Exploring circRNA_19601 as a therapeutic target in diabetic neuropathic pain: Insights from rat and cell models

Abstract

Background:

Diabetic neuropathic pain (DNP) is a common complication of diabetes with significant impact on patients’ quality of life. Current treatments have limitations, and exploring new mechanisms and targets is crucial.

Objective

Our study aims to explore the role and mechanism of circRNA_19601 (circ19601) in regulating miR-324-5p in DNP within the dorsal root ganglion (DRG).

Methods:

This study used streptozotocin (STZ) and high-fat diet-induced diabetic rat models, as well as high-sugar treated DRG neurons to construct cell models. The STZ and high-fat diet induced a diabetic neuropathic pain model in rats, and high-glucose conditions were used to extract DRG neurons. The effects of the positive control drug pregabalin on STZ rats were monitored under different dosing conditions by measuring body weight, blood glucose, mechanical paw withdrawal threshold, and thermal paw withdrawal latency. The study also analyzed the expression of circular RNA in DRG neurons affected by diabetic neuropathic pain. Next, we also examined the effects of knocking down circ19601 with or without the miR-324-5p inhibitor on DNP. The expression of pain-related membrane proteins was analyzed using Western blot.

Results:

STZ treatment in diabetic rats led to reduced body weight, elevated blood glucose, and decreased pain sensitivity. Pregabalin effectively improved mechanical hyperalgesia but mainly influenced mechanical sensitivity long-term. Transcriptomic analysis revealed upregulation of circ19601 in diabetic rats, which was reversed by pregabalin. Knockdown of circ19601 improved body weight, reduced blood glucose, and alleviated pain sensitivity by increasing miR-324-5p levels and decreasing neurotransmitter and pain-related protein levels. MiR-324-5p inhibition reversed these effects, highlighting its role in regulating pain pathways in diabetic neuropathy.

Conclusion:

Pregabalin mitigates mechanical and thermal pain in diabetic rats. It does so by reversing decreased pain thresholds and modifying the expression of circ19601. This, in turn, impacts miR-324-5p and pain-related proteins, leading to improvements in body weight and blood glucose levels.

Keywords

Diabetic neuropathic pain dorsal root ganglia circ19601 miR-324-5p

Highlights

Pregabalin shows efficacy in diabetic rats, improving mechanical and thermal pain thresholds, indicating its potential for diabetic neuropathic pain treatment.

The expression of circRNA_19601 is significantly altered in diabetic neuropathic pain models, being elevated in the model group and reduced by pregabalin treatment.

Knockdown of circRNA_19601 and its interaction with miR-324-5p influence pain -related proteins and physiological parameters, uncovering a potential regulatory mechanism for diabetic neuropathic pain.

Introduction

Diabetic neuropathic pain (DNP), a prevalent complication of diabetes, is characterized by persistent aching, tingling, or burning sensations, severely affecting patients’ daily activities and mental health.^1–3 The pathogenesis of DNP is complex, involving a variety of biochemical and neurobiological changes.^4,5 Pathologically, hyperglycemic states initially cause vascular dysfunction in peripheral nerves, leading to nerve ischemia and hypoxia, and ultimately to nerve fiber damage.⁶ Additionally, high blood sugar levels can directly damage nerve cells, triggering inflammatory responses and the release of cytokines and chemokines, which can further exacerbate nerve damage and pain.⁷

Current treatments strategies for diabetic neuropathic pain primarily include pharmacological and physical therapies. Commonly prescribed medications include antidepressants, anticonvulsants, and opioids, which aim to alleviate pain symptoms.^8,9 However, these interventions typically offer only temporary relief and fail to halt the progression of nerve damage. Furthermore, prolonged use of these drugs is associated with various adverse effects, including dependence, tolerance, and systemic complications.^10,11

Pregabalin, an anticonvulsant, binds to the α2δ subunit of calcium channels, inhibiting neurotransmitter release and reducing pain signals.^12,13 Clinical trials demonstrate its efficacy in alleviating DNP symptoms and improving functional status.¹⁴ Clinical trials demonstrate its efficacy in alleviating DNP symptoms and improving functional status. In addition to diabetic neuropathic pain, pregabalin has gained widespread clinical use for the treatment of postherpetic neuralgia (PHN), a chronic pain condition following herpes zoster. Its mechanism of action involves binding to the α2δ subunit of voltage-gated calcium channels in the central nervous system, thereby reducing the release of neurotransmitters such as glutamate, norepinephrine, and substance P, and ultimately diminishing pain transmission.^15,16 Pregabalin’s ability to modulate pain pathways and potentially promote nerve repair makes it a promising treatment option.¹⁷

Circular RNAs (circRNAs) represent a distinct class of non-coding RNAs that play critical roles in the regulation of gene expression. Recent studies have shown that circRNAs are particularly important in neurological diseases, potentially affecting neural function by influencing the activity of miRNAs and participating in protein coding.^18–20 In the context of diabetic neuropathic pain, specific circRNAs, such as circ19601 (Chr4: 95149591-95205604, NCBI Reference Sequence: XM_039108695.1), are believed to affect the onset of neuropathic pain by regulating miRNAs like miR-324-5p.^21–23 Moreover, transcriptional and post-transcriptional regulation of pain mediators and receptors has been shown to involve circRNAs. For instance, by modulating miR-324-5p expression, circ19601 may influence the activity of voltage-gated ion channels and inflammatory cytokines, both of which are critical drivers of neuronal sensitization in DNP.²⁴ Our study not only reveals how circ19601 regulates pain-related protein expression by modulating miR-324-5p but also demonstrates its role in regulating mechanical and thermal pain thresholds. Additionally, the research examines the efficacy of pregabalin in alleviating diabetic neuropathic pain, further uncovering the potential role of circ19601 in the modulation of pain mechanisms.

The dorsal root ganglion (DRG) is essential in neuropathic pain, acting as a key center for sensory neurons that relay pain signals from the peripheral nervous system to the central nervous system. Within the DRG, neuronal responses to injury or disease, such as diabetic neuropathy, can become hyperactive, leading to exaggerated pain sensations.²⁵ miRNAs are key regulators of these processes, as they modulate the expression of genes involved in neuronal excitability and inflammatory responses, thereby influencing pain sensitivity.^26–28 Targeting dysregulated miRNAs in the DRG under disease conditions may thus provide new therapeutic opportunities for neuropathic pain.

In light of the complexity of DNP and the constraints of current treatments, there is a pressing need for novel therapeutic strategies. This study focuses on Pregabalin as a positive control drug and aims to screen for potential circRNA therapeutic targets. By investigating the molecular mechanisms underlying the role of circ19601/miR-324-5p in diabetic neuropathic pain, we hope to uncover new avenues for more effective treatment of DNP.

Methods and materials

Animals

Adult Sprague-Dawley rats weighing 220–230 g were used in this study. Animals were housed in a pathogen-free environment under a 12 h light/dark cycle with controlled temperature and humidity. They were allowed free access to food and water and acclimatized for at least 1 week before any experimental procedures. Baseline behavioral assessments were conducted using a double-blind method. The animal grouping Settings were as follows: control group (Veh), model group, and positive drug (pregabalin) group. Each group consisted of six rats from which DRG tissues were collected.

Induction of a diabetic neuropathy rat model and treatment

After a 12 h fasting period, male SD rats (220–230 g) were intraperitoneally injected with streptozotocin (STZ, 35 mg/kg) and simultaneously fed a high-fat diet to accelerate the development of diabetes and neuropathic pain. Following STZ injection, rats received penicillin sodium to prevent infection, bedding was replaced daily, and food and water were provided ad libitum. At week 7 post-STZ injection, pregabalin treatment (50 mg/kg) was initiated in the 9th week and maintained for 12 h. Eight days after treatment, all rats were euthanized. During the study, rat body weight, blood glucose levels, mechanical paw withdrawal threshold (PWT), and thermal paw withdrawal latency (TWL) were measured weekly.²⁹

Blood glucose levels were measured using a handheld glucose meter with compatible test strips. Select the rat tail vein blood collection, disinfect it with an alcohol cotton ball before blood collection, and after the alcohol evaporates, squeeze the tail to fill the tail vein, pierce it quickly with a disposable blood collection needle, drop a drop of blood in the designated position of the test strip, insert the test strip into the blood glucose meter, and wait for the reading according to the operation prompts. During the process, ensure that the test strip is in good contact with the blood glucose meter.

Knockdown of circ19601 in vivo

To selectively knock down circ19601 expression in the central nervous system of diabetic neuropathy rats, a recombinant adeno-associated virus vector (AAV9-U6-sh-circRNA19601) was constructed by Anhui General Biological Co., Ltd. The shRNA sequence targeting circ19601 was 5′-agatctGGGAGAAAGAGTTCAAATAtacctgacccataTATTTGAACTCTTTCTCCCTTTTTggtacc-3′.

Adult rats were anesthetized with sodium pentobarbital (40 mg/kg, intraperitoneally), and the viral suspension (1 × 10¹² vg/ml) was intrathecally injected into the L5–L6 intervertebral space at a rate of approximately 1 μl/10 s, with a total injection volume of 10 μl per rat. To enhance transduction efficiency, 4 μg/ml polybrene was added before injection. Fourteen days post-injection, three rats from each group were sacrificed, and dorsal root ganglia were collected to validate knockdown efficiency by RT-qPCR.

Thermal paw withdrawal latency (TWL)

Rats were placed individually on a plexiglass platform within an observation chamber. Once exploratory activity ceased, a radiant heat source was aligned with the plantar surface of the hind paw. Adjust the heat source to achieve a withdrawal latency of 8–15 s for rats with normal heat pain thresholds, setting a maximum exposure time of 15 s to prevent burns. Maintain the platform and ambient temperature at 26 ± 0.5°C. Start the heat exposure when the rat is still, recording the time until the rat lifts its paw. Measure each foot three times with 10-min intervals, and calculate the average latency.

Mechanical paw withdrawal threshold (PWT)

Determine the mechanical withdrawal threshold using Von Frey filaments on the rat’s hind paws. Place the rat on a metal mesh covered with a plexiglass box. After a 30-min acclimation period, apply increasing intensities of filaments (2–26 g) vertically to the central area of the hind paw, pressing for 6–9 s. Record a positive response (paw lift) as “X” and proceed to the next filament if there is no response. Repeat this three times per hind paw with 10-min intervals and calculate the threshold.

Omics data analysis and bioinformatics analysis

The locally conducted searches for the sequences of the identified genes with differential expression were carried out using InterProScan to pinpoint analogous sequences. Following this, the Gene Ontology (GO) terms were plotted, and the sequences received annotation through the Blast2GO software application. Proteins under investigation were processed through BLAST searches in the Kyoto Encyclopedia of Genes and Genomes (KEGG) online repository (accessible at https://www.genome.jp/kegg/) to acquire their KEGG annotations, which were then aligned with the pathways found within the KEGG database. Protein-Protein Interaction (PPI) data for these proteins were sourced from the IntAct molecular interaction database by utilizing their gene symbols or via the STRING application. The acquired data in XGMML file format were transferred into Cytoscape for graphical representation and for an in-depth examination of the functional PPI networks.

ELISA

The concentrations of substance P (catalog number ab288318, sourced from Abcam, United Kingdom), calcitonin gene-related peptide (CGRP; identified by MBS267126, obtained from MyBioSource), norepinephrine (NE; listed as ab287789, Abcam, UK), and cortisol (referred to as Cor; with the product code ab108665, from Abcam, UK) within rat serum were assayed utilizing ELISA kits as per the provided instructions. The absorbance readings were obtained at 450 nm with a microplate reader (3550 model by Bio-Rad), and a calibration curve was generated from the standard wells to determine the levels of the respective markers present in the sample wells.

Cell culture and treatment

The rat dorsal root ganglion (DRG) neurons were purchased from Shanghai Yaji Biotechnology Co., Ltd. (Catalog No. H10995). The neurons were cultured in a humidified atmosphere at 37°C with 5% CO₂. The recommended culture medium was Neurobasal™ medium supplemented with B27 and Glutamax. Cells were passaged every 2–3 days and maintained at a density of approximately 1 × 10⁶ cells/ml to ensure optimal growth conditions. The cell grouping was set as follows: Control group, HG, HG+sh-NC, HG+sh-circ19601, HG+sh-circ19601+NC inhibitor, HG+sh-circ19601+miR inhibitor.

Dual luciferase assay

Neural cells from the dorsal root ganglion were placed in a 24-well culture plate with a seeding density of roughly half. Initially, these cells were simultaneously transfected with either a wild-type or a mutated vector in conjunction with either miR-324-5p or a negative control, paired with the pRL-TK reporter construct. Subsequently, the DRG cells underwent processing with a dual-luciferase assay system (manufactured by Promega, based in the United States) and the luciferase activity was quantified utilizing a multifunctional microplate reader 24 h post-transfection.

Western blot

The dorsal root ganglia were meticulously dissected on ice and subsequently homogenized in a lysis solution supplemented with a protease inhibitor, PMSF (P7626, Sigma-Aldrich, USA), utilizing a high-frequency ultrasonic disruptor (30% amplitude, 10-s pulses with 5-s intervals, repeated six times; Lichen, Beijing, China). Post-homogenization, the resulting lysates underwent centrifugation at a force of 13,500 × g for a duration of 15 min at a temperature of 4°C. The resultant supernatants were adjusted to a uniform protein concentration as determined by the Pierce™ BCA Protein Assay Kit (23225, Thermo Fisher Scientific, USA). These lysate samples were then diluted in lysis buffer containing 5% sample buffer and subjected to boiling for 10 min at a temperature of 100°C. Subsequently, identical quantities of protein were subjected to separation via 15% SDS-PAGE under a steady voltage of 80 V for 90 min. The resolved proteins were subsequently transferred onto PVDF membranes, and to prevent non-specific binding, the membranes were blocked by submersion in a solution of 5% non-fat milk in TBST for 2 h at ambient temperature within a bioshaking incubator. Overnight at 4°C, the membranes were probed with primary antibodies targeting Nav1.7 (1:1000, ab65167, Abcam, UK), TRPV1 (1:1000, ab305299, Abcam, UK), COX-2 (1:1000, ab179800, Abcam, UK) and GAPDH (1:5000, ab8245, Abcam, UK). Following three rinses with TBST buffer, each lasting 5 min, the membranes were exposed to a goat anti-rabbit IgG H&L (HRP) secondary antibody (1:2000, ab7090, Abcam, UK) for 2 h at room temperature, after which they were washed an additional four times with TBST buffer. The protein bands were subsequently made visible using an enhanced chemiluminescence (ECL) detection system, and the band intensities were quantified using Image Lab software version 6.0 (Bio-Rad Laboratories, USA).³⁰

Reverse transcription–quantitative polymerase chain reaction (RT–qPCR)

The rat-derived DRG tissues from the control group were finely minced utilizing an advanced tissue homogenizer. Following homogenization, each tissue sample was supplemented with 1 ml of TRIzol reagent for the isolation of comprehensive RNA. Subsequent to RNA conversion, quantitative reverse transcription polymerase chain reaction was carried out using the TaKaRa reverse transcription kit.³¹ Following this, cDNA synthesis was carried out with a cDNA synthesis kit (D7168S, Beyotime, Beijing, China). PCR reactions were conducted on the ABI 7900 fluorescence quantitative PCR instrument (ABI, USA). The primer sequences used were as follows: circ19601 (rat): forward 5′-CAGGTTTGTGTGCAAGGTCG-3′ and reverse 5′-TGCTTACATGTCCTGGTGTGA-3′, miR-324-5p (rat): forward 5′-AACAATCGCATCCCCTAGGG-3′ and reverse 5′-GTCGTATCCAGTGCAGGGT-3′, GAPDH (rat): forward 5′-TGTGTCCGTCGTGGATCTGA-3′ and reverse 5′-CCTGCTTCACCACCTTCTTG-3′. The procedure began with an initial denaturation step at 95°C for 10 min. This was followed by 40 cycles consisting of denaturation at 95°C for 10 s, annealing at either 52.8°C or 56.2°C for 15 s, and extension at 72°C for 20 s. A temperature gradient from 72°C to 95°C was employed, increasing in 1°C increments per step, and gain calibration was automatically performed before the first run. GAPDH expression served as an internal control for mRNA quantification using the 2^-ΔΔCT method.

Data analysis

Differentially expressed circRNAs (DE-circRNAs) were identified using the DESeq2 package in R. Raw read counts were first normalized to account for library size and sequencing depth. CircRNAs with extremely low expression (average counts <10 across all samples) were filtered out to reduce noise. Statistical significance was determined using the Benjamini–Hochberg method to control the false discovery rate (FDR). CircRNAs with an adjusted p-value < 0.05 and an absolute log2 fold change (|log2FC|) ≥1 were considered significantly differentially expressed. DE-circRNAs were identified separately for each comparison (Veh vs Model and Model vs Pregabalin), and these were subsequently used for downstream analyses including Venn diagrams, volcano plots, heatmaps, GO, and KEGG pathway analyses.

For other experimental data (body weight, blood glucose, behavioral tests, ELISA, etc.), statistical analysis was conducted using Prism and SPSS software. Results are expressed as means ± SD (standard deviation). A t-test was employed to compare variances between two groups, while one-way ANOVA was used to assess differences among three or more groups. A significance level of p < 0.05 was considered statistically significant.

Results

Effects of streptozotocin (STZ) on body weight, blood glucose levels, and pain sensitivity in diabetic rats

Figure 1(a) is the construction flow chart of the animal model. Diabetic rats received a streptozotocin (STZ) injection in week 7. In week 9, the rats were treated with pregabalin for 12 h, and 8 days later, all the rats were euthanized. We then observed changes in body weight, blood glucose levels, paw withdrawal threshold, and thermal withdrawal latency in both control and diabetic rats.

Figure 1.

Effects of streptozotocin (STZ) on body weight, blood glucose levels, and pain sensitivity in diabetic rats. (a) Study design: STZ was administered to diabetic rats in week 7, followed by pregabalin treatment in week 9, and sacrifice of rats 8 days later. N = 6 rats/group. (b) Body weight changes in model and veh groups. (c) Blood glucose levels. (d) Paw withdrawal threshold. (e) Thermal withdrawal latency.

In the control group, body weight remained stable and continued to increase. After STZ treatment in week 7, the weight gain in diabetic rats began to slow down. By week 10, the weight of the diabetic rats was significantly lower than that of the control group (p < 0.05, Figure 1(b)). Before week 6, blood glucose levels showed no statistically significant differences between the control and diabetic rats. However, after STZ treatment in week 7, blood glucose levels in the diabetic rats rose sharply. By week 9, these levels were significantly elevated compared to the control group (p < 0.001, Figure 1(c)).

Prior to week 6, there were no statistically significant differences in either paw withdrawal threshold or thermal withdrawal latency between the control and diabetic rats. After STZ treatment in week 7, both the paw withdrawal threshold and thermal withdrawal latency in diabetic rats decreased sharply by week 8. By week 9, these measures were significantly lower in the diabetic rats compared to the control group (p < 0.001, Figure 1(d) and (e)). Overall, diabetic rats displayed a decrease in body weight, an increase in blood glucose levels, and a reduction in both paw withdrawal threshold and thermal withdrawal latency after STZ treatment, showing significant differences from the control rats by week 9.

Pregabalin improves mechanical hyperalgesia in diabetic rats

Pregabalin, a neuroregulatory medication, effectively alleviates mechanical hypersensitivity in rats both in the short and long term. Control group rats and diabetic rats received either pregabalin or a solvent control treatment for 12 h. As shown in Figure 2(a), the paw withdrawal threshold in diabetic rats significantly decreased compared to the control group (p < 0.001). However, pregabalin treatment reversed this reduced threshold after 12 h (p < 0.001). Analysis of the area under the curve (AUC) for the paw withdrawal threshold revealed a significantly lower AUC in diabetic rats (p < 0.001), which was reversed by pregabalin treatment (p < 0.001, Figure 2(b)). Similarly, the thermal withdrawal latency in diabetic rats significantly decreased compared to the control group (p < 0.01). This reduced latency was also reversed by pregabalin treatment after 12 h (p < 0.01, Figure 2(c)). The AUC analysis for thermal withdrawal latency showed a significantly lower AUC in diabetic rats (p < 0.001), which was reversed by pregabalin (p < 0.001, Figure 2(d)).

Figure 2.

The positive drug pregabalin can alleviate mechanical hypersensitivity and thermal hypersensitivity in diabetic rats. (a) Paw withdrawal threshold for 12 h. (b) Analysis of area under the curve (AUC) of paw withdrawal threshold. (c) Thermal withdrawal latency for 12 h. (d) Analysis of AUC of thermal withdrawal latency. (e) Paw withdrawal threshold for 8 days. (f) Analysis of AUC of paw withdrawal threshold. (g) Thermal withdrawal latency for 8 days. (h) Analysis of AUC of thermal withdrawal latency.

We also examined the long-term effects of pregabalin. After 8 days of treatment, the paw withdrawal threshold and its area under the curve (AUC) in diabetic rats were significantly reduced (p < 0.001, Figure 2(e) and (f)), but this reduction was reversed by pregabalin. However, pregabalin did not significantly improve the thermal sensitivity or its AUC in diabetic rats (p < 0.01, Figure 2(g) and (h)).

Pregabalin effectively improves mechanical and thermal sensitivity in diabetic rats in the short term, but mainly improves mechanical sensitivity in the long term.

Transcriptome profile of DRG neurons and identification of circ19601 in diabetic neuropathic pain rats

CircRNAs are a type of RNA molecule with a closed-loop structure, mainly found in the cytoplasm of eukaryotic cells. To investigate circRNA alterations in DRG neurons of diabetic rats, transcriptome analysis was performed. The circRNA distribution across chromosomes revealed that Chromosome 1 contained the highest number of circRNAs (Figure 3(a)). CircRNA lengths varied from hundreds to thousands of base pairs, with the majority ranging from 201 to 300 bp (Figure 3(b)). CircRNA lengths varied from hundreds to thousands of base pairs, with the majority ranging from 201 to 300 bp (Figure 3(b)). Based on their origin and structure, circRNAs were classified into exonic, intronic, and exon-intron types, which are known to regulate gene expression and cell signaling (Figure 3(c)). While these results provided an overview of circRNA characteristics, we further focused on differentially expressed circRNAs associated with DNP.

Figure 3.

Circular RNAomics study of DRG neurons in diabetic rat neuropathic pain. (a) Distribution of circular RNA on chromosomes. (b) Length of circular RNA. (c) Types of circular RNA.

In the comparison between the Model group and the Veh group, 844 circRNAs were upregulated and 831 were downregulated. When comparing the Pregabalin group with the Model group, 871 circRNAs were upregulated and 829 were downregulated (Figure 4(a)). Among them, 753 differentially expressed circRNAs overlapped between the 2 comparisons (Figure 4(b)). The intersection of Model-vs-Veh_up and Pregabalin-vs-Model_down had 531 genes. Further data screening (removing the data with an expression value of 0 in duplicate samples) revealed two circRNAs (circRNA_19601, circRNA_03894). The intersection of Model-vs-Veh_down and Pregabalin-vs-Model_up had 202 genes, and after further screening, one circRNA (circRNA_17205) was found (Figure 4(c)). The volcano plots demonstrated the distribution of differentially expressed circRNAs (Figure 4(d)), and the clustering heatmap illustrated the expression patterns among groups (Figure 4(e)). Importantly, compared with the Veh group, circRNA_19601 was significantly upregulated in the Model group, whereas pregabalin treatment reversed this increase, leading to a significant reduction in circRNA_19601 expression (Figure 4(f)). These findings suggest that circ19601 may play a pivotal role in the pathogenesis of DNP and could represent a potential target of pregabalin’s therapeutic effect.

Figure 4.

Circular RNA expression analysis of DRG neurons in diabetic rat neuropathic pain. (a) Number of differentially expressed circular RNAs. (b) Venn diagram analysis of differentially expressed genes between groups. (c) Venn diagram analysis was used to analyze the number of differentially expressed genes with high or low expression in the comparison between groups. (d) Volcano plot analysis of differentially expressed circRNAs. (e) Heatmap analysis of expression of differentially expressed genes. (f) Expression level of circ19601.

Functional analysis of differentially expressed circRNAs

To further explore the potential biological roles of differentially expressed circRNAs, we performed functional enrichment analyses for the two comparison groups (Veh vs Model and Model vs Pregabalin). GO functional analysis classified these circRNAs into biological processes, molecular functions, and cellular components, providing insight into their potential involvement in regulatory networks, signaling pathways, and DNP mechanisms (Figure 5(a)). Functional analysis of the top 30 go of the total differentially expressed circRNAs in Veh versus Model and Model versus Pregabalin (Figure 5(b)). Similarly, KEGG pathway enrichment analyses were carried out for both comparisons (Figure 5(c)), and the top 20 enriched KEGG pathways were identified (Figure 5(d)). For the Veh versus Model comparison, the most significantly enriched pathways were regulation of actin cytoskeleton, endocytosis, and Ras signaling pathway. For the Model versus Pregabalin comparison, the most significantly enriched pathways were non-homologous end-joining, nucleocytoplasmic transport, and chemical carcinogenesis.

Figure 5.

Functional analysis of differentially expressed circRNAs. (a) GO functional analysis of total differentially expressed circRNAs in Veh versus Model and Model versus Pregabalin groups. (b) GO functional analysis of top 30 differentially expressed circRNAs (Veh vs Model and Model vs Pregabalin). (c) KEGG pathway analysis of differentially expressed circRNAs in both comparisons (Veh vs Model and Model vs Pregabalin). (d) Top 20 ranked KEGG pathway in the Veh versus Model and Model versus Pregabalin groups.

Circ19601 binds with rno-miR-324-5p

Constructing an interaction network of the top 300 miRNAs and circRNAs is a multi-step bioinformatics process designed to reveal the regulatory relationships between these molecules. This network analysis aids in understanding gene expression regulation and disease pathogenesis by selecting the top 300 miRNAs based on high expression levels or significant biological relevance from databases (Figure 6(a)). The binding site of circ19601 and rno-miR-324-5p were predicted using the miRanda tool (https://www.bioinformatics.com.cn/local_miranda_miRNA_target_prediction_120; Figure 6(b)). The dual-luciferase reporter assay is employed to investigate the interaction between circ19601 and rno-miR-324-5p by assessing the effect of their binding on luminescence signals. The findings reveal that the presence of rno-miR-324-5p significantly decreases the ratio compared to the control group, indicating that the miRNA effectively binds to circ19601 and may inhibit its expression (Figure 6(C)).

Figure 6.

Interaction between miRNA and circRNA. (a) Network of interactions between top 300 miRNAs and circRNAs. (b) Binding site of circ19601 with rno-miR-324-5p. (c) Dual-luciferase assay to detect binding between circ19601 and rno-miR-324-5p.

Knockdown of circ19601 alleviates changes in body weight, blood glucose levels, and mechanical hyperalgesia in diabetic rats

To assess the role of circ19601 in diabetic neuropathy, we measured its expression levels across different treatment groups in a rat DNP model using RT-qPCR. Relative expression levels of circ19601 were significantly increased in the Model group compared to the Veh group (p < 0.001), indicating upregulation of circ19601 in the context of DNP (Figure 7(a)). Knockdown of circ19601 resulted in a significant decrease in circ19601 levels compared to the Model group (p < 0.001), suggesting that sh-circ19601 effectively knocked down circ19601 expression. We also used RT-qPCR to assess the impact of knocking down circ19601 on miR-324-5p expression. The expression of miR-324-5p was decreased in the Model group (p < 0.001) but was reversed by sh-circ19601 (p < 0.001, Figure 7(b)), suggesting that circ19601 plays an important role in negatively regulating miR-324-5p expression. Body weight was significantly increased in the Model group after knockdown of circ19601 in week 11 (p < 0.05, Figure 7(c)). Specifically, compared to the Model group, blood glucose levels were significantly decreased in the Model+sh-circ19601 group (p < 0.001, Figure 7(d)). Paw withdrawal threshold and thermal withdrawal latency were significantly increased after knockdown of circ19601 in the Model group during weeks 10-11 (p < 0.001, Figure 7(e) and (f)). In overall, Knockdown of circ19601 in a diabetic neuropathy model reduced its expression, increased miR-324-5p levels, improved body weight, decreased blood glucose, and alleviated mechanical and thermal pain sensitivity.

Figure 7.

Knockdown of circ19601 improves diabetic neuropathic pain. (a) Detection of circ19601 expression levels in rat models using RT-qPCR. (b) RT-qPCR analysis of the effect of circ19601 knockdown on miR-324-5p expression. (c) Impact of circ19601 knockdown on rat body weight. (d) Effect of circ19601 knockdown on rat blood glucose levels. (e) Influence of circ19601 knockdown on mechanical withdrawal threshold in rats. (f) Effect of circ19601 knockdown on thermal withdrawal latency in rats. (g) ELISA measurement of levels of pain neurotransmitters SP and CGRP in model serum. (h) ELISA detection of levels of norepinephrine (NE) and cortisol (Cor) in model serum.

Knockdown of circ19601 alleviates changes in neurotransmitter levels, and hormone levels in diabetic rats

Serum levels of neurotransmitters substance P (SP), calcitonin gene-related peptide (CGRP), norepinephrine (NE), and cortisol were analyzed by ELISA. sh-circ19601 significantly decreased serum levels of SP and CGRP in model rats (p < 0.001, Figure 7(g)). Similarly, serum levels of NE and cortisol were also decreased following knockdown of circ19601 in the model groups (p < 0.001, Figure 7(h)). In overall, knockdown of circ19601 significantly reduced serum levels of SP, CGRP, NE, and cortisol in the model rats, indicating its impact on neurotransmitter regulation in diabetic neuropathy.

Knockdown of circ19601 decreased the levels of pain-related membrane proteins in cells

The expression of circ19601 was increased in the HG-induced cell model (p < 0.001), but this was reduced by sh-circ19601 (p < 0.001, Figure 8(a)). MiR-324-5p expression decreased in the HG-induced cell model (p < 0.001) and was also reversed by sh-circ19601 (p < 0.001, Figure 8(b)). The Western blot results show that the protein levels of Nav1.7, TRPV1, and COX-2 in the HG group are significantly higher compared to the control group, but their expression is significantly reversed after sh-circ19601 treatment (p < 0.001, Figure 8(c)). These findings suggest that circ19601 may influence pain-related membrane protein expression in DNP rat models, indicating its potential role in neuropathic pain pathways.

Figure 8.

The knockdown of circ19601 affects pain-related membrane proteins in cells. (a) RT-qPCR detection of circ19601 expression. (b) RT-qPCR detection of miR-324-5p expression. (c) Western blot analysis of changes in expression of pain-related membrane proteins.

MiR-324-5p inhibitor reversed the downregulation of pain-related membrane proteins induced by sh-circ19601 in cells

Then, we further explore the role of miR-324-5p in the expression of pain-related membrane proteins. A significant decrease in miR-324-5p expression was observed in the miR inhibitor group, indicating successful suppression (p < 0.01, Figure 9(a)). Western blot analysis revealed changes in pain-related membrane proteins, such as Nav1.7, TRPV1, and COX-2. These proteins were decreased in the HG+sh-circ19601 group (p < 0.001), but their levels were reversed by the miR-324-5p inhibitor (p < 0.001, Figure 9(b) and (c)). In overall, the miR-324-5p inhibitor significantly reduced miR-324-5p expression and reversed the decrease in Nav1.7, TRPV1, and COX-2 levels caused by HG+sh-circ19601.

Figure 9.

Rescue experiment of the interaction between circ19601 and rno-miR-324-5p in pain. (a) RT-qPCR detection of miR-324-5p inhibition efficiency. (b) Western blot analysis of changes in expression of pain-related membrane proteins. (c) Statistical analysis of pain-related membrane protein expression.

Discussion

In this study, we conducted an in-depth investigation of the function of circ19601 in the dorsal root ganglia and its role in DNP, specifically how it regulates pain transmission through miR-324-5p.³² We noted a marked upregulation of circ19601 in diabetic rat models, which was strongly associated with increased pain sensitivity. Further functional assays confirmed miR-324-5p as a direct target of circ19601, revealing a regulatory network that may provide novel theoretical insights and potential therapeutic targets for the management of DNP.^33–35

STZ-induced diabetes led to significant metabolic and sensory changes in rats, including reduced body weight, elevated blood glucose levels, and decreased mechanical and thermal pain thresholds. These findings are consistent with previous research demonstrating STZ’s ability to induce diabetic neuropathy characterized by hyperglycemia and altered pain sensitivity.³⁶ The observed decrease in body weight and increase in blood glucose following STZ administration corroborate the results of other studies showing similar effects of STZ in diabetic rodent models.³⁷ Pregabalin treatment for 12 h in diabetic rats resulted in significant improvements in both mechanical hyperalgesia and thermal withdrawal latency, demonstrating its short-term effectiveness in alleviating neuropathic pain. However, after 8 days of treatment, while pregabalin continued to improve mechanical sensitivity, it did not significantly affect thermal pain sensitivity in diabetic rats. This aligns with clinical and preclinical data indicating pregabalin’s efficacy in managing diabetic neuropathic pain by modulating calcium channels and reducing neuronal excitability.¹⁷ The ability of pregabalin to restore both mechanical and thermal pain thresholds supports its role as a valuable therapeutic option for DNP.

Our analysis revealed that circ19601 was significantly upregulated in diabetic rats and that this upregulation was reversed by pregabalin treatment. This observation highlights circ19601 as a potential biomarker for diabetic neuropathy and a target for therapeutic intervention. Previous studies have demonstrated that circRNAs can modulate pain pathways and are involved in neuropathic pain conditions.³⁸ The interaction network analysis and dual-luciferase assays confirmed that circ19601 binds to miR-324-5p, suggesting a regulatory mechanism where circ19601 may sequester miR-324-5p, thereby affecting its availability to target mRNAs. This interaction underscores the importance of circRNA-miRNA networks in regulating pain pathways. Our findings are consistent with recent reports demonstrating that circRNAs can act as miRNA sponges, influencing the expression of genes involved in pain.³⁹

Knockdown of circ19601 led to significant improvements in body weight, blood glucose levels, and pain thresholds in diabetic rats. This suggests that circ19601 contributes to the pathology of DNP and that targeting it may ameliorate diabetes-induced sensory deficits. Moreover, the knockdown of circ19601 resulted in increased levels of miR-324-5p, further supporting its role as a negative regulator of miRNA expression. Our results also showed that knocking down circ19601 decreased the serum levels of neurotransmitters and hormones associated with pain, such as SP, CGRP, NE, and cortisol. This finding aligns with studies indicating that circRNA modulation can influence neurotransmitter levels and pain perception.⁴⁰ In vitro experiments revealed that circ19601 knockdown led to decreased expression of pain-related membrane proteins such as Nav1.7, TRPV1, and COX-2. These proteins are well-documented contributors to pain signaling and neuropathic pain.⁴¹ The restoration of their levels following miR-324-5p inhibition further reinforces the regulatory role of circ19601 and miR-324-5p in neuropathic pain mechanisms.

Notably, the knockdown of circ19601 in this study was achieved by intrathecal injection of an AAV9-U6-sh-circ19601 vector into the L5–L6 intervertebral space, ensuring selective suppression of circ19601 expression in the dorsal root ganglia rather than systemic knockdown. This localized delivery provides stronger evidence that the observed analgesic effect is directly attributable to modulation of DRG neuronal activity. Moreover, in addition to pain relief, circ19601 knockdown mitigated body weight loss and hyperglycemia in diabetic rats. These systemic improvements may be secondary to reduced chronic pain–induced stress responses and lowered circulating levels of stress-associated hormones (e.g. cortisol, norepinephrine), which are known to exacerbate metabolic dysregulation in diabetes.^42–44 Thus, circ19601 may influence not only nociceptive signaling but also neuroendocrine and metabolic pathways, providing a broader therapeutic potential in DNP.

Functional enrichment analysis revealed that the non-homologous end-joining (NHEJ) pathway was significantly enriched in the Model versus Pregabalin comparison. NHEJ is a key DNA double-strand break repair mechanism critical for maintaining genomic stability and preventing neuronal apoptosis.^45,46 In diabetic conditions, hyperglycemia and oxidative stress can induce DNA damage in dorsal root ganglion neurons, and dysregulation of NHEJ may exacerbate neuronal injury, contributing to neuropathic pain.^47,48

This study investigates how dorsal root ganglion circ19601 regulates DNP via miR-324-5p. In a DNP rat model and high-glucose DRG neuronal cells, circ19601 expression increased significantly, suggesting its role in DNP. Reducing circ19601 levels decreased pain sensitivity, indicating its involvement in pain pathways. Bioinformatics and dual-luciferase assays revealed circ19601 inhibits miR-324-5p, affecting pain-related genes like TRPV1 and Nav1.8. This insight offers a potential new therapeutic target for DNP. However, a limitation of the present study is that only male rats were used; since sex hormones, particularly estrogen, may influence pain mechanisms, future studies should include both sexes to comprehensively assess gender-specific effects.

Footnotes

Author contributions

Xin Sun and Hong Xie conceived and designed the study. Jianlin Ge and Zenghui Liu performed the literature search and data extraction. Jianyun Ge, Boxiang Du, Xuefeng Yang, and Qian Su analyzed and interpreted the data. Xin Sun obtained funding and drafted the manuscript. Hong Xie revised the manuscript for important intellectual content. All authors read and approved the final manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Scientific research project of Nantong municipal health commission (No.MB2021021).

Ethics approval and consent to participate

All animal experiments were conducted in accordance with the guidelines and regulations set by The Standard Operating Procedures for Laboratory Animal Center of NTU. The study was approved by the Ethics Committee of the Affiliated Hospital 2 of Nantong University (approval number: IACUC20240505-1005), and all procedures were performed with the highest regard for animal welfare. Every effort was made to minimize animal suffering and to reduce the number of animals used.

Consent for publication

The authors agree to publish in the Journal.

ORCID iDs

Xin Sun

Hong Xie

Availability of data and material

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

Zhao

Chen

Lin

Mei

. Biodegradable microspheres via orally deliver celastrol with ameliorated neuropathic pain in diabetes rats. Regen Biomater 2024; 11: rbae087.

Kajikawa

Yokomi

Narasaki

Kamiya

Miyoshi

Kato

Tsutsumi

. Serotonin-mediated anti-allodynic effect of Yokukansan on diabetes-induced neuropathic pain. J Clin Med 2024; 13(14): 4276.

Richter

Portenoy

Sharma

Lamoreaux

Bockbrader

Knapp

. Relief of painful diabetic peripheral neuropathy with pregabalin: a randomized, placebo-controlled trial. J Pain 2005; 6(4): 253–260.

Saxena

Thanikkal

Chilkoti

Gondode

Sharma

Banerjee

. PPARgamma and AKt gene modulation following pregabalin and duloxetine combination for painful diabetic polyneuropathy. Pain Manag 2024; 14(5–6): 273–281.

Zeng

Pang

Zhou

Cao

. Antidepressant effect of carvedilol on streptozotocin-induced diabetic peripheral neuropathy mice by altering gut microbiota. Biochem Biophys Res Commun 2024; 730: 150374.

Osmanlioglu

HÖ

Naziroğlu

. Resveratrol modulates diabetes-induced neuropathic pain, apoptosis, and oxidative neurotoxicity in mice through TRPV4 channel inhibition. Mol Neurobiol 2024; 61(9): 7269–7286.

Dainese

Wyngaert

De Mits

Wittoek

Van Ginckel

Calders

. Association between knee inflammation and knee pain in patients with knee osteoarthritis: a systematic review. Osteoarthritis Cartilage 2022; 30(4): 516–534.

Fang

Wang

Shen

Zhang

Tang

. Beneficial effects of galanin system on diabetic peripheral neuropathic pain and its complications. Peptides 2020; 134: 170404.

Zakin

Abrams

Simpson

. Diabetic neuropathy. Semin Neurol 2019; 39(5): 560–569.

10.

Baba

Matsui

Kuroha

Wasaki

Ohwada

. Mirogabalin for the treatment of diabetic peripheral neuropathic pain: a randomized, double-blind, placebo-controlled phase III study in Asian patients. J Diabetes Investig 2019; 10(5): 1299–1306.

11.

Papanas

Ziegler

. Emerging drugs for diabetic peripheral neuropathy and neuropathic pain. Expert Opin Emerg Drugs 2016; 21(4): 393–407.

12.

Wettermark

Brandt

Kieler

Bodén

. Pregabalin is increasingly prescribed for neuropathic pain, generalised anxiety disorder and epilepsy but many patients discontinue treatment. Int J Clin Pract 2014; 68(1): 104–110.

13.

Taylor

Angelotti

Fauman

. Pharmacology and mechanism of action of pregabalin: the calcium channel α2-δ (alpha₂-delta) subunit as a target for antiepileptic drug discovery. Epilepsy Res 2007; 73(2): 137–150.

14.

Rosenstock

Tuchman

LaMoreaux

Sharma

. Pregabalin for the treatment of painful diabetic peripheral neuropathy: a double-blind, placebo-controlled trial. Pain 2004; 110(3): 628–638.

15.

Cross

Viswanath

Sherman

. Pregabalin. Treasure Island, FL: StatPearls, 2025.

16.

Kim

Seo

Abdi

Huh

. All about pain pharmacology: what pain physicians should know. Korean J Pain 2020; 33(2): 108–120.

17.

Verma

Singh

Singh Jaggi

. Pregabalin in neuropathic pain: evidences and possible mechanisms. Curr Neuropharmacol 2014; 12(1): 44–56.

18.

Wang

Zhang

Dai

Guan

. Factors predicting the recurrence of atrial fibrillation after catheter ablation: a review. Heliyon 2024; 10(13): e34205.

19.

Jun

Wang

Liao

Qin

Guo

. Circular RNAs as potential biomarkers for male severe sepsis. Open Life Sci 2024; 19(1): 20220900.

20.

de Abreu

FMC

de Oliveira

de Araujo Romero Ferrari

E Silva

KHCV

Titze-de-Almeida

. Exploring circular RNAs as biomarkers for Parkinson’s disease and their expression changes after aerobic exercise rehabilitation. Funct Integr Genomics 2024; 24(4): 130.

21.

Misir

Ozer Yaman

Petrović

Šami

Akidan

Hepokur

Aliyazicioglu

. Identification of a novel hsa_circ_0058058/miR-324-5p axis and prognostic/predictive molecules for acute myeloid leukemia outcome by bioinformatics-based analysis. Biology (Basel) 2024; 13(7): 487.

22.

Lin

Zhou

Zhang

Liu

. MiR-324-5p reduces viability and induces apoptosis in gastric cancer cells through modulating TSPAN8. J Pharm Pharmacol 2018; 70(11): 1513–1520.

23.

Wang

Deng

. CircATIC contributes to multiple myeloma progression via miR-324-5p-dependent regulation of HGF. Biochem Genet 2022; 60(6): 2515–2532.

24.

Ludhiadch

Bhardwaj

Gotra

Kumar

Munshi

. Common microRNAs in epilepsy and migraine: their possibility as candidates for biomarkers and therapeutic targets during comorbid onset of both conditions. CNS Neurol Disord Drug Targets 2023; 22(5): 698–710.

25.

Krames

. The role of the dorsal root ganglion in the development of neuropathic pain. Pain Med 2014; 15(10): 1669–1685.

26.

Kaur

Kotru

Singh

Munshi

. Role of miRNAs in diabetic neuropathy: mechanisms and possible interventions. Mol Neurobiol 2022; 59(3): 1836–1849.

27.

Wang

Yin

Zhao

Chen

Zhi

Wei

Cheng

Xiao

Yang

Kuang

Liu

Zhou

Lin

Yao

. Histone modifications and Sp1 promote GPR160 expression in bone cancer pain within rodent models. EMBO Rep 2024; 25(12): 5429–5455.

28.

Kuang

Luo

Zhu

Zhao

Zhou

Yao

. Upregulation of LncRNA71132 in the spinal cord regulates hypersensitivity in a rat model of bone cancer pain. Pain 2023; 164(1): 180–196.

29.

Ahmed

Al-Nuaimi

Mustafa

Zeidan

Agouni

Djouhri

. K_v7 channel activators flupirtine and ML213 alleviate neuropathic pain behavior in the streptozotocin rat model of diabetic neuropathy. J Pain Res 2024; 17: 2267–2278.

30.

Kang

Wang

Zheng

Shao

Jiang

Fang

. Electroacupuncture alleviates diabetic neuropathic pain and downregulates p-PKC and TRPV1 in dorsal root ganglions and spinal cord dorsal horn. Evid Based Complement Alternat Med 2023; 2023: 3333563.

31.

Huang

Yang

Long

Gao

Lin

. Proteomic analysis of brain tissue from ducks with meningitis caused by Riemerella anatipestifer infection. Poult Sci 2024; 103(10): 104059.

32.

Wang

Luo

Bao

. Intrathecal circHIPK3 shRNA alleviates neuropathic pain in diabetic rats. Biochem Biophys Res Commun 2018; 505(3): 644–650.

33.

Rosenberger

Blechschmidt

Timmerman

Wolff

Treede

. Challenges of neuropathic pain: focus on diabetic neuropathy. J Neural Transm (Vienna) 2020; 127(4): 589–624.

34.

Calcutt

. Diabetic neuropathy and neuropathic pain: a (con)fusion of pathogenic mechanisms? Pain 2020; 161(Suppl 1): S65–S86.

35.

Xie

Luo

Zhang

Chen

Guo

Qiu

Liu

Chen

Zhen

Tian

Chen

Han

Duan

Shen

Yang

. GPR177 in A-fiber sensory neurons drives diabetic neuropathic pain via WNT-mediated TRPV1 activation. Sci Transl Med 2022; 14(639): eabh2557.

36.

Khan

Wang

Shal

Khan

Zahra

Haq

Khan

Rengasamy

. Anti-neuropathic pain activity of Ajugarin-I via activation of Nrf2 signaling and inhibition of TRPV1/TRPM8 nociceptors in STZ-induced diabetic neuropathy. Pharmacol Res 2022; 183: 106392.

37.

Furman

. Streptozotocin-induced diabetic models in mice and rats. Curr Protoc Pharmacol 2015; 70: 5.47.1–5.47.20.

38.

Jiang

Wang

Feng

Wang

. The etiological roles of miRNAs, lncRNAs, and circRNAs in neuropathic pain: a narrative review. J Clin Lab Anal 2022; 36(8): e24592.

39.

Song

Yang

Guo

Zheng

Wang

. Interactions among lncRNAs/circRNAs, miRNAs, and mRNAs in neuropathic pain. Neurotherapeutics 2020; 17(3): 917–931.

40.

Fang

Liao

Tang

Liu

Zhang

Lin

Zhou

Liu

Shen

. Interactions among non-coding RNAs and mRNAs in the trigeminal ganglion associated with neuropathic pain. J Pain Res 2022; 15: 2967–2988.

41.

Strichartz

. Cell signaling and the genesis of neuropathic pain. Sci STKE 2004; 2004(252): reE14.

42.

Park

Kim

Lee

Sharma

Suh

. Depletion of norepinephrine of the central nervous system Down-regulates the blood glucose level in d-glucose-fed and restraint stress models. Neurosci Lett 2016; 620: 121–126.

43.

Panagiotou

Lambadiari

Maratou

Geromeriati

Artemiadis

Dimitriadis

Moutsatsou

. Insufficient glucocorticoid receptor signaling and flattened salivary cortisol profile are associated with metabolic and inflammatory indices in type 2 diabetes. J Endocrinol Invest 2021; 44(1): 37–48.

44.

Sakamoto

Butera

Zhou

Maurizi

Chen

Ling

Shawkat

Patlolla

Thakker

Calle

Morgan

Rahmouni

Schwartz

Tahiri

Buettner

. Overnutrition causes insulin resistance and metabolic disorder through increased sympathetic nervous system activity. Cell Metab 2025; 37(1): 121–137.e6.

45.

Davis

Chen

. DNA double strand break repair via non-homologous end-joining. Transl Cancer Res 2013; 2(3): 130–143.

46.

Pannunzio

Watanabe

Lieber

. Nonhomologous DNA end-joining for repair of DNA double-strand breaks. J Biol Chem 2018; 293(27): 10512–10523.

47.

Pannunzio

Watanabe

Lieber

. Non-homologous end joining often uses microhomology: implications for alternative end joining. DNA Repair (Amst) 2014; 17: 74–80.

48.

Carballar

Martinez-Lainez

Samper

Bru

Ballega

Mirallas

Ricco

Clotet

Jimenez

. CDK-mediated Yku80 phosphorylation regulates the balance between non-homologous end joining (NHEJ) and homologous directed recombination (HDR). J Mol Biol 2020; 432(24): 166715.