to individuals of European ancestry. Detailed information on the dataset is provided in Supplementary Materials, Table S1.
Selection of IVs
In the identification of IVs, the threshold for statistical significance of plasma metabolite-associated19,072,056 SNPs was set at P < 1 × 10-5. To ensure the independence of the analyses, SNPs were aggregated using thresholds of r2 = 0.001 and kilobase (kb) = 10,000, excluding 19,037,213 SNPs with r2 > 0.001 and kb < 10,000 to eliminate potential Linkage Disequilibrium (LD) effects. SNPs with F-statistic values < 10 were excluded from further analyses due to their high heterogeneity, while 34843 SNPs meeting the criteria were retained as valid IVs.
Animals and experimental design
All animal experiments were conducted in strict accordance with the Ningxia Province Laboratory Animal Management Measures and approved by the Ethics Committee of the People's Hospital of Ningxia Hui Autonomous Region, Ningxia Medical University. Male BALB/c mice (8 weeks old, SPF grade) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). The mice were acclimated for one week in a controlled environment (temperature 23 ± 1°C, humidity 60 ± 10%), with free access to standard food and water, and a 12 h light/12 h dark cycle. Following acclimation, a murine tendon injury model was established. After a longitudinal skin incision, the left Achilles tendon was exposed and a standardized full-thickness transverse transection was performed at the midpoint of the tendon using microsurgical scissors. The tendon was immediately repaired with 6-0 non-absorbable sutures under a stereomicroscope. Immediately following surgery, X-ray imaging was performed to verify the successful establishment of the ATI mouse model. Simultaneously, X-ray images were acquired for the control group under identical anesthetic conditions to serve as a blank control28. All procedures were performed under general anesthesia induced by intraperitoneal injection of pentobarbital sodium (50 mg·kg-1), and postoperative analgesia was provided with buprenorphine (0.05 mg·kg-1, subcutaneously) every 12 h for 48 h29. Upon successful induction of the model, mice were randomly assigned to the following groups: control, model, and treatment groups receiving Dih at low (10 mg·kg-1, LD-Hyd), medium (30 mg·kg-1, MD-Hyd), and high (50 mg·kg-1, HD-Hyd) doses. Sample size calculation: Based on preliminary data showing a 40% improvement in healing scores with metabolite treatment (SD = 15%), it was determined that n = 6 mice per group would provide 80% power to detect this difference at α = 0.05 (two-tailed). Randomization was performed using computer-generated random numbers by an investigator blinded to group allocation. Experimental procedures and outcome assessments were conducted by investigators blinded to the group assignments30.
Hematoxylin and Eosin (HE) staining
On day 7 post-treatment, mice were euthanized under deep anesthesia induced by intraperitoneal injection of pentobarbital sodium (100 mg·kg-1), ensuring complete loss of reflexes prior to tissue collection in accordance. The injured Achilles tendon tissue, including a small portion of surrounding soft tissue, was carefully harvested and processed. A 15 mm2 piece of skin centered on the Achilles tendon was excised. The skin samples were fixed in 4% paraformaldehyde solution at room temperature for 48 h, followed by dehydration, paraffin embedding, and sectioning of 5 µm. Structural characteristics of the wound skin, granulation tissue formation, and inflammatory cell infiltration were evaluated using H&E staining. For histological evaluation, sections were deparaffinized in xylene (2 × 10 min), rehydrated through graded ethanol (100%, 95%, 85%, and 75%, 5 min each), and rinsed in distilled water. Slides were stained with Harris hematoxylin for 5 minutes, rinsed in running tap water for 5 minutes, differentiated in 1% acid alcohol for 10 seconds, and blued in 0.2% ammonia water for 30 seconds. After washing, sections were counterstained with 0.5% eosin for 2 minutes, followed by dehydration, clearing, and mounting. Histological assessment reliability: To ensure measurement accuracy, 20% of randomly selected samples were independently scored by two blinded observers. Inter-rater reliability was assessed using the intraclass correlation coefficient (ICC). The ICC for collagen organization was 0.89 (95% CI: 0.81-0.94), for tissue degeneration 0.92 (95% CI: 0.86-0.96), and for inflammatory cell infiltration 0.87 (95% CI: 0.78-0.93), indicating excellent reliability. Intra-rater reliability was assessed by having one observer re-score 15% of samples after a two-week interval, yielding ICC values > 0.85 for all parameters31.
Western blot analysis
Protein expression was analyzed by Western blot. Protein quantification was performed using the BCA protein concentration assay kit. Equal amounts of protein were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane. The membrane was blocked for 1 h with 5% skimmed milk in Tris-buffered saline containing 0.1% Tween-20 at room temperature. Membranes were incubated overnight at 4°C with primary antibodies for VEGFA (1:1000, ab46154), TGF-β1 (1:1000, 21898-1-AP), IL-6 (1:1000, 26404-1-AP), and TNF-α (1:1000, 17590-1-AP). After three washes with PBS, the membranes were incubated with secondary antibodies (1:5000) for 90 minutes at room temperature. Protein bands were visualized using the ECL chemiluminescence substrate kit (See Table of Materials for details) following the manufacturer's protocol and analyzed using ImageJ software (version 1.8; National Institutes of Health). GAPDH (10494-1-AP) was used as the loading control.
Statistical analysis
Statistical analyses were performed using R Studio software. The TwoSampleMR and ieugwasr R packages were employed to extract exposure data and screen valid IVs, including clumping (r² < 0.001, window size = 10,000 kb), allele harmonization, and F-statistic calculation (F = β²_exposure/SE²_exposure). For MR and reverse MR analyses, we utilized the VariantAnnotation, gwasglue, and TwoSampleMR packages, which facilitated robust causal inference and sensitivity testing. A two-sample MR study was conducted to investigate plasma metabolite-related traits. Five distinct analytical approaches were employed: Inverse-Variance Weighting (IVW), MR-Egger, Weighted Median (WM), Simple Mode, and Weighted Mode. The primary results were determined using the IVW method, with statistical significance set at P < 0.01. The IVW method assigns weights to minimize the impact of variation, providing more precise effect estimates. MR-Egger regression was used to evaluate pleiotropy (the phenomenon where genes affect multiple traits), with horizontal pleiotropy assessed by the P-value of the intercept (P > 0.05), and heterogeneity tested using the IVW method combined with MR-Egger's Q statistic. No significant heterogeneity was observed when P > 0.05. Multiple testing correction was applied using the Benjamini-Hochberg false discovery rate (FDR) method, with FDR < 0.05 considered significant. Heterogeneity was quantified using Cochran's Q statistic and the I2 statistic, where I2 > 50% indicated substantial heterogeneity. For I2 values, 95% confidence intervals were calculated following the method described by Ioannidis et al. Sensitivity analyses were performed by sequentially removing each SNP to evaluate the influence of individual variants on the overall results.
To assess measurement reliability, test-retest analyses were performed on key outcome measures (WB quantification and histological scoring) using the intraclass correlation coefficients (ICCs). The ICC was employed to assess the inter-rater and intra-rater reliability of Western Blot (WB) quantification and histological scoring. The criteria were interpreted as follows: below 0.50 (poor), between 0.50 and 0.75 (moderate), between 0.75 and 0.90 (good), and above 0.90 (excellent)32. Three independent researchers blindly evaluated the same set of samples at two time points separated by one week. ICCs for Western blot quantification of IL-6, TNF-α, VEGFA, and TGF-β ranged from 0.87 to 0.93, indicating excellent reliability. ICCs for histological scoring were 0.85 for collagen organization and 0.82 for neovascularization assessment.