ABSTRACT
Background
Inflammation is increasingly recognized in the pathogenesis of diabetic neuropathy (DN). Circulating miR‐27a‐3p levels are closely related to inflammation and increased in patients with diabetic nephropathy and retinopathy. miR‐27a‐3p has a predicted binding site in the 3 UTR of Netrin‐1 mRNA, an anti‐inflammatory nerve growth factor whose levels correlate with small nerve fiber loss in patients with DN.
Methods
Eighty‐two participants with (DN+) and without (DN‐) small nerve fiber damage and 45 healthy controls underwent measurement of plasma miR‐27a‐3p, PBMC levels of NF‐kB and NLRP3, serum Netrin‐1, TNFα, IL‐1β, and IL‐10 levels.
Results
Corneal nerve fiber density (CNFD), branch density (CNBD), and fiber length (CNFL) were significantly lower in the DN‐ and DN+ groups compared to controls (p < 0.001). Plasma miR‐27a‐3p and PBMC NF‐kB and NLRP3 expression were significantly higher in the DN+ group, and miR‐27a‐3p identified DN+ with an area under the curve (AUC) of 89%. Serum Netrin‐1 and IL‐10 levels were lower, while TNFα and IL‐1β levels were higher in the DN+ group. miR‐27a‐3p levels correlated negatively with CNFL and Netrin‐1 and positively with NF‐kB and NLRP3.
Conclusion
miR‐27a‐3p levels are elevated in subjects with DN and correlate with markers of inflammation, Netrin‐1, and corneal nerve loss. miR‐27a‐3p could be a biomarker for DN, and miR‐27a‐3p/Netrin‐1 crosstalk may be a promising therapeutic target for DN.
Keywords: biomarker, corneal confocal microscopy, diabetic neuropathy, miR‐27a‐3p, netrin‐1, neuroinflammation
1. Introduction
Diabetic Neuropathy (DN) is associated with pain, numbness, foot ulceration, and amputation [1, 2]. Based on an assessment of symptoms and signs, many patients remain undiagnosed [3, 4, 5, 6]. However, corneal confocal microscopy (CCM) allows quantification of corneal nerves [7] that enables the identification of sub‐clinical DN and predicts incident DN [8, 9] as well as progression or improvement of DN [10].
There are currently no FDA‐approved disease‐modifying treatments for DN [11], with many failed clinical trials of putative therapies like aldose reductase inhibitors and PKC inhibitors [12]. Whilst the pathogenesis of DN is multifactorial, the emerging role of inflammation is increasingly recognized [13]. Netrin‐1, a laminin‐related protein, was initially discovered as the main cue for axonal guidance [14], but recent studies have shown that it plays an important role in promoting Schwann cell proliferation and migration, supports nerve regeneration [15], and protects from neuroinflammation [16]. Netrin‐1 levels are reduced in subjects with type 2 diabetes [17], and our recent study showed that circulating Netrin‐1 levels correlate with corneal nerve fiber loss in subjects with DN [18].
MicroRNAs (miRNAs) are small endogenous non‐coding RNAs that regulate gene expression by binding to the 3′ untranslated region (3′ UTR) of the target gene mRNA, repressing translation [19]. It has been clarified that miR‐27a might play a pivotal part in the insulin signaling pathways related to glucose metabolism and IR [20]. miR‐27a plays a crucial role in the regulation of glucose metabolism by targeting the GLUT4 signaling pathway in L6 cells [21]. A plausible association of miR‐27a with metabolic syndrome and type 2 diabetes patients has also been found [22]. miR‐27a‐3p has been shown to be upregulated in patients with T2D [23], gestational diabetes mellitus (GDM) [24], and is linked with pathways involved in both T2D and cancer [25]. Moreover, high urinary miR‐27a‐3p has been shown to be a biomarker for diabetic kidney disease [26] and diabetic retinopathy [27]. miR‐27a, together with miR‐27b, has been shown to play an important role in an in vitro model of neurovegetative disease [28]. However, its role in diabetes‐related small fiber damage has not been studied yet.
Bioinformatics analysis showed that miR‐27a‐3p predicted target region lies in the 3’UTR of the netrin‐1 gene (targetscan.org). Hence, we explored the possible association between miR‐27a‐3p and netrin‐1 in subjects with early neuropathy determined by CCM and the involvement and interaction of miR‐27a‐3p, Netrin‐1, and inflammation in subjects with DN.
2. Materials and Methods
2.1. Participants
The study was approved by the Institutional Ethics Committee of IPGME&R and SSKM Hospital (Kolkata, India) (Memo No: IPGME&R/IEC/2022/69). Participants were recruited from the diabetic outpatient clinic, Institute of Post Graduate Medical Education & Research (IPGME&R), Kolkata, India, and compared with 45 age‐ and sex‐matched healthy controls after written informed consent. Subjects with type 1 diabetes, pregnancy, thyroid disease, gastrointestinal disease, chronic kidney disease (CKD, eGFR < 60 mL/min/1.73 m2 or known clinical diagnosis), significant hepatic disease, or major systemic inflammatory/autoimmune conditions were excluded. Medication histories were collected; however, glucose‐lowering regimens were heterogeneous and not uniformly stratified, which is acknowledged as a limitation of the present study, as well as cardiovascular disease, progressive malignant tumors, history of neurological disorders due to non‐diabetic causes, corneal surgery, or dystrophy were excluded from this study.
2.2. Demographic and Clinical Assessment
All study participants underwent measurement of height, weight, BMI, body fat percentage (Bioelectrical impedance method), and blood pressure. HbA1c was analyzed with the HPLC method; fasting plasma glucose (FPG) levels were measured with the GOD‐POD method, and the lipid profile and liver function tests were measured using a semi‐autoanalyzer. Patients underwent assessment with the Douleur Neuropathique 4 (DN4) questionnaire, with a DN4 score > 4 indicating painful diabetic neuropathy. Warm and cold perception thresholds (WPT and CPT) and vibration perception threshold (VPT) were assessed on each foot.
2.3. In Vivo Corneal Confocal Microscopy
All study participants underwent corneal nerve imaging with a Corneal Confocal Microscope (CCM) [Heidelberg Retina Tomograph 3 (HRT3) Rostock Cornea Module (RCM)]. The subject's eyes were anesthetized with proparacaine hydrochloride (0.5% w/v) ophthalmic solution, followed by the application of a drop of hydroxypropyl methylcellulose (Hypromellose) 0.3% ophthalmic gel between the lens and the Tomocap. The lens was moved towards the cornea until it touched, and after focal plane ring adjustment, images from various depths were captured. Three non‐overlapping good‐quality images from the corneal sub‐basal nerve plexus of each eye were analyzed using fully automated image analysis software (ACCMetrics, V.2.0, University of Manchester, UK). An annotated CCM image panel indicating main nerve trunks and branches is shown in Supplemental Figure S1. Corneal nerve fiber density (CNFD; the number of main nerve fibers/mm2), branch density (CNBD; the total number of main nerve branches/mm2), and fiber length (CNFL; corneal nerve fiber length/mm2) were analyzed. Participants were divided into those without (DN‐) and with (DN+) based on small nerve fiber damage defined by a CNFL of 2 Standard Deviations (SD) below the mean of the control group.
2.4. Serum Netrin‐1 and Cytokines
Serum Netrin‐1 levels were determined using an enzyme‐linked immunosorbent assay (ELISA) method with a commercial kit from Cusabio (Wuhan Huamei Biotech Co Ltd., China) following the manufacturer's protocol. Serum levels of pro‐inflammatory cytokines i.e., TNFα and IL‐1β, and the anti‐inflammatory cytokine IL‐10 were determined by sandwich ELISA with a commercial kit from Bioassay Technology Laboratory (Korain Biotech, China). Absorbances were read using a 450 nm filter on a microplate reader by Erba Lisa Scan EM (Transasia Bio‐medical Limited, India).
2.5. RNA Extraction and Quantitative Real‐Time Polymerase Chain Reaction (qRT‐PCR)
Total RNAs from plasma and Peripheral Blood Mononuclear Cells (PBMCs) were extracted using the TRIzol Method (Invitrogen). Prior studies have demonstrated upregulation of key inflammatory mediators such as NF‐κB and NLRP3 in PBMCs in individuals with diabetic nephropathy and retinopathy, justifying the study of PBMCs as a marker of inflammation in DN. Complementary DNA (cDNA) for microRNA was reverse transcribed by reverse transcription reagents (Thermo Scientific) using Universal Stem Loop Primer (USLP). mRNA‐specific cDNAs were synthesized from PBMC total RNAs using the RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific). qRT‐PCR was performed with the DyNAmo ColorFlash SYBR Green qPCR Kit (Thermo Scientific) in ABI 750, QuantStudio 5 (Applied Biosystems). The 2−ΔΔCt method was used to calculate relative expression levels. U6 snRNA and 18 s were used as internal controls for miRNA and mRNA, respectively. The primer sequences are as follows: USLP: 5′‐GAAAGAAGGCGAGGAGCAGATCGAGGAAGAAGACGGAAGAATGTGCGTCTCGCCTTC‐3′, miR‐27a‐3p Forward: 5′‐TCGTGTTCACAGTGGCTAAG‐3′, Universal Reverse: 5′‐CGAGGAAGAAGACGGAAGAAT‐3′, U6 Forward: 5′‐GCTTCGGCAGCACATATACTAAAAT‐3′, U6 Reverse: 5′‐CGCTTCACGAATTTGCGTGTCAT‐3′, NF‐kB Forward: 5′‐AGATGGCCCATACCTTCAA‐3′, NF‐kB Reverse: 5′‐GGCATGCAGGTGGGATATTTT‐3′, NLRP3 Forward: 5′‐GGACTGAAGCACCTGTTGTGCA‐3′, NLRP3 Reverse: 5′‐TCCTGAGTCTCCCAAGGCATTC‐3′, 18 s Forward: 5′‐GTAACCCGTTGAACCCCATT‐3′, 18 s Reverse: 5′‐CCATCCAATCGGTAGTAGCG‐3′.
2.6. Statistical Analysis
The ‘Kolmogorov–Smirnov test’ was performed to determine the assumption of normality for the data, obtained from the three groups. Data are presented as mean ± SD (standard deviation). The One‐way Analysis of Variance (ANOVA) followed by Tukey's posthoc test (parametric) or Kruskal Wallis (non‐parametric) test followed by Dunn's multiple comparisons test (non‐parametric) was used to test for significant differences between groups. The Pearson's or Spearman's correlation test was used to determine the correlation between variables. Receiver operating characteristic (ROC) curve and area under the curve (AUC) analysis was performed to assess the diagnostic utility of miR‐27a‐3p for DN. Yates' chi‐square test was used to obtain the negative and positive predictive values. Statistical analysis was performed using GraphPad Prism software (V.9.5.0, San Diego, CA, USA). A value of p < 0.05 was considered to be statistically significant.
3. Results
3.1. Demographic and Clinical Findings
The demographic and clinical parameters in the study population are shown in Table 1. A total of 82 subjects with type 2 diabetes, without [DN‐, n = 42] and with [DN+, n = 40] corneal nerve loss, were compared to 45 age‐ and sex‐matched individuals without diabetes. There was no significant difference in age, BMI, body fat%, blood pressure, lipid profiles, and liver function tests between the study groups. There was no significant difference in the duration of diabetes between the DN‐ and DN+ groups. FPG and HbA1c levels were higher in the DN‐ and DN+ groups compared to controls (p < 0.0001), but were comparable between DN‐ and DN+.
TABLE 1.
Demographic and clinical parameters among the study groups.
| Control N = 45 | DN‐ N = 42 | DN+ N = 40 | p | |
|---|---|---|---|---|
| Age (years) | 47.35 ± 8.52 | 49.37 ± 10.56 | 52.79 ± 7.84 | ns |
| BMI | 24.45 ± 3.28 | 24.70 ± 2.35 | 24.36 ± 2.23 | ns |
| Body fat % | 29.74 ± 9.81 | 28.99 ± 8.56 | 29.88 ± 8.87 | ns |
| SBP (mmHg) | 122.70 ± 11.94 | 131.90 ± 17.00 | 126.90 ± 18.21 | ns |
| DBP (mmHg) | 80.00 ± 7.75 | 84.61 ± 9.96 | 79.32 ± 9.88 | ns |
| FPG (mg/dL) | 79.09 ± 10.07 | 131.20 ± 48.87 | 151.60 ± 67.81 | < 0.0001 |
| HbA1c % | 5.19 ± 0.27 | 7.49 ± 1.17 | 7.94 ± 1.79 | < 0.0001 |
| HbA1c (mmol/mol) | 33.2 ± 3.0 | 58.4 ± 12.8 | 63.3 ± 19.6 | < 0.0001 |
| Duration of diabetes (years) | 6.75 ± 5.528 | 9.52 ± 6.88 | ns | |
| Alkaline phosphatase (IU/L) | 76.05 ± 20.77 | 83.98 ± 32.21 | 92.33 ± 39.01 | ns |
| SGPT (U/L) | 25.72 ± 21.85 | 32.95 ± 22.05 | 21.63 ± 11.05 | ns |
| SGOT (U/L) | 28.35 ± 13.31 | 35.90 ± 26.17 | 27.38 ± 9.77 | ns |
| Cholesterol (mg/dL) | 180.20 ± 33.81 | 178.90 ± 40.33 | 168.70 ± 52.00 | ns |
| TG (mg/dL) | 117.50 ± 36.86 | 133.70 ± 49.12 | 133.70 ± 48.63 | ns |
| HDL (mg/dL) | 42.02 ± 10.20 | 42.12 ± 13.96 | 42.04 ± 11.48 | ns |
| LDL (mg/dL) | 122.80 ± 32.59 | 110.30 ± 40.87 | 101.00 ± 37.49 | ns |
| DN4 score | 0.06 ± 0.25 | 0.20 ± 0.47 | 0.30 ± 0.75 | ns |
| Warm perception threshold (°C) | 35.23 ± 0.58 | 37.07 ± 1.86 | 36.93 ± 2.85 | 0.017 |
| Cold perception threshold (°C) | 26.39 ± 1.05 | 24.55 ± 1.82 | 24.52 ± 3.02 | 0.0183 |
| VPT (V) | 7.81 ± 2.68 | 15.16 ± 10.36 | 18.24 ± 9.30 | < 0.0001 |
| CNFD (no/mm2) | 28.83 ± 3.73 | 23.22 ± 4.88 | 16.91 ± 3.35 | < 0.0001 |
| CNBD (no/mm2) | 31.50 ± 10.01 | 23.34 ± 12.07 | 13.37 ± 4.71 | < 0.0001 |
| CNFL (mm/mm2) | 16.64 ± 2.54 | 13.36 ± 1.18 | 10.16 ± 1.10 | < 0.0001 |
| Netrin‐1 (pg/mL) | 1145.00 ± 313.4 | 767.10 ± 300.0 | 529.20 ± 182.70 | < 0.0001 |
| TNFα (ng/L) | 63.07 ± 7.94 | 78.10 ± 15.46 | 93.80 ± 27.63 | < 0.001 |
| IL‐1β (pg/mL) | 831.30 ± 282.8 | 1041.00 ± 277.60 | 1291.00 ± 345.10 | < 0.0001 |
| IL‐10 (pg/mL) | 209.50 ± 125.00 | 204.50 ± 90.23 | 137.80 ± 29.89 | 0.0017 |
Note: Data are presented as the mean ± standard deviation and P value (ns = not significant).
Abbreviations: BMI: body mass index, CNFD: corneal nerve fiber density, CNBD: corneal nerve branch density, CNFL: corneal nerve fiber length, DBP: diastolic blood pressure, FPG: fasting plasma glucose, HDL: high density lipoprotein, LDL: low density lipoprotein, SBP: systolic blood pressure, SGOT: serum glutamic‐oxaloacetic transaminase, SGPT: serum glutamic pyruvic transaminase, TG: triglyceride, VPT: vibration perception threshold.
3.2. Neuropathy Evaluation
The DN4 score did not differ significantly between DN‐, DN+, and controls. Warm perception threshold was significantly higher in DN‐ (35.23 ± 0.58 vs. 37.07 ± 1.86, p < 0.05) and DN+ (35.23 ± 0.58 vs. 36.93 ± 2.85, p < 0.05) compared to controls, but did not differ between DN‐ and DN+ (p = 0.8). Cold perception threshold was significantly lower in DN‐ (26.39 ± 1.05 vs. 24.55 ± 1.82, p < 0.05) and DN+ (26.39 ± 1.05 vs. 24.52 ± 3.02, p < 0.05) groups compared to controls, but did not differ between DN‐ and DN+ (p = 0.9). Vibration perception threshold was significantly higher in DN‐ (15.16 ± 10.36 vs. 7.81 ± 2.68; p < 0.001) and DN+ (18.24 ± 9.30 vs. 7.81 ± 2.68; p < 0.0001) groups compared to controls, but did not differ between DN− and DN+ (p = 0.2). (Figure S2).
3.3. Corneal Confocal Microscopy
CNFD (28.83 ± 3.73 vs. 16.91 ± 3.35, p < 0.0001), CNBD (31.50 ± 10.01 vs. 13.37 ± 4.71, p < 0.0001), and CNFL (16.64 ± 2.54 vs. 10.16 ± 1.10, p < 0.0001) were significantly lower in the DN+ group compared to controls. CNFD (28.83 ± 3.73 vs. 23.22 ± 4.88, p < 0.0001), CNBD (31.50 ± 10.01 vs. 23.34 ± 12.07, p < 0.01), and CNFL (16.64 ± 2.54 vs. 13.36 ± 1.18, p < 0.0001) were significantly lower in the DN‐ group compared to controls. CNFD (23.22 ± 4.88 vs. 16.91 ± 3.35, p < 0.0001), CNBD (23.34 ± 12.07 vs. 13.37 ± 4.71, p < 0.001), and CNFL (13.36 ± 1.18 vs. 10.16 ± 1.10, p < 0.0001) were significantly lower in the DN+ group compared to the DN‐ group (Figure 1).
FIGURE 1.
(A) Corneal confocal image from a control, (B) patient with DN‐ and (C) DN+ and (D) bar charts (mean and standard deviation) showing a progressive significant decrease in corneal nerve fiber density (CNFD), (E) corneal nerve branch density (CNBD) and (F) corneal nerve fiber length (CNFL) in patients with DN‐ and DN+ compared to controls. ****p < 0.0001, ***p < 0.001, **p < 0.01.
3.4. Serum Netrin‐1 and Cytokine Levels
Serum Netrin‐1 levels (pg/mL) were significantly lower in DN‐ (1145.00 ± 313.40 vs. 767.10 ± 300.00, p < 0.0001) and DN+ (1145.00 ± 313.40 vs. 529.20 ± 182.70, p < 0.0001) groups compared to controls and were significantly lower in DN+ compared to the DN‐ group (p < 0.001). TNFα levels (ng/L) were significantly higher in the DN‐ (78.10 ± 15.46 vs. 63.07 ± 7.94, p < 0.05) and DN+ (93.80 ± 27.63 vs. 63.07 ± 7.94, p < 0.001) groups compared to controls and were significantly higher in the DN+ compared to the DN‐ group (p < 0.05). IL‐1β levels (pg/mL) were significantly higher in DN‐ (1041.00 ± 277.60 vs. 831.30 ± 282.80, p < 0.01) and DN+ (1291.00 ± 345.10 vs. 831.30 ± 282.80, p < 0.0001) groups compared to controls and were significantly higher in DN+ compared to DN‐ groups (p < 0.01). Serum IL‐10 levels (pg/mL) were significantly lower in the DN+ group (137.80 ± 29.89 vs. 209.50 ± 125.00, p < 0.01), but not the DN‐ group (204.5 ± 90.23 vs. 209.50 ± 125.00, p > 0.9), compared to controls (Figure 2). Serum Netrin‐1 did not correlate with BMI or body fat% in the overall cohort (Supplemental Table S3).
FIGURE 2.
(A) Bar chart expressed as the mean and standard deviation showing a progressive significant decrease in serum netrin‐1, and (B) IL‐10, and (C) an increase in TNFα and (D) IL‐1β levels in DN+ compared to DN‐ and controls. ****p < 0.0001, ***p < 0.001 **p < 0.01, *p < 0.05.
3.5. Plasma miR‐27a‐3p and PBMC NF‐kB and NLRP3
There was a significant increase in the expression of plasma miR‐27a‐3p in type 2 diabetes, which did not correlate with clinical and demographic variables (Supplemental Table S4). There was ~twofold upregulation in the DN‐ (p < 0.05) group and a ~ fourfold upregulation in the DN+ (p < 0.0001) group compared to controls, with a significant difference between DN+ and DN‐ (p < 0.05). PBMC NF‐kB expression was ~threefold higher in the DN+ group compared to controls (p < 0.01), with no difference between DN‐ and controls or DN‐ and DN+. NLRP3 expression was ~fourfold higher in the DN+ group compared to controls (p < 0.01), with no difference between the DN‐ group and controls or DN‐ and DN+ groups (Figure 3).
FIGURE 3.
(A) Bar chart expressed as the mean and standard deviation showing a progressive significant increase in plasma miR‐27a‐3p in DN‐ and (B) DN+ and an increase in PBMC NF‐kB and (C) NLRP3 in the DN+ group compared to controls. ****p < 0.0001, **p < 0.01, *p < 0.05.
3.6. ROC Analysis and Diagnostic Utility of Plasma miR‐27a‐3p for DN
A > 2.95 fold change in plasma miR‐27a‐3p revealed an area under the curve (AUC) of 0.89 (p < 0.0001; 95% Confidence interval 0.72–0.96) with a sensitivity of 78%, specificity of 89%, and likelihood ratio of 7.04 for DN+. Furthermore, Yates' chi‐square correction test demonstrated a positive predictive value of 86% and a negative predictive value of 83% (p < 0.0001) (Figure 4).
FIGURE 4.
Receiver operating characteristic curve analysis of plasma miR‐27a‐3p for DN+. miR‐27a‐3p with an area of 89% (sensitivity 78% and specificity 89%) significantly distinguished DN+ (p < 0.0001).
3.7. Association of Plasma miR‐27a‐3p With CCM Parameters and Netrin‐1
Plasma miR‐27a‐3p expression levels correlated with CNFD (r = −0.40, p < 0.05), CNFL (r = −0.49, p < 0.01), and serum Netrin‐1 levels (r = −0.50, p < 0.01) in the DN+ group (Figure 5). To further address severity, we have added an ordinal severity analysis by CNFL tertiles across all participants, demonstrating a stepwise increase in miR‐27a‐3p (p = 0.04) and a decrease in Netrin‐1 (p = 0.01) with worsening CNFL. These graphs are depicted in the Figure S5.
FIGURE 5.
(A–C) Correlation analysis between miR‐27a‐3p with CCM parameters and (D) Netrin‐1.
3.8. Association of Plasma miR‐27a‐3p and Netrin‐1 With NF‐kB and NLRP3 Expression
Plasma miR‐27a‐3p expression correlated with PBMC expression of NF‐kB (r = 0.44, p < 0.01) and NLRP3 (r = 0.44, p < 0.01), whereas serum Netrin‐1 levels showed a negative correlation with NF‐kB (r = −0.42, p < 0.05) and NLRP3 (r = −0.48, p < 0.01) in the DN+ group (Figure 6).
FIGURE 6.
(A, B) Correlation analysis of plasma miR‐27a‐3p and (C) serum Netrin‐1 with PBMC expression of NF‐kB and (D) NLRP3.
4. Discussion
In this study, we show that circulating miR‐27a‐3p levels were significantly higher in subjects with small nerve fiber damage and showed excellent diagnostic utility for DN, confirming previous studies in diabetic retinopathy and nephropathy [27, 29]. Furthermore, miR‐27a‐3p levels correlated with CNFD (r = −0.40, p < 0.05) and CNFL (r = −0.49, p < 0.01) in subjects with DN. In our recent study, we showed downregulation of Netrin‐1, an anti‐inflammatory neurotropic factor in subjects with early DN, which correlated with CNFL loss [18]. In this study, we also show that serum Netrin‐1 levels were significantly lower in the DN+ compared to the DN‐ and control groups. Target analysis for miR‐27a‐3p using targetscan.org revealed a predicted target region in the 3’UTR (position 435–442) of Netrin‐1 mRNA, and indeed miR‐27a‐3p expression correlated with Netrin‐1 levels in subjects with DN.
Neuroinflammation appears to be critical in the development and progression of DN [30], and indeed, levels of pro‐inflammatory cytokines, especially TNFα and IL‐1β, are closely associated with DN [30, 31]. We confirm that circulating levels of TNFα are significantly higher and levels of the anti‐inflammatory cytokine IL‐10 are lower in subjects with DN. Indeed, a recent study has shown that miR‐27a‐3p promotes inflammation and is associated with elevated TNFα, IL‐1β, IL‐6, and IFNγ [32], whilst Netrin‐1 suppresses the production of inflammatory cytokines (IL‐6, IL‐1β, TNF‐α) by helper T‐lymphocytes [33]. NF‐κB plays a key role in inflammation [34], mediates the production of TNFα, IL‐6, and IL‐1, and may contribute to the development of neurodegeneration [35]. We show that NF‐κB expression was almost threefold higher in subjects with DN. miR‐27a‐3p may induce inflammation by increasing NF‐κB and TNFα expression [36], and indeed, there was a correlation between miR‐27a‐3p and NF‐κB expression in subjects with DN+. Netrin‐1 suppresses the production of inflammatory cytokines and chemokines [33] and inhibits NF‐κB and cyclooxygenase‐2 (COX‐2) expression [37], counteracting inflammation. In our study, we found that circulating netrin‐1 levels were negatively associated with NF‐κB expression, consistent with previous findings.
Upregulation of NLRP3 (NLRP3—NOD‐, LRR‐ and pyrin domain‐containing protein 3) inflammasome in the sciatic nerve and dorsal root ganglion (DRG) has been reported in painful diabetic neuropathy [38]. The priming signal (Signal1) is the NF‐kB pathway that upregulates NLRP3 and pro‐IL‐1β expression [39]. Once formed, the NLRP3 inflammasome recruits and activates caspase‐1, which cleaves pro‐IL‐1β and pro‐IL‐18 to produce IL‐1β and IL‐18 [40], leading to inflammatory cell death. In this study, we show that NLRP3 expression and IL‐1β levels were significantly higher in subjects with small nerve fiber loss, and miR‐27a‐3p expression positively correlated with NLRP3 expression and negatively with Netrin‐1 level.
Our study has several limitations. First, although BMI and body fat% were not different between groups and did not correlate with molecular markers, residual confounding by obesity cannot be excluded. Second, detailed treatment stratification (e.g., insulin, metformin, SGLT2i, GLP‐1RA, statins) was not available for all participants and could not be systematically analyzed; future work should address treatment‐specific effects. Third, while individuals with overt CKD were excluded, we did not systematically measure eGFR/albuminuria in all participants; thus, subtle renal effects on Netrin‐1 cannot be ruled out. Finally, the single‐center cross‐sectional design limits generalizability. Despite these limitations, the observed associations between miR‐27a‐3p, Netrin‐1, and corneal nerve damage remained robust.
Altogether, our experiments show that Netrin‐1 acting via the NF‐kB/NLRP3/IL‐1β pathway, together with increased miR‐27a‐3p expression acting via NF‐kB and NLRP3, plays a key role in inflammation and small nerve fiber loss. The assessment of miR‐27a/Netrin‐1 levels may aid in the diagnosis of DN and may also act as a therapeutic target for preventing neuroinflammation and DN.
Author Contributions
Asif Mondal: conceptualization, methodology, investigation, formal analysis, data curation, writing – original draft. Sucharita Banerjee: methodology, investigation, data curation, funding acquisition, formal analysis. Somnath Naskar: validation, supervision, visualization, methodology. Chiranjit Bose: methodology, formal analysis, investigation, data curation. Chinmay Saha: software, validation, formal analysis. Subhasish Pramanik: methodology, software, formal analysis. Bidisha Mukherjee: data curation, investigation. Piyali Jati: data curation, investigation. Pranab Kumar Sahana: validation, supervision. Rayaz A. Malik: supervision, writing – review and editing. Satinath Mukhopadhyay: supervision, conceptualization, validation, visualization, writing – review and editing.
Funding
The study was funded by the Indian Council of Medical Research (ICMR). Sanction No. ICMR/22/3356/SGP‐2024.
Disclosure
Approval of the research protocol: The study was approved by the Institutional Ethics Committee.
Approval Date of Registry and the Registration No. of the Study/Trial: (Ethical approval number: IPGME&R/IEC/2024/0821 dated 16th Dec, 2024).
Animal Studies: N/A.
Consent
All study participants provided written informed consent.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1: A representative CCM image of the central corneal sub‐basal nerve plexus (A) and the corresponding image.
Figure S2: Warm (WPT), cold (CPT), and vibration (VPT) perception thresholds in controls, DN–, and DN+.
Table S3: Correlation of Netrin‐1 with demographic and clinical variables.
Table S4: Correlation of miR‐27a‐3p with demographic and clinical variables.
Figure S5: Ordinal severity analysis by CNFL tertiles across all participants, demonstrating a stepwise increase.
Acknowledgments
The authors thank all the volunteers for their participation. The authors thank the Department of Biochemistry, IPGMER, Kolkata, India, for their assistance with biochemical analysis. The authors also thank the Multi‐Disciplinary Research Unit (MRU) of IPGME&R for access to the qRT‐PCR system. Some of the data were presented as an abstract at the 60th EASD annual meeting in Madrid, Spain, 2024.
Mondal A., Banerjee S., Naskar S., et al., “ miR‐27a‐3p Upregulation and Netrin‐1 Deficiency Drive Inflammatory Responses and Nerve Degeneration in Patients With Diabetes,” European Journal of Neurology 32, no. 12 (2025): e70449, 10.1111/ene.70449.
Contributor Information
Asif Mondal, Email: asif.mondal262@gmail.com.
Satinath Mukhopadhyay, Email: satinath.mukhopadhyay@gmail.com.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Abbott C. A., Malik R. A., van Ross E. R., Kulkarni J., and Boulton A. J., “Prevalence and Characteristics of Painful Diabetic Neuropathy in a Large Community‐Based Diabetic Population in the U.K.,” Diabetes Care 34, no. 10 (2011): 2220–2224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Ponirakis G., Elhadd T., Al Ozairi E., et al., “Prevalence and Risk Factors for Diabetic Peripheral Neuropathy, Neuropathic Pain and Foot Ulceration in the Arabian Gulf Region,” Journal of Diabetes Investigation 13, no. 9 (2022): 1551–1559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Katulanda P., Ranasinghe P., Jayawardena R., Constantine G. R., Sheriff M. H., and Matthews D. R., “The Prevalence, Patterns and Predictors of Diabetic Peripheral Neuropathy in a Developing Country,” Diabetology & Metabolic Syndrome 4, no. 1 (2012): 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Kuate‐Tegueu C., Temfack E., Ngankou S., Doumbe J., Djientcheu V. P., and Kengne A. P., “Prevalence and Determinants of Diabetic Polyneuropathy in a Sub‐Saharan African Referral Hospital,” Journal of the Neurological Sciences 355, no. 1–2 (2015): 108–112. [DOI] [PubMed] [Google Scholar]
- 5. Malik R. A., Andag‐Silva A., Dejthevaporn C., et al., “Diagnosing Peripheral Neuropathy in South‐East Asia: A Focus on Diabetic Neuropathy,” Journal of Diabetes Investigation 11, no. 5 (2020): 1097–1103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gad H., Kalra S., Pinzon R., et al., “Earlier Diagnosis of Peripheral Neuropathy in Primary Care: A Call to Action,” Journal of the Peripheral Nervous System 29, no. 1 (2024): 28–37. [DOI] [PubMed] [Google Scholar]
- 7. Malik R. A., Kallinikos P., Abbott C. A., et al., “Corneal Confocal Microscopy: A Non‐Invasive Surrogate of Nerve Fibre Damage and Repair in Diabetic Patients,” Diabetologia 46, no. 5 (2003): 683–688. [DOI] [PubMed] [Google Scholar]
- 8. Sivaskandarajah G. A., Halpern E. M., Lovblom L. E., et al., “Structure‐Function Relationship Between Corneal Nerves and Conventional Small‐Fiber Tests in Type 1 Diabetes,” Diabetes Care 36, no. 9 (2013): 2748–2755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Petropoulos I. N., Alam U., Fadavi H., et al., “Rapid Automated Diagnosis of Diabetic Peripheral Neuropathy With in Vivo Corneal Confocal Microscopy,” Investigative Ophthalmology & Visual Science 55, no. 4 (2014): 2071–2078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Gad H., Elgassim E., Lebbe A., et al., “Corneal Confocal Microscopy Detects Early Nerve Regeneration After Pharmacological and Surgical Interventions: Systematic Review and Meta‐Analysis,” Journal of the Peripheral Nervous System 29, no. 2 (2024): 173–184. [DOI] [PubMed] [Google Scholar]
- 11. Malik R. A., “Wherefore Art Thou, O Treatment for Diabetic Neuropathy?,” International Review of Neurobiology 127 (2016): 287–317. [DOI] [PubMed] [Google Scholar]
- 12. Dewanjee S., Das S., Das A. K., et al., “Molecular Mechanism of Diabetic Neuropathy and Its Pharmacotherapeutic Targets,” European Journal of Pharmacology 833 (2018): 472–523. [DOI] [PubMed] [Google Scholar]
- 13. Eid S. A., Rumora A. E., Beirowski B., et al., “New Perspectives in Diabetic Neuropathy,” Neuron 111, no. 17 (2023): 2623–2641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Rajasekharan S. and Kennedy T. E., “The Netrin Protein Family,” Genome Biology 10, no. 9 (2009): 239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Dun X. P. and Parkinson D. B., “Role of Netrin‐1 Signaling in Nerve Regeneration,” International Journal of Molecular Sciences 18, no. 3 (2017): 491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Podjaski C., Alvarez J. I., Bourbonniere L., et al., “Netrin 1 Regulates Blood‐Brain Barrier Function and Neuroinflammation,” Brain 138, no. Pt 6 (2015): 1598–1612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Yim J., Kim G., Lee B. W., et al., “Relationship Between Circulating Netrin‐1 Concentration, Impaired Fasting Glucose, and Newly Diagnosed Type 2 Diabetes,” Front Endocrinol (Lausanne) 9 (2018): 691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Mondal A., Bose C., Pramanik S., et al., “Circulating Netrin‐1 Levels Are Reduced and Related to Corneal Nerve Fiber Loss in Patients With Diabetic Neuropathy,” Journal of Diabetes Investigation 15, no. 8 (2024): 1068–1074. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Catalanotto C., Cogoni C., and Zardo G., “MicroRNA in Control of Gene Expression: An Overview of Nuclear Functions,” International Journal of Molecular Sciences 17, no. 10 (2016): 1712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Chen T., Zhang Y., Liu Y., et al., “MiR‐27a Promotes Insulin Resistance and Mediates Glucose Metabolism by Targeting PPAR‐γ‐Mediated PI3K/AKT Signaling,” Aging (Albany NY) 11, no. 18 (2019): 7510–7524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Zhou T., Meng X., Che H., et al., “Regulation of Insulin Resistance by Multiple MiRNAs via Targeting the GLUT4 Signalling Pathway,” Cellular Physiology and Biochemistry 38, no. 5 (2016): 2063–2078. [DOI] [PubMed] [Google Scholar]
- 22. Karolina D. S., Tavintharan S., Armugam A., et al., “Circulating miRNA Profiles in Patients With Metabolic Syndrome,” Journal of Clinical Endocrinology & Metabolism 97, no. 12 (2012): E2271–E2276. [DOI] [PubMed] [Google Scholar]
- 23. He Y., Ding Y., Liang B., et al., “A Systematic Study of Dysregulated MicroRNA in Type 2 Diabetes Mellitus,” International Journal of Molecular Sciences 18, no. 3 (2017): 456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Radojičić O., Dobrijević Z., Robajac D., et al., “Gestational Diabetes Is Associated With an Increased Expression of miR‐27a in Peripheral Blood Mononuclear Cells,” Molecular Diagnosis & Therapy 26, no. 4 (2022): 421–435. [DOI] [PubMed] [Google Scholar]
- 25. Natalicchio A., Montagnani M., Gallo M., et al., “MiRNA Dysregulation Underlying Common Pathways in Type 2 Diabetes and Cancer Development: An Italian Association of Medical Oncology (AIOM)/Italian Association of Medical Diabetologists (AMD)/Italian Society of Diabetology (SID)/Italian Society of Endocrinology (SIE)/Italian Society of Pharmacology (SIF) Multidisciplinary Critical View,” ESMO Open 8, no. 3 (2023): 101573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Han L. L., Wang S. H., Yao M. Y., and Zhou H., “Urinary Exosomal microRNA‐145‐5p and microRNA‐27a‐3p Act as Noninvasive Diagnostic Biomarkers for Diabetic Kidney Disease,” World Journal of Diabetes 15, no. 1 (2024): 92–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Lazzara F., Trotta M. C., Platania C. B. M., et al., “Stabilization of HIF‐1α in Human Retinal Endothelial Cells Modulates Expression of miRNAs and Proangiogenic Growth Factors,” Frontiers in Pharmacology 11 (2020): 1063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Kim J., Fiesel F. C., Belmonte K. C., et al., “miR‐27a and miR‐27b Regulate Autophagic Clearance of Damaged Mitochondria by Targeting PTEN‐Induced Putative Kinase 1 (PINK1),” Molecular Neurodegeneration 11, no. 1 (2016): 55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Wu L., Wang Q., Guo F., et al., “Involvement of miR‐27a‐3p in Diabetic Nephropathy via Affecting Renal Fibrosis, Mitochondrial Dysfunction, and Endoplasmic Reticulum Stress,” Journal of Cellular Physiology 236, no. 2 (2021): 1454–1468. [DOI] [PubMed] [Google Scholar]
- 30. Cox A. A., Sagot Y., Hedou G., et al., “Low‐Dose Pulsatile Interleukin‐6 as a Treatment Option for Diabetic Peripheral Neuropathy,” Front Endocrinol (Lausanne) 8 (2017): 89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Huang Y., Zang Y., Zhou L., Gui W., Liu X., and Zhong Y., “The Role of TNF‐Alpha/NF‐Kappa B Pathway on the Up‐Regulation of Voltage‐Gated Sodium Channel Nav1.7 in DRG Neurons of Rats With Diabetic Neuropathy,” Neurochemistry International 75 (2014): 112–119. [DOI] [PubMed] [Google Scholar]
- 32. Chen Y., Li S., and He H., “miR‐27a‐3p Promotes Inflammatory Response in Infectious Endophthalmitis via Targeting TSC1,” Scientific Reports 14, no. 1 (2024): 19353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Tadagavadi R. K., Wang W., and Ramesh G., “Netrin‐1 Regulates Th1/Th2/Th17 Cytokine Production and Inflammation Through UNC5B Receptor and Protects Kidney Against Ischemia‐Reperfusion Injury,” Journal of Immunology 185, no. 6 (2010): 3750–3758. [DOI] [PubMed] [Google Scholar]
- 34. Mattson M. P. and Camandola S., “NF‐kappaB in Neuronal Plasticity and Neurodegenerative Disorders,” Journal of Clinical Investigation 107, no. 3 (2001): 247–254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Smith J. A., Das A., Ray S. K., and Banik N. L., “Role of Pro‐Inflammatory Cytokines Released From Microglia in Neurodegenerative Diseases,” Brain Research Bulletin 87, no. 1 (2012): 10–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Wang Z., Ruan Z., Mao Y., et al., “miR‐27a Is Up Regulated and Promotes Inflammatory Response in Sepsis,” Cellular Immunology 290 (2014): 190–195. [DOI] [PubMed] [Google Scholar]
- 37. Natura G., Bär K. J., Eitner A., et al., “Neuronal Prostaglandin E2 Receptor Subtype EP3 Mediates Antinociception During Inflammation,” Proceedings of the National Academy of Sciences of the United States of America 110, no. 33 (2013): 13648–13653. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Chen W., Wang X., Sun Q., et al., “The Upregulation of NLRP3 Inflammasome in Dorsal Root Ganglion by Ten‐Eleven Translocation Methylcytosine Dioxygenase 2 (TET2) Contributed to Diabetic Neuropathic Pain in Mice,” Journal of Neuroinflammation 19, no. 1 (2022): 302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Jo E. K., Kim J. K., Shin D. M., and Sasakawa C., “Molecular Mechanisms Regulating NLRP3 Inflammasome Activation,” Cellular & Molecular Immunology 13, no. 2 (2016): 148–159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Chen C. and Smith M. T., “The NLRP3 Inflammasome: Role in the Pathobiology of Chronic Pain,” Inflammopharmacology 31, no. 4 (2023): 1589–1603. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: A representative CCM image of the central corneal sub‐basal nerve plexus (A) and the corresponding image.
Figure S2: Warm (WPT), cold (CPT), and vibration (VPT) perception thresholds in controls, DN–, and DN+.
Table S3: Correlation of Netrin‐1 with demographic and clinical variables.
Table S4: Correlation of miR‐27a‐3p with demographic and clinical variables.
Figure S5: Ordinal severity analysis by CNFL tertiles across all participants, demonstrating a stepwise increase.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.