Neurophysiological mechanisms of electroacupuncture in regulating pancreatic function and adipose tissue expansion

Abstract

BACKGROUND

Electroacupuncture (EA) has been recognized for its beneficial effects on glucolipid metabolism, potentially through the regulation of sensory nerve coordination. The expandability of peripancreatic adipose tissue (PAT) is implicated in the transition from obesity to type 2 diabetes mellitus (T2DM). However, the specific pancreatic responses to EA require further elucidation.

AIM

To investigate the influence of EA on pancreatic glucolipid reduction level in a high-fat diet (HFD) rat model.

METHODS

To delineate the precise pathway through which EA mediates interactions between PAT and islets, we assessed the expression levels of NGF, TRPV1, insulin, as well as other proteins in the pancreas and PAT. This approach enabled us to identify the acupoints that are most conducive to optimizing glycolipid metabolism.

RESULTS

The ST25, LI11 and ST37 groups attenuated HFD-induced obesity and insulin resistance (IR) to distinct degrees, with ST25 group having the greatest effect. EA at ST25 was found to modify the local regulatory influence of PAT on the pancreatic intrinsic nervous system. Specifically, EA at ST25 obviously activated the TRPV1-CGRP-islet beta cell pathway, contributing to the relief of glucolipid metabolic stress. The beneficial effects were abrogated following the chemical silencing of TRPV1 sensory afferents, confirming their indispensable role in EA-mediated regulation of islet and PAT function. Furthermore, in TRPV1 knockout mice, a reduction in PAT inflammation was observed, along with the recovery of islet beta cell function. EA at LI11 and ST37 demonstrated anti-inflammatory properties and helped ameliorate IR.

CONCLUSION

The PAT ecological niche influenced the progression from obesity to T2DM through various immunometabolic pathways. EA at ST25 could regulate glucolipid metabolism via the TRPV1-CGRP-islet beta cell pathway.

Key Words: Insulin resistance; Electroacupuncture; Sensory nerve; Transient receptor potential vanilloid 1 receptor; Pancreatic beta cell; Peripancreatic adipose tissue

Core Tip: This study underscored the pivotal role of peripancreatic adipose tissue expansion in the transition from obesity to type 2 diabetes mellitus (T2DM), influenced by the crosstalk between islet beta cells and the pancreas islet nerve system. Our findings revealed that electroacupuncture (EA), particularly when applied at ST25, effectively regulates glucolipid metabolism and pancreatic function via the TRPV1-CGRP signaling axis, thereby mitigating metabolic stress and inflammation. This provides a foundation for the development of precision therapeutic strategies for T2DM management. Ongoing exploration of the detailed mechanisms and therapeutic targets of EA at various acupoints, including LI11 and ST37, is essential for refining clinical methodologies aimed at T2DM treatment.

INTRODUCTION

As of 2020, China had the highest prevalence of diabetes globally, with 11.2% of its adult population affected[1]. Accurate diagnosis, prevention, and management of diabetes are crucial for significantly reducing the risk of associated infections. Insulin resistance (IR) is a hallmark of type 2 diabetes mellitus (T2DM). The pathogenesis of IR is linked to the insufficient compensatory capacity of islet beta cells and the relentless supply of nutrients from adipose tissue (AT)[2]. IR has the potential to compromise the functionality of islet beta cells and trigger abnormal AT lipolysis[3]. The pancreas, an integral endocrine organ, encompasses various elements that can influence systemic metabolism. Islet inflammation, prevalent in obese rodents, results in excessive adiposity, IR, and ultimately, the transition to obese T2DM[4]. AT secretion plays a crucial role in insulin signaling and the feedback mechanisms of insulin, being intricately involved in this metabolic interplay. In the early stages of disease, the storage capacity of AT is primarily dependent on its compensatory enlargement and volume expansion[5]. Insulin surges can be temporarily sequestered within cellular vesicles, thereby mitigating the lipolytic effects of insulin. However, excessive accumulation of insulin in AT leads to a diminished capacity for insulin binding, inducing IR. This adaptive response serves to protect AT integrity but subsequently triggers an inflammatory cascade, characterized by the infiltration of immune cells[6]. Peripancreatic AT (PAT) is specifically featured by adipocyte enrichment, as well as distinct features such as multilocularity and the small size of lipid droplets[2]. PAT exhibits robust metabolic activity and is highly responsive to insulin, demonstrating significant vulnerability to the adverse effects of obesity induced by diet or conditions associated with obesity[7]. As a result, the interplay between PAT and islet beta cells is vital for the modulation of glycolipid metabolism. The normal role of PAT as well as its intricate link with islet beta cells remains a subject of complex inquiry.

Electroacupuncture (EA) is known to induce remote physiological changes through somatosensory-autonomic reflexes[8,9]. It effectively regulates blood glucose levels[10,11] and mitigates the adverse effects associated with pharmacological interventions[12]. Clinical studies have shown that EA applied to the bilateral Tianshu acupoint (ST25) enhances insulin sensitivity in patients[13]. Furthermore, EA at Quchi acupoint (LI11) reduces fasting blood glucose levels in obese females[14]. Chronic high-fat diet (HFD) induces obesity combined with IR[15], EA at Shangjuxu acupoint (ST37) has been shown to encourage AT redistribution, ameliorate inflammation, and enhance endocrine function[16]. During pathological progression, dysregulated glycolysis can exacerbate lipid metabolism disruptions, creating a detrimental cycle[17], and precipitating a cascade of detrimental consequences, including oxidative stress[18]. Thus, EA may disrupt glycolipid metabolism via neuroendocrine and neuroimmune axes, presenting a promising therapeutic approach for the restoration of islet beta cell capacity in obese individuals with T2DM. EA applied to distinct acupoints can evoke diverse somatic afferent nerve responses, thereby triggering a variety of downstream cascade reactions[13]. These differential neural afferents explain the varied effects attributed to different acupoints. The segmental innervation of rats, which closely resembles that of humans[19], offers a neurophysiological basis for understanding the differential hypoglycemic impacts of EA at ST25, LI11, as well as ST37.

Owing to its central role in modulating metabolic equilibrium, the pancreas is widely regarded as a pivotal target in the treatment of obesity and T2DM[4]. Typical pancreas islet nerve system (PINS) were identified in rodent pancreatic tissue sections, and their contribution to the regulation of pancreatic function was further elucidated in human samples[20]. PINS comprises a diverse array of neurotransmitters interacting synergistically to establish a complex, endogenous neural framework within the pancreas[21]. PINS specifically senses and adapts to fluctuations in insulin levels. For instance, activation of the pancreatic vagus nerve during hypoglycemia rapidly induces an increase in glucagon secretion[22]. Impaired function of TRPV1-positive nerves has been implicated as a contributing factor to the pathogenesis of T2DM[23]. The development of novel TRPV1-targeted drugs shows promise for regulating glycolipid metabolism and represents a promising avenue for future research[23]. Our recent studies have demonstrated that EA at ST25 mediates hypoglycemic effects, significantly enhanced by the TRPV1-CGRP-islet beta cell pathway[24]. Sensory neurons, including those sensitive to capsaicin or capsazepine, innervate the pancreas and release CGRP to aid in the modulation of glycemic levels. Nonetheless, the precise interplay between each subtype of pancreatic nerve activated by EA and the various components of the pancreatic response requires further elucidation.

In the present study, we employed a HFD rat model to replicate the conditions of obesity and associated IR. Our team has elucidated the pivotal role of EA in modulating brown AT to combat obesity and had unravel part of the underlying neurological mechanisms[25,26]. This subsequent experiment aims to refine the application of EA for the management of obesity. Our objective was to compare the efficacy of various EA, with the aim of refining the therapeutic strategy. Additionally, we delved into the specific mechanisms by which EA modulates the pancreas, encompassing multiple facets of its function.

MATERIALS AND METHODS

HFD rat model

Four-week-old male Sprague Dawley rats were sourced from the Model Animal Research Center at Nanjing University of Chinese Medicine, licensed under permit number 1100112011052760 and authorized by the certificate No. SCXK(JING) 2016-0006. The animals were housed and maintained in the Laboratory Animal Center of the same institution. Rats were randomly assigned to either the control group, which received a standard diet, or the HFD model group, which was fed an HFD as detailed in Supplementary Table 1. The allocation was conducted using a random number table method to ensure unbiased group formation. Following a four-week dietary regimen, the HFD-fed animals were further randomized into three subgroups for intervention: The untreated model group and three EA treatment groups targeting specific acupoints, ST25 and LI11, as well as ST37. Twenty-six per group, of which 10 were used for subsequent electrophysiological experiments. Except for the control group, all groups were maintained on the HFD throughout the intervention period. The experimental procedures were conducted in strict accordance with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals, as promulgated by the National Science Council of China, under the ethical approval number 202105A041.

Acute hyperglycemia model

Male adult TRPV1 knockout (TRPV1^-/-, B6.129X1-TRPV1tm1Jul/NJU, J003770) mice were procured from the Model Animal Center at Nanjing University. C57BL/6 mice served as wild-type controls under the authorization of grant number 202108A002. The study design included 10 animals per experimental group. To advance our understanding of the neural mechanisms underlying the regulation of pancreatic endocrine function, we utilized an acute hyperglycemia model. Mice were subjected to an intraperitoneal injection of a glucose solution (10 g/kg, as referenced in[27], sourced from Sinopharm Chemical Reagent Co., Ltd. Shanghai, China), which induced islet beta cell dysfunction and acute hyperglycemia.

Pancreas intraductal injection

TRPV1 agonists and inhibitors were administered via retrograde pancreas intraductal injection to elucidate the specific effects of pancreatic TRPV1 activation or inhibition on glucose and lipid regulation in response to EA (Supplementary Figure 1). The detailed methodology is provided in the Supplementary material.

Western blot

Rapid isolation and preservation of pancreatic and PAT at -80 °C were conducted to facilitate subsequent analysis. The methods for Western blotting, described in the Supplementary material, were employed to quantify the expression of TRPV1, CGRP, insulin, and other relevant markers. The detailed information of antibodies can be found in Supplementary Table 2.

Immunofluorescence staining

Pancreatic and PAT tissue was fixed in a 4% paraformaldehyde solution for 24 hours, followed by dehydration in a 30% sucrose solution overnight. Frozen sections of the pancreas, 20 μm in thickness, and dorsal root ganglias (DRGs), 40 μm in thickness, were prepared. Primary antibodies specific for TRPV1 (1:200, Abcam, United Kingdom) and insulin (1:200, Signalway, United States) were applied and incubated at 4 °C for 40 hours (The detailed information of antibodies can be found in Supplementary Table 2). Excess primary antibodies were removed using 0.1 M phosphate-buffered saline (Biosharp Life Sciences, China), after which the sections were incubated with secondary antibodies conjugated with Alexa Fluor 488 (for mouse) and 594 (for rabbit) at a dilution of 1:500, both sourced from Abcam, United Kingdom, for 1 hour at 37 °C. Fluorescence imaging was performed using an Olympus BX60 microscope (Japan). In addition to calculating the Pearson correlation coefficients, the co-localization analysis was also described using Plot Profile to calculate the change curves of different fluorescence along this rectangular distance within the range of interest (ROI), with a view to clarifying the relationship between the two metrics within the ROI.

EA intervention

Rats in the EA group were subjected to stimulation at specific bilateral acupoints: ST25, located 5 mm lateral to the intersection of the upper two-thirds and lower one-third of the line connecting the xiphoid process to the upper border of the pubic symphysis; LI11, found in the depression medial to the extensor carpi radialis muscle, at the lateral end of the cubital crease; and ST37, situated 5 mm below the knee joint and 1 mm lateral to the anterior crest of the tibia. The detailed methodology for these procedures is delineated in the Supplementary material.

Recording of electrophysiological activity in PINS

To facilitate the isolation of the pancreatic intrinsic nervous system, rats were induced into anesthesia using isoflurane (2%-5%) via a precision vaporizer (RWD Life Science Co., Ltd., Shenzhen, China). Anesthesia depth was confirmed by the absence of corneal and hind paw withdrawal reflexes. A laparotomy incision, approximately 3 cm in length, was made lateral to the abdominal midline. The longitudinal muscle and myenteric plexus were carefully dissected from the duodenum, ensuring the integrity of the connective tissue linking the bowel and pancreas. The pancreatic nerves, which course along the splenic artery and the superior and inferior pancreatic arteries, were identified. A branch of the pancreatic intrinsic nervous nerve was isolated and connected to a positive electrode (PFA-Coated Platinum, A-M system, United States, 772000). A reference electrode was placed on adjacent tissue. To minimize electromagnetic interference, the experimental rats were housed in Faraday cages[24]. Only the electrodes in contact with the nerves were exposed. Neural spikes were captured using a preamplifier (A-M systems, Carlsborg, WA, United States; band-passing: 10-1000 Hz, sampling frequency: 20000 Hz, amplification: 1000 ×) and interfaced with a biosignal acquisition and analysis system (Micro1401-3, CED, United Kingdom). Data analysis was performed using Spike2 software.

Statistical analysis

Data from all experiments are presented as mean ± SE. Statistical comparisons within the same group before and after EA intervention were performed using a paired t-test, while comparisons between two distinct groups were conducted with an independent t-test. For multiple group comparisons, one-way analysis of variance (ANOVA) was employed. Data analyses were executed using SPSS software version 22.0 (IBM Corp., United States) and GraphPad Prism version 8.0 (GraphPad Inc., United States). Statistical significance was determined at the threshold of P < 0.05.

RESULTS

EA improved systemic glucose-lipid metabolism and pancreatic adipose ectopic deposition

Rats administered a HFD for 4 weeks showed a tendency toward obesity. After EA intervention, a significant reduction in body weight was observed in all groups, with varying degrees of weight loss, as shown in Supplementary Figure 2A. Concurrently, elevated levels of triglyceride as well as total cholesterol in the model group, indicative of dysregulated lipid metabolism, were ameliorated post-EA treatment, with the most pronounced improvement observed in the ST25 group, as depicted in Supplementary Figure 2B-D. Additionally, the rats showed disturbances in glucose metabolism, characterized by elevated serum glucose levels (Supplementary Figure 2E) and increased insulin secretion (Supple- mentary Figure 2F).

Abnormal lipolysis in adipocytes is recognized as a significant contributor to IR and impaired glucose tolerance[28]. In the model group, adipocytes displayed an increase in diameter and variability in size. The treatment group demonstrated varying degrees of improvement (Supplementary Figure 2G). Furthermore, skeletal muscle, a pivotal site for energy metabolism, is crucial for glucose and fatty acid metabolism. Impairment of insulin receptors, characterized by reduced ability of insulin to effectively bind to its receptors and trigger downstream signaling, results in accumulation of insulin within the gastrocnemius muscle. The model group in our study showed IR, which corroborated the findings obtained from serum analysis, as shown in Supplementary Figure 2H-J. The ST25, ST37, and LI11 groups exhibited varying degrees of improvement in IR, highlighting the capacity of EA to enhance systemic metabolic homeostasis.

Islet morphology was assessed through histological examination, as shown in Figure 1A. The model group exhibited signs of islet beta cell damage, including moderate degeneration and cellular swelling. Additionally, islet atrophy was observed, characterized by irregular contours and the presence of prominent vacuoles. The treatment groups showed varying degrees of islet restoration. Furthermore, ORO staining revealed ectopic AT deposition within the pancreas of the model mice, as shown in Figure 1B. This ectopic AT was improved to varying extents in the EA treatment groups, with the ST25 group demonstrating the most significant reduction in lipid deposition.

Figure 1 Electroacupuncture modulates the trophic influence of peripancreatic adipose tissue on the pancreatic intrinsic nervous system. A: Histological alterations in pancreatic tissue and ectopic fat deposition in the pancreas across experimental groups (200 × magnification). Blue arrows indicate vacuolar-like changes; B: Comparative analysis of peripancreatic adipose tissue (PAT) blood perfusion between the control and model groups; C: Immunohistochemical staining for NGF in PAT under high magnification for the different groups (400 × magnification). Scale bar represents 50 μm; D: Illustrative immunofluorescence images showing co-expression of NGF and tyrosine hydroxylase (TH) in all groups (200 × magnification). Red fluorescence denotes NGF, while green fluorescence indicates TH. A uniform scale bar is applied to all groups; D: Pearson correlation analysis between NGF and TH expression (n = 5); E: Illustrative immunofluorescence images showing co-expression of NGF and TH in all groups (200 × magnification); F and G: Representative Western blot bands for NGF; F and H: Vascular endothelial growth factor (VEGF) in PAT. Vinculin served as a loading control; I-L: PAT blood flow changes during electroacupuncture stimulation at acupoints ST25, LI11, and ST37 in normal rats, respectively. ^aP < 0.05 compared to the normal control group; ^bP < 0.05 compared to the model or alternative treatment group. TH: Tyrosine hydroxylase; ST25: Tianshu acupoint; LI11: Quchi acupoint; ST37: Shangjuxu acupoint.

EA enhanced AT remodeling to impact the PINS homeostasis

The preservation and restoration of islet morphology are closely associated with the integrity of the nervous system[29]. As a subtype of white AT (WAT), PAT’s metabolic activity is regulated by the plasticity of its nerve endings[30]. In the control group, a minimal presence of AT was observed, which implies a supportive nutritional role of PAT for the pancreas, as illustrated in Figure 1B. The effect of NGF signaling blockade on the activation state of adjacent pancreatic nerves remains to be fully understood. Our findings showed a significant downregulation of NGF expression in the model group, as depicted in Figure 1C, with subsequent restoration observed in the intervention groups to varying extents. Given the correlation between islet morphology maintenance and nervous system abundance[29], the recovery of NGF expression following EA at ST25 aligned with the observed trend in islet morphology repair. Additionally, the model group showed elevated co-expression of NGF and tyrosine hydroxylase (TH), suggesting NGF's involvement in neuronal remodeling, which leads to increased TH activity in the pancreas under HFD conditions. Notably, EA at the ST25 acupoint demonstrated the most pronounced effect on the neuroinhibition of pancreatic TH, as shown in Figure 1D-G.

The microvascular niche is essential for maintaining the functional homeostasis of AT[31]. We investigated the modulatory effects of EA on the expansion and functional shifts of PAT in response to a HFD environment. In the model group, PAT volume increased significantly (Figure 1B), indirectly expanding the cellular space surrounding microvessels. The reduction in VEGF as well as platelet endothelial CD31 within PAT mirrored that observed from the pancreas, suggesting compromised secretion of insulin together with a reduction in the mass of beta cells, as depicted in Figure 1F and H and Supplementary Figure 3A and B. EA treatment was found to effectively enhance blood perfusion in PAT, both in normal and model groups, with pronounced and rapid effects observed in Figure 1I-L. Additionally, abnormal VEGF levels have been shown to correlate with inflammatory responses[32]. Collectively, these findings suggest that EA may restore the balance of NGF secretion and transport by improving the vascular ecosystem within PAT, thereby facilitating coordinated adaptive changes in the PINS.

EA alleviated the inflammatory infiltration associated with adiposity and improved metabolic health

Adipose cell expansion facilitates the recruitment of macrophages and their polarization towards a pro-inflammatory phenotype upon chemokine stimulation[33]. Consequently, the pathogenesis of T2DM is influenced by the signal transduction between macrophages and adipocytes, with a particular focus on PAT. In the model group, F4/80, a macrophage marker, showed elevated expression with statistical significance (P < 0.05, Figure 2A and B), indicating a substantial macrophage infiltration. Additionally, there was a marked increase in the production of MCP-1 (P < 0.05, Figure 2A and C), and an imbalance in the ratio of interleukin (IL)-10 to IL-1β was observed (P < 0.05, Figure 2A, D and E), leading to inflammatory infiltrate. The significant decrease in scavenger receptor cysteine-rich type 1 protein M130 (CD163) and the abnormal increase in IL-12 also indicated that the pancreatic tissue was being invaded by an inflammatory response (Supplementary Figure 3A, C and D). Stimulation of acupoints ST25 or LI11 through EA effectively reversed the levels of these key inflammatory mediators within AT, as depicted in Figure 2A-E. In contrast, EA at ST37 exhibited a less pronounced anti-inflammatory effect, with no significant modulation observed on the expression of F4/80 (P > 0.05, Figure 2A and B). It is suggested that EA improved the inflammatory environment associated with adiposity.

Figure 2 Electroacupuncture modulates inflammatory mediators in peripancreatic adipose tissue, enhancing metabolic function. A and B: Electroacupuncture (EA)'s impact on the expression of F4/80 (also known as EGF-like module-containing mucin-like hormone receptor-like 1); A and C: Influence of EA on MCP-1 levels; A and D: Changes in interleukin (IL)-1β expression due to EA treatment; A and E: EA's effect on IL-10 expression; F and G: Expression levels of PPAR-γ following EA intervention; F and H: Adipose tissue expression of FABP4 in response to EA; F and I: TRPV1 expression in PAT after EA treatment. Vinculin served as a loading control; J: Representative immunofluorescence images of co-expression of FABP4 and TRPV1 in the pancreas of the model group (200 × magnification). Blue fluorescence representing FABP4, and TRPV1 is represented by purple fluorescence; K: Quantitative analysis of the co-localization of FABP4 and TRPV1 immunofluorescence, indicating a correlation between the two proteins. ^aP < 0.05 compared to the normal control group; ^bP < 0.05 compared to the model or another treatment group.

Stimulation of ST25 through EA had TRPV1-dependent dependent efficacy in managing glucolipid metabolism

Adipocyte FABP4 is uniquely localized to fully differentiated adipocytes. Additionally, PPAR-γ serves as a key regulator of adipogenesis[34]. These biomarkers are crucial for tracking the developmental trajectory of adipocytes. In the model group, elevated FABP4 levels combined with reduced PPAR-γ levels confirmed disturbances in adipocyte differentiation and metabolic equilibrium, as shown in Figure 2F-H. The upregulation of FABP4 heightened the concentration of lipids while attenuated the sensitivity of insulin; In contrast, an inverse correlation was observed between PPAR-γ levels and FABP4 expression. EA was capable of normalizing the expression of these markers, suggesting their potential as therapeutic targets for T2DM[35].

Further investigation using whole tissue imaging of the pancreas revealed extensive co-localization of FABP4 and TRPV1 in the model group, as shown in Figure 2F and I-K. This observation implies that TRPV1-expressing neural axons within the pancreas may serve as key mediators, innervating and sensing alterations in the PAT state, as indicated by FABP4 labeling. Consistent with this hypothesis, modulation of TRPV1 levels in both the model and treatment groups paralleled the changes in FABP4 and PPAR-γ (Figure 2F and I). Collectively, these findings suggest that EA, particularly at the ST25 acupoint, exerts a pronounced regulatory influence on lipid metabolism, potentially through the modulation of TRPV1-expressing nerve axons.

To investigate the role of TRPV1 in mediating the improvement of pancreatic insulin release by the PINS, we selected acute hyperglycemic mouse models treated with TRPV1 agonists and antagonists, respectively. These models allowed us to separately analyze the involvement of TRPV1. Post-surgical blood glucose responses were then monitored in these mice. EA at ST25 was found to rely on pancreatic TRPV1 to exert its effects, (Supplementary Figure 4A-D). TRPV1 knockout (TRPV1^-/-) mice exhibited similar phenomena, as shown in Supplementary Figure 4E and F, while the sham EA group did not demonstrate a comparable effect (Supplementary Figure 4G). These findings indicated that TRPV1 ameliorates glycemic metabolism and underpins EA’s efficacy at acupoint ST25.

We concurrently compared the fat content across multiple depots in TRPV1^-/- mice. The findings revealed an obvious reduction in the wet weight of fat in TRPV1^-/- mice, particularly in visceral fat depots such as PAT (Supplementary Figure 4H). This suggests that TRPV1 deletion plays a substantial role in maintaining lipid metabolism and promoting fat redistribution. The regulatory influence of TRPV1 on PAT was replicable post-ablation of pancreatic TRPV1-expressing sensory afferent neurochemicals; however, at this juncture, EA at ST25 was ineffective (Supplementary Figure 4H). This implies the existence of bidirectional neuroendocrine signaling of PAT with islet beta cells, with TRPV1-expressing sensory nerves serving as conduits that modulate EA’s effects at the acupoint ST25. The minimal impact of manual acupuncture (MA) at acupoint ST25 on PAT activity provides corroborating evidence for this hypothesis (Figure 3). Since only the pancreatic TRPV1 afferent nerves were ablated, the weight-reducing effect of EA at ST25 on epididymal WAT was preserved (Supplementary Figure 4H). The unaltered expansion of subcutaneous AT in TRPV1^-/- mice helped to maintain inflammatory response levels, partially elucidating the lack of significant change in subcutaneous AT in these mice. In summary, these findings indicate that TRPV1 is a key regulator of lipid metabolism homeostasis and may help prevent the development of IR through its modulatory effects on signaling pathways within islet β cells and PAT.

Figure 3 TRPV1-mediated acupuncture modulates the discharge patterns of the pancreatic intrinsic nervous system. A and B: The impact of manual acupuncture (MA) at Tianshu (ST25) acupoint on pancreatic intrinsic nervous system (PINS) activity and the corresponding discharge frequency in wild-type (WT) mice; C and D: The influence of MA at ST25 on PINS activity and discharge frequency in TRPV1 knockout (TRPV1^-/-) mice; E and F: The effect of MA at ST25 on peripancreatic adipose tissue (PAT) activity and associated discharge frequency in WT mice; G and H: The alteration in PAT activity and discharge frequency following MA at Tianshu (ST25) acupoint in TRPV1^-/- mice (n = 7). ^aP < 0.01 compared to the dur-MA group; ^bP < 0.01 compared to the pre-MA group and MA, manual acupuncture. TRPV1: Transient receptor potential vanilloid subfamily member 1; MA: Manual acupuncture.

To elucidate the essential role of TRPV1 in sustaining the therapeutic effects of acupuncture, we assessed the effects of MA on TRPV1^-/- mice. The results indicated that the response of TRPV1^-/- mice to MA was obviously attenuated in contrast with that of wild-type mice. MA increased the frequency of PINS discharge 3.47 ± 0.16 Hz, in contrast with the pre-acupuncture baseline of 0.39 ± 0.02 Hz (P < 0.05, Figure 3A and B). In contrast, following TRPV1 ablation, the electrophysiological activity of PINS post-MA intervention remained largely unchanged (0.31 ± 0.02 Hz pre-MA in contrast with 0.43 ± 0.07 Hz post-MA, P > 0.05, Figure 3C and D), indicating a significant reduction in the sensitivity of PINS to MA.

Stimulation of ST25 through EA enhanced endocrine function by engaging TRPV1 sensory afferents and the TRPV1-CGRP-Beta cell pathway

Rats with dysregulated glucolipid metabolism exhibited a range of neuroadaptive alterations within the pancreas, notably an increase in TRPV1-positive nerve fibers, as shown in Figure 4A and Supplementary Figure 5. Corresponding changes were observed in the DRG, where the expression of TRPV1, a molecule involved in glucose metabolism, was significantly downregulated by the HFD. This reduction was partially reversed by EA, as illustrated in Figure 4B-D. These findings suggest a link between TRPV1 and the regulation of glucose metabolism by EA in both central and peripheral compartments. We further explored the upward regulatory mechanism of TRPV1 sensory nerve regulation of pancreatic function by injecting capsazepine into the T10 DRG. The low concentration of capsaicin used in this experiment inhibits calcium influx and reduces the production of electrical signals by cells, which travel along afferent nerves to the brain. The results showed that TRPV1 inhibition by capsaicin significantly reduced CGRP and insulin positive nerves in the DRG of rats in the EA group (Figure 5A). This suggests that the altered response of CGRP and INS neurons in the DRG may be one of the cascade effects following TRPV1 activity changes. Meanwhile, capsazepine injection significantly suppressed the response of T10 DRG and TRPV1-mediator CGRP insulin by EA at ST25 (Figure 5B-H). These findings suggest that the upstream neurological mechanisms by which EA regulates pancreatic function are partially mediated by TRPV1.

Figure 4 TRPV1-mediated enhancement of metabolic function by electroacupuncture. A: Representative immunofluorescence images of pancreatic tissue across experimental groups. The nuclei are visualized in blue by 4',6-diamidino-2-phenylindole (DAPI), with green immunofluorescence indicating the presence of TRPV1. Islets were examined at 630 × magnification. A uniform scale bar of 25 μm applies to all groups; B: Representative immunofluorescenc images of the dorsal root ganglion across groups. Nuclei are stained blue by DAPI, with green immunofluorescence representing TRPV1 and red immunofluorescence representing insulin; C: Percentage of pancreatic TRPV1+ nerve fibers in each group; D: Illustrative immunofluorescence images showing co-expression of TRPV1 and INS in all groups. ^aP < 0.05 compared to the normal control group; ^bP < 0.05 compared to the model group. TRPV1: Transient receptor potential vanilloid subfamily member 1; INS: Insulin; DAPI: 4',6-diamidino-2-phenylindole.

Figure 5 Modulation of pancreatic function by TRPV1 neuronal pathways in electroacupuncture-treated t10 dorsal root ganglion. A: Representative immunofluorescence images illustrating co-expression of insulin (INS) and CGRP in the pancreas at 200 × magnification. Nuclei are stained blue with 4',6-diamidino-2-phenylindole, with red representing INS and green representing CGRP immunofluorescence; B-D: Electrophysiological activity measurements of the T10 DRG in model mice under different conditions: Untreated model, TRPV1 knockout (TRPV1^-/-) in T10 DRG, and TRPV1^-/- in T10 DRG followed by electroacupuncture (EA) at the ST25 acupoint; E: Discharge frequency associated with the groups mentioned in B-D; F-H: Impact of EA on the expression levels of TRPV1 and INS in the pancreas following capsazepine injection into the T10 DRG. GAPDH served as an internal reference protein. ^aP < 0.05 compared to the model or another treatment group. INS: Insulin; DAPI: 4',6-diamidino-2-phenylindole; ST25: Tianshu acupoint.

To elucidate the specific mechanisms underlying TRPV1-mediated crosstalk of pancreatic beta cells with PAT, we assessed the expression levels of tissue proteins among TRPV1^-/- mice. Results showed that the targeted knockdown of TRPV1 induced an obvious reduction of hyperinsulinemia within the pancreas and concurrently mitigated the inflammatory response in PAT, as demonstrated in Figure 6A-F. Having established that sensory afferent TRPV1 serves as a crucial neurophysiological foundation for maintaining islet function stability, we further investigated the molecular cascade triggered by TRPV1 activation. Our focus was on the expression of CGRP, a molecule known to be modulated by TRPV1 and that exerts a direct inhibitory effect on insulin secretion[23]. Compared to the normal control group, the model group exhibited elevated TRPV1 levels alongside reduced CGRP levels. EA at LI11 or ST37 did not significantly reduce the abnormal expression of TRPV1. In contrast, the ST25 group demonstrated a more pronounced increase in CGRP levels compared to the LI11 and ST37 groups (Figure 6G-I). These results support our findings that EA at ST25 is the most effective, promoting insulin secretion regulation through upregulation of CGRP expression (Figure 6J-K). The beneficial effects of EA on insulin secretion were partially counteracted (Figure 6F-H).

Figure 6 TRPV1 knockout preserves immunometabolic homeostasis in peripancreatic adipose tissue and pancreas. A-F: Impact of electroacupuncture (EA) on the levels of insulin receptor, EGF-like module-containing mucin-like hormone receptor-like 1 (F4/80), interleukin (IL)-10, IL-1β, and insulin in peripancreatic adipose tissue following TRPV1 knockout; G-I: Influence of EA on the expression levels of TRPV1 and calcitonin gene-related peptide-receptor component protein (CGRP) in the pancreas. Vinculin served as loading controls; J: Representative immunofluorescence images of co-expression of TRPV1 and CGRP in the islets of the ST25 group (200 × magnification). Nuclei were stained with 4',6-diamidino-2-phenylindole (blue), TRPV1 is represented by red immunofluorescence, and CGRP by green immunofluorescence; K: Correlation between TRPV1 and CGRP expression as revealed by quantitative analysis of the co-localization of TRPV1 and CGRP immunofluorescence. ^aP < 0.05 compared to the normal control group; ^bP < 0.05 compared to the model or another treatment group. IL: Interleukin; EA: Electroacupuncture

DISCUSSION

EA at ST25 Reversed AT Expansion and Arrested the Progression to T2DM

The HFD model more closely mirrors the weight gain process observed in most human populations than the monogenic model[36]. The quantity and quality of dietary fat significantly influence the severity of diabetes[37]. Long-term consumption of a high-energy diet, such as 45-60 kcal, leads to weight gain, a gradual increase in blood glucose levels, and various behavioral abnormalities in rodents, such as hyperalgesia[38]. This suggests that, in addition to obesity traits, diabetes and its complications can be observed in these animals. While diet-induced obesity models can mimic certain aspects of human obesity and diabetes pathogenesis, there are notable differences. For instance, humans are highly responsive to changes in dietary structure, particularly carbohydrate intake, which greatly affects blood glucose levels[39]. This sensitivity may not be replicated in food-borne obesity rat models. Furthermore, animals typically consume food ad libitum throughout the day and night, often with a more uniform diet[40], which does not accurately reflect the diverse eating habits and patterns of humans. Despite these limitations, HFD-induced obesity models are more widely adopted than single-gene genetic models or surgically induced obesity models[41]. The conditions for these models can be refined over time to better approximate the human condition. Dysfunctional AT is intricately linked to the pathogenesis of T2DM, encompassing adiposity-induced disruptions in islet beta cell function, alterations in AT biology, and multi-organ IR, which are often mitigated and may even be reversed with significant weight loss[42]. Despite this, the underlying mechanisms of IR remain incompletely understood. Our study provides preliminary evidence that PAT regulated islet β-cell function in an NGF-dependent manner. NGF played a crucial role in coordinating the functional homeostasis of the pancreatic islet neurovascular system. Moreover, islet beta cells constitute a pivotal target regulated by PINS. The inflammatory status within the islets is intricately linked to the severity of underlying metabolic disturbances. The accumulation of aberrant lipid deposits within the pancreas may catalyze an excessive secretion of insulin, which in turn intensifies inflammation and sets off a harmful feedback loop. The inflammatory response induces a significant remodeling of the pancreatic microenvironment, with PINS playing a key role. This transformation serves as a compensatory adaptation to counteract the effects of metabolic stress. Our findings lead us to hypothesize that the expandability of PAT underlies the development of obesity and associated IR. Abnormal lipolytic processes are one of the key factors that exacerbate the IR phenotype[43]. Lipid metabolites are generated and can modulate glucose-stimulated insulin secretion through feedback mechanisms involving islet beta cell lipolysis[44]. Therefore, modulating the activity of particular lipolytic enzymes could potentially optimize insulin sensitivity. Even when islet morphology is compromised, the body can still keep the concentration of glucose in the blood within a normal range, suggesting the presence of redundancy and compensatory mechanisms within islet function. In summary, early IR events primarily occur in peripheral tissues such as adipose and muscle. Ectopic lipid deposition and accumulation are crucial in the overall pathological process. Thus, strategically modulating the functional interplay between PAT and other tissues represents a promising therapeutic strategy for managing the disease progression of T2DM. EA might offer an innovative therapeutic approach for metabolic control.

The proliferation of visceral AT is intimately connected with the escalation of metabolic stress and inflammation[45]. Autonomic regulatory mechanisms within adipocytes play a crucial role in maintaining their physiological equilibrium. Adipocyte growth is further influenced by a range of signaling pathways. Infiltrating macrophages and other immune cells can modify and reshape the adipocyte microenvironment[46]. Systemic hormones, including insulin, exert their influence on adipocyte expansion and proliferation via paracrine signaling as well as other cellular communication pathways, thereby impacting overall energy equilibrium[47]. The initial acute response can also summon additional local macrophages to restructure the extracellular matrix and cytoskeleton of adipocytes, countering the stress of adipocyte hyperplasia[48]. This phenomenon partly explains the link between inflammatory mediator buildup in PAT and abnormal lipid storage. AT is richly vascularized, featuring a dense network of capillaries that facilitates optimal delivery of essential nutrients and oxygen to support its metabolic functions. The microvascular system is vital for AT growth, can combat hypoxia, and supports the process by which precursor cells from the tissue's tiny blood vessels transform into mature fat cells[49]. Hypoxia together with the ensuing inflammatory cascade is implicated in the progression of obesity and IR[50]. Consequently, AT expansion is also governed by vascular blood flow, a relationship strongly supported by our findings that EA can enhance blood circulation within fat deposits. Alterations in MCP-1 levels were positively correlated with the extent of tissue inflammatory infiltration[51] and insulin damage. The accumulation of inflammatory cytokines can contribute to IR in affected individuals[6], thereby increasing the degree of lipolysis. As a result, the self-preservation mechanisms of PAT help maintain the balance of insulin production and uptake, preserving pancreatic endocrine homeostasis. The efficacy of EA in regulating glucolipid metabolism is linked to its ability to enhance PAT blood supply and reduce insulin receptor damage. Furthermore, the pancreatic vascular system can mediate feedback mechanisms via adipokines, facilitating homeostatic regulation[52]. Chronic glucose or lipid overload can escalate islet microvascular pressure, precipitating endothelial cell damage and aberrant islet blood flow regulation, culminating in pancreatic islet blood flow coupled with the progressive functional deterioration of islet cells (Supplementary Figure 6)[53]. EA improves the structural reorganization of islet beta cells while simultaneously inhibiting the formation and development of new fat cells. Despite of that, its effectiveness are closely tied to the precise selection of acupoint points, with ST25 showing significant benefits. interestingly, EA at either ST25 or LI11 can reduce macrophage infiltration into AT, lessening their effect on adipocyte remodeling and size reduction. The therapeutic effects of EA depend on the integrity of neural structures and functions, which exhibit ganglionic specificity. The neurophysiological basis for EA at ST25 in alleviating pancreatic steatosis is linked to the partial anatomical overlap between the acupuncture site and the ganglionic segments innervating the target organ[54]. ST37 is primarily innervated by nerve projections from L5 to S2[19], while LI11 is associated with the C5-T11 spinal segments[55]. This neuroanatomical distinction suggests that pancreatic tissue is not a direct target of EA at ST37 or LI11. Instead, the beneficial effects of EA at these acupoints on pancreatic inflammation may be secondary to improvements in the condition of other tissues. The vagus nerve mediates the systemic anti-inflammatory effects of EA at ST37, overcoming the limitations of ganglionic segmentation.

The interaction between PAT and pancreas underpins EA at ST25 in regulating glucolipid metabolism via TRPV1

Adipokines play a pivotal role in intercellular communication, energy storage sensing, and the regulation of adipocyte size and distribution[56]. A reduction in PPAR-γ transcriptional activity can impair adiponectin secretion, thereby exacerbating IR[57]. Consistent with findings in HFD-fed mice, pancreatic tissue from obese patients exhibited typical steatosis, with excessive PAT accumulation primarily localized around blood vessels and nerves[21]. Adipocytes have the capacity to adaptively reshape the pancreatic neurovascular network. When adipose cells release NGF, it can trigger remodeling of the pancreatic neurovascular network, potentially disrupting the pancreatic environment and worsening pathological conditions. Our research shows that in the model group, islet structural damage often coincided with reduced NGF secretion from PAT. EA was effective in restoring the expression of NGF within PAT (Figure 1). NGF, emanating from PAT[58], can alter nerve fiber regulation in the pancreas and enhance the resilience of the PINS to metabolic stress. This observation aligns with the observed decline in sensory afferent markers, such as TRPV1, in the model group. Besides, sensory nerves play a crucial role in conveying metabolic information about AT reserves to the brain while simultaneously releasing regulatory neuropeptides that modulate local adipocyte function[54]. Sensory nerve fibers within and surrounding AT can rapidly detect and transmit changes in metabolic signals. Therefore, exploring the functional interplay between PAT and pancreatic islet β-cells is crucial, especially for understanding the role of TRPV1⁺ sensory nerve fibers. In the present study, we demonstrate that EA at ST25 can remodel pancreatic tissue nerve innervation (Supplementary Figure 5), which may underlie the glucose-lowering effects of EA. Furthermore, we reveal that TRPV1⁺ sensory afferent fibers can directly inhibit abnormal insulin release by modulating CGRP secretion. In the absence of TRPV1, this regulatory mechanism is impaired, rendering EA ineffective and highlighting the necessity of TRPV1 in EA-mediated pancreatic endocrine regulation. Furthermore, TRPV1, typically implicated in inflammatory response studies, may also participate in AT metabolism and homeostasis[55]. The lack of trpv1 reduces pat inflammation, which might improve islet function. Adipocytes and islet beta cells interact under the influence of inflammatory pathways as well as neural regulatory mechanisms. However, the mechanisms underlying the functional communication between pancreatic β-cells and PAT remain incompletely understood, particularly in relation to the pathological remodeling of the PINS under metabolic stress. In our previous work, we characterized the PINS based on the pathology of T2DM in rats and elucidated the potential mechanisms by which the PINS adapts to abnormal glucose metabolism[24]. This study builds upon those findings, further substantiating that neural reconfiguration can stem from impaired fat processing pathways, encompassing a spectrum of inflammatory events triggered by AT. Nonetheless, the precise sequence initiating the transmission of the CGRP signal to islet beta cells remains elusive, warranting additional exploration into the function of CGRP-associated receptors within the pancreas. Concurrently, identifying the principal target organs influenced by the acupoints LI11 and ST37 and elucidating their mechanisms in the management of T2DM is essential. Such research will contribute valuable insights for the development of clinical intervention strategies.

CONCLUSION

In summary, this study underscored the pivotal role of PAT expansion in the transition from obesity to T2DM, influenced by the crosstalk between islet beta cells and the PINS. Our findings revealed that EA, particularly when applied at ST25, effectively regulates glucolipid metabolism and pancreatic function through the TRPV1-CGRP signaling axis, thereby mitigating metabolic stress and inflammation. Moreover, the study emphasized the significance of understanding the neurophysiological mechanisms by which EA modulates glucolipid metabolism, offering a foundation for the development of precision therapeutic strategies for T2DM management. Ongoing investigation into the detailed mechanisms and therapeutic targets of EA at various acupoints, including LI11 and ST37, is essential for refining clinical methodologies aimed at optimizing T2DM treatment.

ACKNOWLEDGEMENTS

We sincerely thank the staff of the Key Laboratory of Acupuncture and Medicine Research, Ministry of Education, and the Experiment Center for Science and Technology at Nanjing University of Chinese Medicine for their valuable support during the preparation of this manuscript.