Redox-Responsive Surface-Enhanced Raman Scattering Microneedles for Monitoring Acupuncture Responses in Rats With Impaired Glucose Tolerance

Summary

This exploratory rat model study presents a redox-sensitive surface-enhanced Raman scattering (SERS)-active microneedle for simultaneous acupuncture intervention and ratiometric tracking of SERS indices related to local oxidative capacity and redox potential in rats with impaired glucose tolerance.

Abstract

Impaired glucose tolerance (IGT) is associated with disturbed redox homeostasis and increased oxidative stress, but dynamic in vivo tracking of redox-associated changes during acupuncture intervention remains technically challenging. This study developed a redox-sensitive surface-enhanced Raman scattering (SERS)-active microneedle system for simultaneous acupuncture intervention and relative ratiometric tracking of oxidative capacity (OC)- and redox potential (RP)-related SERS indices in rats with IGT. Oxidative- and redox-sensitive SERS probes were prepared by functionalizing gold nanoshells (GNS) with p-phenylenediamine (p-PDA) for OC responsiveness and anthraquinone for RP responsiveness, respectively. The probes were loaded into two etched grooves on acupuncture needles to construct SERS-active microneedles. Specific-pathogen-free male Sprague–Dawley rats were used in this exploratory study. Three rats served as normal controls, and 20 rats were fed a high-fat diet and administered streptozotocin (STZ) to induce IGT. After screening, 18 eligible IGT rats were divided into high-fat-diet and normal-diet groups, with 9 rats per diet group. Each diet group was further divided into acupuncture-point twisting, non-acupuncture-site twisting, and acupuncture-point direct-insertion subgroups, with 3 rats per subgroup. Needling was performed twice daily for 7 consecutive days, and fasting blood glucose (FPG), 2 h postprandial blood glucose (2hPG), and OC- and RP-related SERS indices were recorded. The results showed that needling at acupuncture points with twisting was associated with lower FPG and 2hPG values and altered OC- and RP-related SERS indices in this IGT rat model. Because external calibration curves, independent biochemical oxidative-stress assays, probe-leaching tests, and biocompatibility assays were not included, the SERS ratios should be interpreted as relative redox-associated optical indices rather than absolute biochemical concentrations or absolute redox-potential values. These findings support the feasibility of SERS-active microneedles as an exploratory preclinical platform for combining acupuncture intervention with local ratiometric SERS tracking in IGT rats.

Introduction

Impaired glucose tolerance (IGT), a decrease in the ability of the body to tolerate glucose, represents a transitional stage between normal blood glucose and diabetic blood glucose¹. IGT is thus a key early-warning signal and a reversible stage in the development of type 2 diabetes². Clinical data show that approximately 30% of the population with IGT will progress to type 2 diabetes within 5–10 years³; however, with timely intervention, this percentage can be reduced to less than 10%⁴.

IGT is a dynamic and complex process, and a persistent high-glucose environment is a key trigger for oxidative metabolism disorders⁵. When the body is in a prolonged state of hyperglycemia, the mitochondrial respiratory chain functions abnormally, and the rate of reactive oxygen species generation increases by 30–50%, directly enhancing oxidative power in the body⁶; simultaneously, high sugar levels inhibit the activity of key enzymes, such as glutathione reductase, preventing the antioxidant system from removing excess reactive oxygen species in a timely manner and further disrupting the oxidation–reduction balance. In addition, IGT is accompanied by pathological changes, including insulin resistance and mild impairment of β-cell function, resulting in a range of biological and biochemical responses^7,8. Together with the evolution of the redox state, these changes further exacerbate the disruption of glucose metabolism and elevate the potential risk of diabetes complications, including retinopathy, nephropathy, and cardiovascular disease^9,10. Therefore, monitoring oxidative capacity (OC) and redox potential (RP) in individuals or animal models with IGT should improve our understanding of the mechanisms underlying IGT development and guide the design of targeted therapeutic strategies.

Acupuncture has a history of more than 3,000 years and has shown good clinical efficacy in the treatment of many diseases¹¹. Moreover, numerous experimental studies in animals have demonstrated that acupuncture exerts beneficial effects in regulating oxidative stress states in animal models by enhancing the antioxidant enzyme system and reducing lipid peroxidation¹². As a traditional Chinese medical practice, acupuncture has also demonstrated significant benefits in the treatment of diabetes and IGT. Guanyuan (CV4) and Sanyinjiao acupuncture points (SP6) have been reported to activate the SIRT1/PGC-1α pathway in skeletal muscle, thereby alleviating insulin resistance of obesity-diabetic mice¹³. Moreover, Tianshu acupuncture points (ST25) modulate the activity of glucose-inhibitory neurons located in the lateral hypothalamic region and contribute to the regulation of glycolipid metabolism¹⁴. However, most studies on the mechanisms of intravascular acupuncture have focused primarily on the neuromodulatory, immunomodulatory, and gastrointestinal systems, in which acupuncture may affect glucose metabolism through modulation of autonomic function or indirectly regulate blood glucose levels through enhancing the balance of intestinal flora^15,16. The specific mechanism of the effect of acupuncture in IGT treatment and its effects on oxidative stress have lacked in-depth investigation and experimental data support.

Surface-enhanced Raman scattering (SERS) provides great sensitivity and specificity, which enables the detection of a wide variety of analytes. SERS nano-probes based on nano-fiber optics¹⁷, chitosan membranes¹⁸, and previous studies have demonstrated that acupuncture needles¹⁹ can be used for the detection of biomarkers, neurotransmitters, and other aspects of cellular biochemistry in vivo. Acupuncture needles, in particular, represent promising tools for deep tissue testing, enabling minimally invasive in situ monitoring followed by ex vivo analysis. Acupuncture, as a well-established therapeutic method, has an extremely wide range of clinical applications. Recent studies have shown that SERS-enabled optical fibers, wound dressings, and microneedle devices can support in situ or minimally invasive tracking of redox-associated signals in biologically relevant settings^20,21. In parallel, SERS-active acupuncture needles and related microneedle systems have emerged as promising carriers for integrating sensing capability with tissue intervention¹⁹, resulting in SERS-active microneedles with in vivo response and in vitro SERS detection capabilities.

The present study builds on the previously reported SERS-active microneedle strategy for detecting oxidative-capacity- and redox-potential-related signals in glucose-induced stress models. However, the present work differs from the previous study in both biological context and experimental purpose. Specifically, this study applies the SERS-active microneedle platform to a rat model of impaired glucose tolerance and integrates local ratiometric SERS tracking with acupuncture intervention. The current study further compares acupuncture points with adjacent non-acupuncture sites, evaluates different needling manipulations, and monitors redox-associated SERS indices during a 7-day intervention period. Compared with previously reported SERS-based optical fibers, wound dressings, and SERS-active microneedles designed primarily for single-site or single-function biochemical detection, the present platform provides two advances. First, OC- and RP-responsive probes are integrated into two independent etched grooves on a single acupuncture needle, enabling simultaneous monitoring of complementary redox parameters. Second, the acupuncture needle functions not only as a SERS probe carrier but also as a therapeutic device, allowing redox monitoring to be performed during acupuncture intervention rather than after treatment. As shown in Figure 1, after the microneedles were inserted into tissues, tiny wounds are produced, and the probes in the two grooves react with tissues, enabling them to detect levels of OC and RP in vivo. The aim of the study was to use these SERS-active microneedles to examine changes in the OC and RP in rats with IGT and determine the effects of acupuncture matched to acupuncture points in these animals, thereby obtaining valuable insights into the mechanism of effects of acupuncture in treating IGT and the relationship between redox and abnormal hyperglycemic states.

Protocol

Animals and models of IGT

All animal procedures were approved by the Animal Ethics Committee of Southeast University Suzhou Research Institute (approval no. SEU-IACUC-20250318007) and were performed in accordance with institutional guidelines for the care and use of laboratory animals. The rats were obtained and housed under specific-pathogen-free (SPF) conditions at 25 °C with a 12 h light/12 h dark cycle. For procedures requiring anesthesia, rats were anesthetized using isoflurane delivered in oxygen through an induction chamber and maintained using a nose cone. Isoflurane was used at 3% for induction and 1% for maintenance, with the concentration adjusted according to respiratory rate and procedural response.

Adequate anesthetic depth was confirmed before needling or injection by loss of the righting reflex and absence of response to toe pinch. During anesthesia, respiratory pattern, body position, and recovery were monitored, and animals were placed on a warming pad to maintain body temperature. Animals were housed 2 per cage with free access to food and water unless fasting was required for glucose testing. After arrival, rats were acclimatized for 1 week before model induction. Animal welfare was monitored daily, including general activity, grooming, food intake, body weight, injection-site condition, and recovery after anesthesia. At the end of the study, rats were euthanized according to the approved animal protocol. Euthanasia was performed by gradual-fill CO₂ inhalation, followed by confirmation of death using the cervical dislocation method. Animals that met humane endpoint criteria, including severe weight loss, persistent immobility, abnormal respiration, inability to eat or drink, or severe distress, were euthanized immediately rather than being maintained until the scheduled endpoint. Specific-pathogen-free-grade male Sprague–Dawley rats weighing 200 ± 20 g were used in this study. Following 1 week of acclimatization, the rats were randomly assigned into two groups: the three rats in the normal group were fed with normal chow, whereas the 20 rats in the model group were fed with high-fat chow (50% basal chow + 10% lard + 25% sucrose + 15% yolk powder) for 3 weeks. Animals in the model group received intraperitoneal injection of 20 mg/kg of 2% STZ solution (solubilized in 0.1 mol/L, pH 4.4 citrate buffer) for 12 h, following 12 h of fasting with no water restriction. Those in the normal group received an intraperitoneal injection of the corresponding dose of citrate buffer solution after being fed normal chow. An intraperitoneal glucose tolerance test (IPGTT) was performed 1 week later. Reagents and equipment used in this study are listed in the Table of Materials.

Fasting blood glucose (FPG) was measured by blood collection from the tail tip of rats in each group after 12 h of fasting with no water restriction, and 2 h postprandial blood glucose (2hPG) was recorded following intraperitoneal injection of 20% dextrose 2 g/kg. In the normal group, the FPG range was 5.1–6.6 mmol/L, and the 2hPG glucose range was 7.3–9.4 mmol/L, compared with an FPG of ≤7.0 mmol/L and a 2hPG range of 10–16 mmol/L in the model group. Among the 20 rats assigned to IGT model induction, 18 met the predefined IGT criteria after IPGTT and were included in the subsequent intervention experiment. Two rats were excluded because their FPG or 2hPG values fell outside the predefined IGT range. The 18 eligible rats were then randomly assigned to a high-fat-diet group and a normal-diet group, with 9 rats per group. Each diet group was further divided into three subgroups: acupuncture-point twisting, non-acupuncture-site twisting, and acupuncture-point direct insertion, with three rats per subgroup. This was an exploratory method-development study designed to evaluate the feasibility of combining acupuncture intervention with SERS-based ratiometric tracking in a preclinical IGT rat model. No formal statistical power calculation was performed. The subgroup size of three rats was selected based on feasibility, preliminary SERS-microneedle studies, and the animal-reduction principle. Rats were randomly assigned by the experimenter; formal allocation concealment and blinding were not performed.

The acupuncture treatment protocols were screened with reference to evidence-based clinical practice guidelines for traditional Chinese medicine, and the feasibility of animal experiments²². The selected acupoints were Zhongwan (RN12/CV12), Shuifen (RN9/CV9), Yinjiao (RN7/CV7), bilateral Tianshu (ST25), bilateral Shuidao (ST28), and bilateral Zusanli (ST36). RN12, RN9, and RN7 were located on the anterior midline and were treated as single midline points, whereas ST25, ST28, and ST36 were treated bilaterally. The acupuncture point pairings were based on those used for clinical treatment of IGT²³. Adaptation was based on the literature^24,25, using the dichotomous method. For rats, acupoints were localized according to anatomical landmarks and proportional conversion from standard human acupoint locations. The abdominal RN/CV points were identified along the anterior midline using the xiphoid process, umbilicus, and pubic symphysis as anatomical landmarks. ST25 was marked bilaterally at the umbilical level, and ST28 was marked bilaterally in the lower abdominal region at the same lateral distance from the anterior midline. ST36 was identified on the anterolateral hind limb in the tibialis anterior region, lateral to the tibial tubercle. Adjacent non-acupuncture comparison sites were marked 5 mm lateral to the corresponding acupoints and outside the recognized meridian line, while avoiding visible vessels, scars, and previous puncture sites. Before needling, rats were anesthetized with isoflurane according to the approved animal protocol and placed in the supine position on a warming pad. The abdominal and hindlimb skin over each target site was shaved and disinfected with 75% ethanol. SERS-active microneedles were inserted approximately perpendicular to the skin to a depth of approximately 10 mm. For the acupuncture-point twisting groups, SERS-active microneedles were inserted into the selected acupoints and gently rotated bidirectionally during the 20 min retention period. For the non-acupuncture-site groups, the same procedure was performed at the 5 mm adjacent non-acupuncture comparison sites. For the direct-insertion groups, SERS-active microneedles were inserted into the acupoints and retained for 20 min without twisting. All needling procedures were performed by the same trained operator to minimize inter-operator variation.

Needling was performed twice daily, at 10:00 and 15:00, for 7 consecutive days. Before each day’s procedure, FPG was measured after 8 h of fasting without water restriction. After each day’s needling procedure, OC and RP responses were recorded. On day 7, after acupuncture treatment, IPGTT was performed, and 2hPG values were recorded.

Cautions

Streptozotocin is hazardous and should be handled as a toxic chemical. Prepare the streptozotocin solution in a chemical fume hood or a certified biosafety cabinet, while wearing a laboratory coat, nitrile gloves, protective eyewear, and a mask. Streptozotocin-contaminated syringes, needles, wipes, bedding, cages, and animal-related waste should be handled in accordance with institutional hazardous biological and chemical waste procedures. Used needles and syringes should be discarded immediately into puncture-resistant sharps containers without recapping.

Fabrication of SERS-active microneedles for the detection of OC and RP

Acupuncture needles were prepared as previously reported²⁶. SERS-active microneedles were fabricated using the same type of commercial stainless-steel acupuncture needles (0.25 mm × 26 mm) throughout the study. The needles were inserted approximately perpendicular to the skin to a depth of 10 mm. For twisting groups, manual bidirectional rotation was applied gently and consistently for 20 min during the retention period by the same trained operator. The manipulation was not instrument-controlled; therefore, exact rotation frequency and amplitude were not quantified. This limitation has been acknowledged in the Discussion.

Initially, the needles were soaked in a 20% (w/v) PMMA solution in MMA, followed by drying at room temperature (25 °C) for 1 day. Next, two parallel indentations of uniform size and similar shape and depth in the PMMA protective layer (5 mm apart)were made using a small craft knife blade that had been cleaned with ethanol; then, the needles were submerged in 0.5 M sulfuric acid serving as the electrolyte for 10 s to undergo electrochemical etching at a voltage of 12 V, thereby generating two nearly circular grooves at the scoring sites. Third, after rinsing on three occasions, the needles were immersed at room temperature for 24 h in an ethanol solution containing 0.1 mol L⁻¹ MPTES and 0.3 mol L⁻¹ APTES, thereby conferring sulfhydryl functionality to the grooved surfaces. Fourth, the coating was stripped off by dipping the needle into MMA. Redox-potential-responsive SERS-active microneedles were prepared based on a previously reported method²⁷. The SERS substrate was adsorbed onto one of the etched grooves of each acupuncture needle, after which the resulting SERS-active microneedles were submerged in an aqueous p-PDA solution at a concentration of 1 × 10^-4 mol/L for 60 min at 37 °C, yielding oxidation-responsive SERS-active microneedles. SERS substrates based on gold nanoshells were synthesized as described previously²⁸. In the present work, these microneedles were dripped into the other groove of each acupuncture needle, which were then allowed to dry naturally. This yielded SERS-active microneedles with OC and RP response.

Ex vivo evaluation of OC and RP response to SERS-active microneedles

The responses of the OC and RP probes were evaluated by inserting the SERS-active microneedles into ex vivo tissue. To verify the responsiveness of the OC probe to oxidative changes, SERS spectra of p-PDA were recorded before and after oxidation. Spectra of the redox SERS probes were also acquired in a strongly reducing NaBH₄ solution and a strongly oxidizing HAuCl₄ solution. The selectivity of the OC probe was evaluated using reducing agents, including 1 × 10^-4 M FeSO₄, 1 × 10^-4 M Na₂SO₄, and 1 × 10^-4 M ascorbic acid, and oxidizing agents, including 1 × 10^-4 M CuSO₄, 1 × 10^-4 M H₂O₂, 1 × 10^-4 M FeCl₃, and 1 × 10^-4 M HAuCl₄. All solutions were prepared in 20 mM PBS (pH 7.4), and the probes were incubated with each solution for 20 min at 37 °C. The robustness of the SERS-active microneedles was assessed by recording OC responses before and after reaction with 1 × 10^-4 mol/L FeCl₃ at room temperature. These ex vivo and solution-based tests were used to confirm probe responsiveness under controlled conditions. However, they do not fully reproduce the complexity of the in vivo tissue microenvironment, where local pH variation, protein adsorption, extracellular matrix components, and other biological substances may affect SERS signals. Therefore, the in vivo OC and RP readouts were interpreted as relative redox-associated SERS indices rather than as matrix-independent biochemical measurements.

OC and RP detection at acupuncture and non-acupuncture sites

To evaluate the performance of the OC and RP probes at acupuncture points and non-acupuncture points, the SERS-active microneedles were inserted into healthy rats at the ST36 and RN12 acupuncture points, and 5 mm adjacent to each of these points, at a depth of approximately 10 mm, for the same amount of time.

Effects of needle twisting on OC and RP detection

To assess the responsiveness of the OC and RP probes under different needling techniques, SERS-active microneedles were inserted into the ST36 and RN12 acupuncture points and 5 mm adjacent to these points in healthy rats to a depth of about 10 mm. At each acupuncture point, one needle was twisted, whereas the corresponding needle adjacent to the point was not, and vice versa (an acupuncture-point-adjacent needle was twisted, whereas its counterpart at the acupuncture point was not). SERS signals from the probes were collected at the same insertion time for each needle.

Changes in OC and RP in rats with IGT with respect to acupuncture points, needle twisting, and dietary conditions

To assess the effects of acupuncture points, needle twisting, and dietary conditions on OC and RP of rats with IGT, rats in the high-fat diet group were randomly assigned to three groups (A1, A2, and A3; three rats in each group), as were those in the normal diet group (groups B1, B2, and B3; three rats in each group). For rats in groups A1 and B1, SERS-active microneedles were inserted into acupuncture points after sedation and twisted for 20 min; the same procedure was carried out for rats in groups A2 and B2, except that the microneedles were inserted 5 mm adjacent to the acupuncture points. For rats in A3 and B3, the SERS-active microneedles were inserted into rat acupuncture points without twisting; the needles were left in place for 20 min. Needling was performed twice per day, at 10:00 and 15:00, for 7 days. Before each day’s procedures, a glucose meter was used to record FPG values for each group of rats (after fasting but no water restriction for 8 h). After each day’s needling, OC and RP responses were measured and recorded. On day 7, following acupuncture, IPGTT was performed on each group of rats, and 2hPG values were recorded.

Characterization and measurement

The morphology of SERS-active microneedles was examined with scanning electron microscopy (SEM). SERS spectra were acquired at 25 °C, using a Raman spectrometer configured with a 50× long working distance objective (NA: 0.5) and a 785 nm laser. For all measurements, the acquisition time was 10 s, and the Raman system was operated at a nominal laser-power setting of 600 µW under the same optical configuration. The actual laser power delivered at the tissue surface was not directly measured in this study. Therefore, the SERS spectra were used for relative comparisons between groups under identical acquisition settings, and no conclusions were drawn regarding absolute tissue photothermal safety. The intensity ratios of two SERS probes were calculated using the following approach: the ratio of intensity values at 1450 cm⁻¹ to that at 1500 cm⁻¹ was designated as the indicator for OC (I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹), after subtracting the intensity at 1700 cm⁻¹ (as background ) from each of the two peaks, respectively, as in previous reports ¹⁷; and using I₁₆₀₆ cm⁻¹/I₁₆₆₆ cm⁻¹ as an indicator of RP, the intensity at 1700 cm⁻¹ (as background ) was subtracted from each of the two peaks, respectively. The calculated I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹ and I₁₆₀₆ cm⁻¹/I₁₆₆₆ cm⁻¹ values were analyzed as relative SERS-derived indices of OC- and RP-associated responses, respectively. These ratios were used to compare changes between groups and time points under identical acquisition settings.

No external calibration curve was established in the present study to convert these ratios into absolute concentrations of oxidative species or absolute electrochemical redox-potential values. Each spectrum was recorded twice at five different points on the SERS probe in three replicate samples. Origin was used to process data and produce charts. To minimize acquisition-related variability, the same laser wavelength, objective, acquisition time, laser-power setting, and spectral-processing workflow were applied to all groups. Potential photothermal or photochemical effects caused by laser irradiation were not independently quantified and are acknowledged as a limitation. Statistical data are presented as mean ± standard error of the mean (SEM). The sample size and the statistical test used for each experiment or figure panel are specified in the corresponding figure legends. For ex vivo probe-validation experiments, the SERS probe or independently prepared microneedle was used as the analysis unit. For in vivo blood-glucose and SERS-index analyses, the rat was used as the biological analysis unit. Spectra collected from multiple points on the same probe were treated as technical measurements and averaged before group-level analysis. Statistical analyses were conducted using SPSS software. Comparisons between groups were performed using the statistical tests indicated in the corresponding figure legends. Serial fasting blood glucose and SERS-index measurements over the 7-day intervention period were interpreted as exploratory longitudinal trends because of the small subgroup size, and no formal repeated-measures or mixed-effects model was applied. Statistical significance was defined as p < 0.05. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Results

Fabrication of SERS-active microneedles for the detection of OC and RP

To identify changes in OC and RP, SERS-active microneedles containing the corresponding probes were fabricated. Representative SEM images confirmed the two-groove structure of the SERS-active microneedles. Figure 2A shows the acupuncture needle body with two etched grooves, Figure 2B shows a groove loaded with GNS, and Figure 2C shows an enlarged view of GNS distributed within the groove. These images confirmed that the etched grooves could serve as localized loading sites for the OC- and RP-responsive SERS probes. The ratio of peak intensities I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹ was used to represent the degree to which p-PDA was oxidized. Figure 3A displays SERS spectra of p-PDA before and after interaction with 1 × 10⁻⁴ M FeCl₃, whereas Figure 3B presents spectra for the redox probe in 1 × 10⁻³ M NaBH4 and 1 × 10^-4 M HAuCl₄ aqueous solutions. The RP probe response was expressed as a peak intensity ratio (I₁₆₀₆ cm^-1/I₁₆₆₆ cm^-1), which decreased with increasing RP.

To evaluate the selectivity of the oxidative-capacity-responsive probe, the I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹ ratio was measured after exposure to different reducing and oxidizing agents. Independent probe measurements were performed with n = 3 per group, and the SERS probe or independently prepared microneedle was used as the analysis unit (Figure 3C). Compared with the CK group (Kruskal–Wallis test followed by Dunn's multiple comparisons test), the reducing agents FeSO₄ and Na₂SO₄ did not significantly alter the I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹ ratio (p > 0.05). Ascorbic acid (AA, p < 0.001) also produced an increased ratio under the present assay condition, despite being included as a reducing agent. Therefore, the AA response was interpreted cautiously as a probe-specific spectral response in this solution-based assay rather than as evidence of oxidizing behavior. In contrast, oxidizing or oxidation-associated reagents increased the ratio to varying degrees. The ratio was increased after exposure to CuSO₄ (p < 0.0001), H₂O₂ (p < 0.0001), FeCl₃ (p < 0.0001) and HAuCl₄ (p < 0.0001) To assess the stability of the OC probe at room temperature, SERS spectra were acquired at hourly intervals over a 6 h period before reaction of the probe with 1 × 10⁻⁴ M FeCl₃ (Figure 3D), and the corresponding I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹ ratios were calculated. After probe adsorption, the initial value remained stable at approximately 0.20 ± 0.02, as did that for the reaction with FeCl₃ at about 0.9 ± 0.10, over the 6 h period. These results suggest that p-PDA on GNS is stable in air.

OC and RP responses at acupuncture and non-acupuncture points

Previous studies²⁹ have shown that the OC probe achieves its maximum response 10 min after insertion. Figures 4A and B illustrate the response of SERS-active microneedles in healthy rats placed in the ST36 and RN12 acupuncture points, as well as those placed 5 mm adjacent to the points, for 10 min. For these experiments, n = 3 rats per group, and spectra collected from multiple points on the same probe were treated as technical replicates and averaged before group-level analysis. The SERS-active microneedles responded effectively to changes in OC and RP in vivo, with no significant difference in OC and RP in the acupuncture points group relative to the non-acupuncture points group (Independent samples t-test, p > 0.05).

To further determine the effects of needle twisting on local OC and RP in vivo, SERS-active microneedles were inserted into healthy rats at the ST36 and RN12 acupuncture points, and 5 mm beyond these points, to a depth of 10 mm. During the twisting maneuver on each acupuncture point, the needle adjacent to the point was untwisted, and vice versa, and changes in OC and RP values at the point of puncture were recorded (Figures 4C–F). There were no significant changes in the OC and RP according to whether twisting was performed at either acupuncture points or non-acupuncture points in healthy rats, nor was there any effect of twisting at an adjacent point (p > 0.05).

Changes in OC and RP in rats with IGT associated with acupuncture, needle twisting, and dietary conditions

To confirm the successful establishment of a rat model of IGT, FPG and 2hPG values were recorded using an IPGTT assay in model and control rats, respectively. The control group included n = 3 rats, and the model group included n = 18 rats that met the predefined criteria for impaired glucose tolerance. The model rats showed higher fasting blood glucose than control rats (6.92 mmol/L versus 5.54 mmol/L; Mann–Whitney U test, p < 0.01) and higher 2 h postprandial blood glucose than control rats (11.52 mmol/L versus 7.92 mmol/L; Mann–Whitney U test, p < 0.001) (Figure 5), indicating successful induction of IGT in the model animals.

To verify the effects of the combination of acupuncture points, needle twisting, and dietary conditions on IGT rats, a typical set of acupuncture point pairs was selected. The 18 eligible model rats were then divided into high-fat-diet and normal-diet groups, with n = 9 rats per diet group. Each diet group was further divided into three intervention subgroups: acupuncture-point twisting, non-acupuncture-site twisting, and acupuncture-point direct insertion, with n = 3 rats per subgroup. FPG values were recorded before each day of needling (Figures 6A–C), and 2hPG (Figure 6D) values were measured after 7 days of needling by IPGTT assay in each group. In this exploratory dataset, no clear difference in FPG or 2hPG was observed between the high-fat-diet and normal-diet model groups after needling at acupuncture points. Comparisons between acupuncture-point and non-acupuncture-site needling suggested that acupuncture-point twisting was associated with lower blood-glucose values under the present experimental conditions (independent samples t-tests, p < 0.05). Because the subgroup size was small and diet, acupoint location, and twisting manipulation could not be fully statistically separated, these findings were interpreted as preliminary animal model observations.

Changes in oxidative-capacity-related and redox-potential-related SERS indices were monitored during 1 week of acupuncture intervention in IGT rats. For these analyses, n = 3 rats were included per subgroup, and each rat was treated as the biological analysis unit. In high-fat-diet rats, the oxidative-capacity-related SERS index decreased over time in the acupuncture-point twisting and acupuncture-point direct-insertion groups, whereas it remained relatively higher in the non-acupuncture-site twisting group (Figure 7A). The redox-potential-related SERS index increased over time in the acupuncture-point twisting and acupuncture-point direct-insertion groups, whereas the non-acupuncture-site twisting group showed a comparatively weaker increase (Figure 7B). Similar trends were observed in normal-diet IGT rats. The oxidative-capacity-related SERS index decreased over time in the acupuncture-point twisting and acupuncture-point direct-insertion groups, with a smaller change in the non-acupuncture-site twisting group (Figure 7C). The redox-potential-related SERS index increased over time in the acupuncture-point twisting and acupuncture-point direct-insertion groups, whereas the non-acupuncture-site twisting group remained comparatively lower (Figure 7D).

To further compare dietary conditions under the same acupuncture manipulation, the acupuncture-point twisting subgroups from the high-fat-diet and normal-diet groups were compared. In this comparison, the oxidative-capacity-related SERS index decreased over time in both diet conditions, with a larger decrease observed in the normal-diet acupuncture-point twisting subgroup by day 7 (Figure 7E). The redox-potential-related SERS index increased over time in both diet conditions, with a higher day 7 value in the normal-diet acupuncture-point twisting subgroup (Figure 7F). These findings suggest that acupuncture-point twisting and dietary normalization were associated with changes in local redox-associated SERS indices in IGT rats. Because each subgroup contained only three rats and no independent biochemical oxidative-stress assays were performed, these results should be interpreted as exploratory animal-model observations rather than definitive biochemical evidence of oxidative-stress improvement.

DATA AVAILABILITY:

All numerical raw data used for figure generation and statistical analysis, including blood-glucose measurements, SERS-derived ratio values, and source data for quantitative figure panels, have been organized and uploaded as Supplementary Table 1. For continuous probe-monitoring experiments generated directly by the analytical instrument, the exported values and representative instrument-generated outputs used to evaluate probe performance are included in the Supplementary File 1.

Figure 1: Methodological workflow for acupuncture intervention and local ratiometric SERS tracking using a dual-groove SERS-active microneedle. The schematic illustrates insertion of the two-groove SERS-active microneedle into an acupuncture point or adjacent non-acupuncture site, followed by Raman spectral acquisition. The authors created this figure using the WPS software, and no copyright license is required. Abbreviations: SERS = surface-enhanced Raman scattering. Please click here to view a larger version of this figure.

Figure 2: Representative SEM images of SERS-active microneedles. (A) Acupuncture needle body featuring two grooves; scale bar = 200 µm; magnification = 77×. (B) A groove integrated with GNS; scale bar = 20 µm; magnification = 1,000×. (C) An enlarged image of the groove in B; scale bar = 2 µm; magnification = 30,000×. Abbreviations: SEM = scanning electron microscopy; GNS = gold nanoshells. Please click here to view a larger version of this figure.

Figure 3: Characterization of OC-responsive and RP-responsive SERS probes. (A) Normalized spectra of p-PDA before and after interaction with FeCl₃. (B) Normalized SERS spectra of RP probes in solutions of NaBH₄ and HAuCl₄, respectively. (C) I₁₄₅₀ cm⁻¹/I₁₅₀₀ cm⁻¹ ratios for OC probes in various oxidizing and reducing solutions. n=3 independent probe measurements per group. Kruskal–Wallis test followed by Dunn’s multiple comparisons test versus the CK group. Data are presented as mean ± standard error of the mean (SEM). (D) Six-hour stability assessment of the OC-responsive probe before and after reaction with FeCl₃. Abbreviations: SERS = surface-enhanced Raman scattering; CK = control check; p-PDA = p-phenylenediamine; AA = ascorbic acid. ***p < 0.001, ****p < 0.0001. Please click here to view a larger version of this figure.

Figure 4: OC- and RP-related SERS indices measured with different needling methods at acupuncture and non-acupuncture sites in healthy rats. (A) OC-related SERS index after direct insertion at acupuncture points and adjacent non-acupuncture sites. (B) RP-related SERS index after direct insertion at acupuncture points and adjacent non-acupuncture sites. (C) OC-related SERS index after direct insertion at non-acupuncture sites and needle twisting at acupuncture points. (D) RP-related SERS index after direct insertion at non-acupuncture sites and needle twisting at acupuncture points. (E) OC-related SERS index after direct insertion at acupuncture points and needle twisting at adjacent non-acupuncture sites. (F) RP-related SERS index after direct insertion at acupuncture points and needle twisting at adjacent non-acupuncture sites. OC was calculated as I_{1450 cm}^-1/I_{1500 cm}^-1, and RP was calculated as I_{1606 cm}^-1/I_{1666 cm}^-1. n=3 each group. Data are presented as mean ± SEM. No significant differences were observed between the paired comparisons. Abbreviations: OC = oxidative capacity; RP = redox potential. Please click here to view a larger version of this figure.

Figure 5: FPG and 2hPG values according to the IPGTT assay in model and control (CK) rats. FPG and 2hPG were measured to verify the establishment of the impaired glucose tolerance rat model. n = 3 rats in the control group and n = 18 rats in the model group. Statistical analysis was performed by the Mann–Whitney U test, and the results are presented as the mean ± SEM. **p < 0.01, ***p < 0.001. Please click here to view a larger version of this figure.

Figure 6: FPG and 2hPG values for IGT rats during acupuncture procedures. (A) FPG values for the group receiving acupuncture with needle twisting at acupuncture points and direct needle insertion at acupuncture points (n = 3 per group). (B) FPG values for group receiving needling with twisting at acupuncture points and non-acupuncture points (n = 3 per group). (C) FPG values for the high-fat diet and normal diet groups (n = 3 per group). (D) Day 7 2hPG values in high-fat-diet and normal-diet rats under the acupuncture-point twisting, non-acupuncture-site twisting, and acupuncture-point direct-insertion conditions. (n = 3 per group). Results are presented as the mean ± SEM. Planned pairwise endpoint comparisons in panel D were analyzed using an unpaired two-tailed Student’s t-test. *p < 0.05. Please click here to view a larger version of this figure.

Figure 7: Changes in OC and RP of rats in each group during 1 week of acupuncture. (A) Changes in OC in high-fat diet IGT rats. (B) Changes in RP in high-fat diet IGT rats. (C) Changes in OC in normal diet IGT rats. (D) Changes in RP in normal diet IGT rats. (E) OC-related SERS index comparison between the high-fat-diet and normal-diet conditions within the acupuncture-point twisting subgroup; (F) RP-related SERS index comparison between the high-fat-diet and normal-diet conditions within the acupuncture-point twisting subgroup. n = 3 rats per subgroup. Data are presented as mean ± SEM. Serial measurements were interpreted as exploratory longitudinal trends. Abbreviations: OC = oxidative capacity; RP = redox potential. Please click here to view a larger version of this figure.

Supplementary File 1: Representative instrument-generated outputs for probe-response evaluation. This file contains representative continuous monitoring outputs generated directly by the analytical instrument and used to evaluate probe-response performance.Please click here to download this file.

Supplementary Table 1: Raw data table for quantitative analyses. This file contains the numerical source data used for figure generation and statistical analysis, including blood-glucose measurements, SERS-derived ratio values, and quantitative figure panel data.Please click here to download this file.

Discussion

Impaired glucose tolerance represents a reversible transitional stage before type 2 diabetes and is closely associated with metabolic imbalance, insulin resistance, and oxidative-stress-related dysregulation. Lifestyle intervention and other early therapeutic strategies can reduce the risk of progression from impaired glucose tolerance to diabetes, but tools for dynamically monitoring local biochemical responses during intervention remain limited³⁰. Acupuncture has been reported to affect glucose metabolism and oxidative-stress-related pathways in experimental and clinical studies. However, most mechanistic studies of acupuncture rely on endpoint blood or tissue measurements, and relatively few methods allow minimally invasive tracking of local redox-associated responses during acupuncture intervention. Previous clinical trials of treatments for IGT have confirmed that the risk of diabetes can be largely reduced through lifestyle interventions, including dietary modifications, body mass reduction, and increased physical activity³¹. Acupuncture has been shown to be effective in improving glucose metabolic indexes and lowering fasting and 2hPG. Therefore, elucidation of the relationship between acupuncture, redox, and IGT may provide approaches to reverse IGT or intervene in its progression, thereby reducing the incidence of diabetes mellitus, as well as providing insight into the pathology of IGT and the mechanisms underlying its development.

The present study proposes a dual-groove redox-sensitive surface-enhanced Raman scattering-active microneedle platform for combining acupuncture intervention with local ratiometric SERS tracking in a rat model of impaired glucose tolerance. In this system, two etched grooves on a single acupuncture needle were used to carry oxidative-capacity-responsive and redox-potential-responsive probes. This design allowed the acupuncture needle to function not only as an intervention tool but also as a sensing carrier for relative local redox-associated optical readouts. The results showed that needling at acupuncture points with twisting was associated with lower fasting blood glucose and 2hPG values and with changes in oxidative-capacity-related and redox-potential-related SERS indices in impaired-glucose-tolerance rats. These findings support the feasibility of using SERS-active microneedles to explore local redox-associated responses during acupuncture intervention. Overall, the results suggest that needling at acupuncture points with twisting was associated with lower FPG and 2hPG values and with changes in OC- and RP-related SERS indices in IGT rats. These findings support the feasibility of using SERS-active microneedles to follow local redox-associated optical responses during acupuncture intervention. However, because standard biochemical oxidative-stress assays, probe-leaching analysis, and tissue biocompatibility tests were not performed, the observed SERS-index changes should not be interpreted as definitive biochemical proof of oxidative-stress improvement.

Several alternative approaches could be used to further investigate the same hypothesis. Standard biochemical oxidative-stress assays, including measurements of reactive oxygen species, malondialdehyde, superoxide dismutase, catalase, glutathione peroxidase, and glutathione/glutathione disulfide ratios, could be used to validate whether changes in SERS-derived indices correspond to conventional oxidative-stress biomarkers^32,33. Electrochemical redox probes, fluorescence-based redox sensors, chemiluminescence assays, microdialysis followed by biochemical analysis, and tissue-based immunohistochemical or histological validation may also provide complementary information. Combining SERS-active microneedle sensing with these independent approaches would help determine whether the observed optical changes reflect specific biochemical redox alterations or broader local tissue responses to needling and metabolic status³⁴.

The potential applications of this method extend beyond the present impaired-glucose-tolerance model. Because redox imbalance is involved in metabolic disease, tissue injury, inflammation, and wound repair, SERS-active microneedles may provide a useful preclinical tool for monitoring local redox-associated responses in diabetes-related models, wound-healing studies, inflammatory disease models, and other minimally invasive intervention settings³⁵. In acupuncture research, the platform may be useful for comparing acupuncture points, non-acupuncture sites, different needling manipulations, and different intervention schedules. However, these applications remain exploratory and require further technical and biological validation.

Several limitations should be noted. First, the OC and RP readouts were calculated as SERS-derived ratiometric indices and were not calibrated against standard concentrations of oxidative species or absolute electrochemical redox-potential values. Therefore, the present results should be interpreted as relative redox-associated optical responses under identical acquisition conditions rather than absolute biochemical measurements. Second, the biological tissue microenvironment may introduce additional sources of signal variation, including local pH changes, protein adsorption, extracellular matrix components, and other matrix effects. These factors were not fully evaluated in serum-containing or tissue-mimicking media. Third, the actual laser power delivered to the tissue surface and the local temperature change during Raman acquisition were not directly measured; therefore, potential photothermal or photochemical effects cannot be completely ruled out. Fourth, batch-level reproducibility of groove geometry and probe loading was not quantitatively evaluated by multi-needle SEM imaging or loading-efficiency analysis. Finally, independent biochemical validation, probe-leaching analysis, and local biocompatibility testing were not performed. Future studies should establish calibration curves in biologically relevant matrices, correlate SERS indices with standard oxidative-stress and redox assays, quantify probe leaching, evaluate local tissue compatibility after repeated insertion, and validate the device in larger animal cohorts. SERS-active microneedles may provide a useful platform for combining acupuncture intervention with relative ratiometric SERS tracking of local redox-associated indices in a preclinical IGT rat model. Further calibration, biochemical validation, probe-leaching assessment, and biocompatibility studies are required before absolute quantitative monitoring or translational application can be claimed.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
3-Aminopropyltriethoxysilane (APTES, 98%)	Tianjin Heowns Bio-Chemical Technology Co., Ltd.	A800524
(3-Mercaptopropyl)trimethoxysilane (MPTES, 95%)	Tianjin Heowns Bio-Chemical Technology Co., Ltd.	M742544
Ascorbic acid	Sinopharm Chemical Reagent Co., Ltd.	CFAD310554
Blood glucose meter	Cofoe Medical Technology Co., Ltd.	Cofoe A03
Blood glucose test strips	Cofoe Medical Technology Co., Ltd.	Cofoe-A03-C
Copper sulfate (CuSO4)	Sinopharm Chemical Reagent Co., Ltd.	PHR1477
Ethyl alcohol (≥99.7%)	Sinopharm Chemical Reagent Co., Ltd.	XW00641752
Ferrous sulfate (FeSO4)	Sinopharm Chemical Reagent Co., Ltd.	XW01772078701
Glucose (Glu)	Tianjin Heowns Bio-Chemical Technology Co., Ltd.	D9434
Hydrogen peroxide (H2O2)	Sinopharm Chemical Reagent Co., Ltd.	P-17-0500
Iron chloride (FeCl3)	Sinopharm Chemical Reagent Co., Ltd.	CP81015228
Methyl methacrylate (MMA)	Shanghai Aladdin Bio-Chem Technology Co., Ltd.	M109626
Micro-Raman spectrometer	Renishaw	InVia Raman Microscope
p-Phenylenediamine (p-PDA)	Shanghai TCI Chemical Industry Development Co., Ltd.	P128784
Polymethyl methacrylate (PMMA)	Tianjin Heowns Bio-Chemical Technology Co., Ltd.	P742581
Scanning electron microscope	Zeiss	ULTRA Plus
Sodium borohydride (NaBH4)	Shanghai Adamas-beta Reagents Co., Ltd.	S432207
Sodium sulfite (Na2SO4)	Sinopharm Chemical Reagent Co., Ltd.	S572396
Stainless steel acupuncture needles	Suzhou Tianxie Acupuncture Instruments Co., Ltd.	X-20-03
Statistical analysis software	International Business Machines Corporation	Version 2021
Streptozotocin (STZ)	Shanghai Aladdin Bio-Chem Technology Co., Ltd.	S408311
Tetrachloroauric acid (HAuCl4)	Tianjin Heowns Bio-Chemical Technology Co., Ltd.	A124024
SPSS software	International Business Machines Corporation, Armonk, New York State, USA	version 26.0
rats	Jiangsu Qinglongshan Biotechnology Co., Ltd.	Specific-pathogen-free male Sprague-Dawley
basal chow	Anuokang Biotech.	SY10001
Lard	MCE	HY-W127601
Sucrose	MCE	HY-B1779
Egg yolk powder	MCE	HY-B2235B
Citric acid monohydrate	Sigma-Aldrich	C1909
Trisodium citrate dihydrate	Sigma-Aldrich	S4641
Citrate buffer	beyotime	P0086
Phosphate-buffered saline	beyotime	ST448-1L
Isoflurane	Bioss	D54468
Small-animal isoflurane anesthesia system	RWD Life Science	R500
Induction chamber and nose cone	RWD Life Science	R500compatible accessories
Warming pad	Shanghai Yuyan Life Science	Y69020
75% ethanol	Boyu Bio.	YB60401
CO2 euthanasia chamber	Lab Animal Tec. Co.	LAT-10-0090
WPS	Beijing Kingsoft Office Software	WPS365