Enhanced Neuromodulation Using PC-CNTs and NIR Photothermal Therapy for Preventing Ventricular Arrhythmias

Abstract

Background

Cardiac sympathetic hyperactivation of the left stellate ganglion (LSG) critically promotes ventricular arrhythmias (VAs) following myocardial infarction (MI). Although neuromodulatory strategies targeting sympathetic activity have shown therapeutic potential, current strategies are limited by invasiveness, incomplete targeting and transient efficacy. Therefore, safe and precise approaches for modulating LSG activity are needed. This study aimed to investigate the enhanced neuromodulatory effects of phospholipid-coated carbon nanotubes (PC-CNTs) combined with near-infrared (NIR) photothermal therapy on the LSG for preventing post-MI VAs.

Methods

PC-CNTs were synthesized and characterized, followed by evaluation of their photothermal properties and in vitro/in vivo biocompatibility. In a canine MI model induced by left anterior descending artery occlusion (n = 18), animals were randomly assigned to PBS, PC-CNTs and PC-CNTs + NIR groups. PBS or PC-CNTs (30 μg mL⁻¹, 0.1 mL) were locally microinjected into the LSG, and animals in the PC-CNTs + NIR group subsequently received 808 nm NIR laser irradiation (1.0 W cm⁻² for 5 min). LSG neural activity, heart rate variability (HRV), ventricular effective refractory period (ERP) and VA incidence were assessed. Transcriptomic profiling of LSG tissue and molecular analysis of peri-ganglionic adipose tissue were performed to evaluate therapeutic effects and potential mechanisms.

Results

PC-CNTs exhibited excellent photothermal conversion efficiency and favorable biocompatibility. In vivo, PC-CNTs treatment suppressed MI-induced LSG hyperactivity, improved cardiac autonomic balance, enhanced ventricular electrophysiological stability and reduced VA susceptibility compared with the PBS group. The addition of NIR photothermal therapy further enhanced these protective effects, resulting in further suppression of sympathetic activation and reduction of post-MI arrhythmias. Molecular analyses suggested that combined PC-CNTs and NIR treatment modulated the LSG microenvironment, as evidenced by reduced expression of inflammation-related genes (CXCL14, CSF3, TIMP1 and CRLF1), indicating attenuation of neuroinflammatory signaling. Additionally, NIR-induced local hyperthermia was associated with thermogenic browning of peri-ganglionic white adipose tissue, characterized by increased expression of thermogenic markers and reduced adipocyte size. These changes may contribute to reduced sympathetic hyperactivity and decreased arrhythmia susceptibility.

Conclusion

The combination of PC-CNTs and NIR photothermal therapy provides enhanced anti-arrhythmic effects associated with suppression of LSG hyperactivity and modulation of neuroinflammatory signaling and peri-ganglionic adipose browning, representing a promising preclinical strategy for preventing VAs.

Introduction

Ventricular arrhythmias (VAs) remain the primary cause of sudden cardiac death in patients with myocardial infarction (MI).^1,2 Cardiac sympathetic hyperactivation plays a critical role in the initiation and perpetuation of post-MI VAs, with the left stellate ganglion (LSG) serving as a pivotal peripheral node for sympathetic outflow to the heart.³ Although cardiac sympathetic innervation from the bilateral stellate ganglia partially overlaps, the LSG plays a particularly important role in ventricular sympathetic regulation and arrhythmogenesis.^4,5 Compared with other broader or more diffusely distributed autonomic targets, the LSG provides a relatively well-defined and anatomically accessible target for localized neuromodulation.^6,7 Following MI, excessive LSG activation enhances sympathetic outflow, destabilizing ventricular electrophysiology and promoting arrhythmogenesis.⁸ Moreover, inflammatory infiltration within the stellate ganglion may contribute to neuronal excitability and sustain sympathetic hyperactivity.⁹ Accordingly, targeted suppression of LSG hyperactivation represents a compelling therapeutic strategy for preventing post-MI VAs. Current strategies to modulate LSG activity, including cardiac sympathetic denervation,¹⁰ stellate ganglion blockade,⁵ vagal nerve stimulation,¹¹ and optogenetic modulation,¹² have demonstrated therapeutic potential but are limited by several clinical challenges. Cardiac sympathetic denervation is invasive and largely irreversible, whereas the effects of stellate ganglion blockade are typically temporary.¹³ Vagal nerve stimulation may have limited target specificity due to its broad autonomic effects, while the clinical translation of optogenetic modulation remains constrained by the need for genetic modification and specialized light-delivery systems.^12,14,15 Therefore, there is an urgent need for novel, safe, minimally invasive and targeted neuromodulatory strategies capable of mitigating sympathetic hyperactivity to prevent VAs.

Carbon nanotubes (CNTs) have attracted considerable attention as multifunctional nanomaterials owing to their exceptional electrical conductivity, high surface area, intrinsic near-infrared (NIR) photothermal conversion efficiency and interactions with neural membranes and neuronal circuits.^16,17 However, the biological responses to CNTs depend on their physicochemical properties, including dimensions, aggregation state, purity, surface functionalization, dose and route of administration.^17,18 Although functionalization can improve dispersion, biodistribution and biocompatibility, concerns regarding tissue persistence, incomplete clearance and long-term inflammatory responses remain.¹⁹ Our group previously developed phospholipid-functionalized CNTs that were shown to modulate cardiac sympathetic activity and reduce susceptibility to malignant arrhythmias, while demonstrating favorable four-week in vivo biosafety without evident biochemical abnormalities or histopathological injury in major organs.²⁰ In addition to their neuromodulatory properties, CNTs possess excellent photothermal conversion efficiency, which is advantageous for photothermal therapy, an emerging technique with the potential for remote neuronal manipulation.^14,21 In photothermal neuromodulation, NIR laser irradiation induces local hyperthermia in the presence of photothermal materials, such as CNTs, activating or inhibiting neuronal firing.^22,23 The inhibitory effect may be attributable to the activation of heat-sensitive potassium channels by moderate hyperthermia, leading to potassium efflux, membrane hyperpolarization and suppression of neuronal excitability. The temperature-dependent response of neurons is critical in determining whether photothermal stimulation results in neuronal activation or inhibition.^13,24 Recently, platinum nanoshell-mediated photothermal therapy was reported to achieve multimodal autonomic neuromodulation, including neural activation and inhibition at different temperature thresholds, conferring significant protection against post-MI VAs in canines, suggesting the transformative potential of photothermal neuromodulation in cardiac electrophysiology.²³ Our previous study also demonstrated that gold nanoparticle-mediated photothermal neuromodulation could suppress stellate ganglion activity and reduce malignant arrhythmias.¹³ Compared with conventional noble metal-based photothermal platforms, CNTs may exhibit broad NIR absorption, favorable thermal stability and inherent neuromodulatory potential, providing a rationale for exploring PC-CNTs as a multifunctional platform for cardiac neuromodulation.^16,18

Intriguingly, the therapeutic potential of photothermal therapy may extend beyond direct neural modulation. The LSG is anatomically closely associated with peri-ganglionic adipose tissue, suggesting that this local neuro-adipose microenvironment may represent a potential interface involved in sympathetic regulation.⁸ Recent studies have suggested that localized hyperthermia can induce white adipose tissue browning in a cell-autonomous manner, independent of sympathetic activation, which may exert cardioprotective effects and influence cardiac remodeling and arrhythmia susceptibility.^14,25 Given the anatomical proximity of the LSG to surrounding adipose tissue, this phenomenon may represent another potential interface influenced by photothermal treatment. Thus, the photothermal effect of CNTs may affect both neuronal activity and the surrounding peri-ganglionic environment, potentially contributing to enhanced cardiac neuromodulation. Building upon this converging evidence, we hypothesized that combining the neuromodulatory properties of CNTs with NIR-mediated photothermal therapy would enhance the suppression of LSG hyperactivity through additional modulation of neuronal activity and potential alterations of the peri-ganglionic microenvironment.

In the present study, we synthesized 1,2-diphytanoyl-sn-glycero-3-phosphocholine (DPhPC)-modified CNTs (PC-CNTs). DPhPC is an amphiphilic phospholipid that interacts with CNT surfaces through hydrophobic interactions, while its hydrophilic headgroups provide improved aqueous dispersibility and biocompatibility.²⁶ We evaluated their therapeutic potential for suppressing LSG activity alone or in combination with NIR laser irradiation in a canine MI model. Our findings demonstrated PC-CNTs suppressed cardiac sympathetic hyperactivity and reduced VA vulnerability, while NIR irradiation further enhanced these therapeutic effects. These enhanced effects may be associated with attenuation of inflammatory signaling in the LSG and remodeling of peri-ganglionic adipose tissue (Scheme 1). This study presents a potential neuromodulatory strategy for protecting against post-MI VAs, representing a novel application of CNTs in cardiac electrophysiology.

Scheme 1.

Schematic overview showing the protective effects of PC-CNTs and NIR photothermal therapy on VAs via neuromodulation. PC-CNTs combined with NIR treatment exert enhanced anti-arrhythmic effects associated with suppression of LSG hyperactivation, altered neuroinflammatory signaling and browning-related changes in peri-ganglionic adipose tissue. Downward arrows indicate suppression or reduction, whereas upward arrows indicate enhancement or increase.

Materials and Methods

Preparation and Characterization of PC-CNTs

Short carboxylated multi-walled carbon nanotubes (MWCNTs; CAS: 308068–56-6; length: 0.5–2 µm; diameter: <8 nm; purity: >95%) and the lipid 1,2-diphytanoyl-sn-glycero-3-phosphocholine (DPhPC; CAS: 207131–40-6) were sourced from Aladdin Reagent Database, Inc. (Shanghai, China). As described previously,²⁰ PC-CNTs were fabricated by dispersing MWCNTs in a DPhPC-rich aqueous environment via intensive bath sonication. The morphology and size of both CNTs and PC-CNTs were characterized by transmission electron microscopy (TEM; JEM-2010, JEOL, Japan). Additionally, nanoparticle tracking analysis (NTA) was employed to determine the hydrodynamic size and zeta potential of the resulting PC-CNTs.

Measurement of the Photothermal Effect

To evaluate the influence of PC-CNTs concentration on the photothermal effect in vitro, different concentrations of PC-CNTs (15, 30, 60, and 90 µg mL⁻¹) were irradiated with an 808 nm NIR laser (Changchun New Industries Optoelectronics Technology Co., Ltd., Changchun, China) at 1.0 W cm⁻² for 10 min. Next, the impact of power density on the photothermal effect was examined by irradiating the same volume of PC-CNTs (30 µg mL⁻¹) with different power densities for 10 min. A digital thermometer (1319A-K type, TES, Taiwan) was used to measure the temperature of the solutions. For in vivo experiments, 0.1 mL of 30 µg mL⁻¹ PC-CNTs was microinjected into the LSG, which was then irradiated with the NIR laser for 5 min. A thermal imaging camera (FLIR C2, USA) was used to capture the thermal images.

Cell Culture and in vitro Cytotoxicity Assay

HT22 hippocampal cells (Procell Life Science, Wuhan, China), a representative neuronal cell line, were maintained in specialized media at 37 °C in a 5% CO₂ atmosphere for in vitro biocompatibility assessment. To assess the photothermal cytotoxicity of PC-CNTs, a CCK-8 assay was performed by seeding cells in 96-well plates (1 × 10⁴ cells/well) and allowing them to adhere for 24 h. After a 12-h incubation with PC-CNTs (15–240 µg mL⁻¹), the cells were irradiated with an 808 nm laser (1.0 W cm⁻²) for 5 min. The cells were then incubated for a further 12 h, after which CCK-8 reagent was added and the cells incubated for approximately 1 h. Absorbance at 450 nm was measured using a microplate reader.

Animal Preparation and Study Protocol

Eighteen adult male beagles (8–12 kg) were obtained from the Experimental Animal Center of Wuhan University Medical College. All animal experiments were approved by the Institutional Animal Care and Use Committee of Renmin Hospital of Wuhan University (Approval No. WDRM20231203A) and conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All beagles were anesthetized with sodium pentobarbital (30 mg kg⁻¹, i.v.) and maintained at 60 mg h⁻¹ throughout the experimental procedure. Animals were ventilated with room air using a positive-pressure respirator (MAO01746, Harvard Apparatus, Holliston, MA, USA). Normal saline was infused intravenously at 100 mL h⁻¹ to compensate for fluid loss. Surface electrocardiography (ECG) was continuously recorded using a computer-based laboratory system (Lead 7000, Jinjiang, Chengdu, China). Core body temperature was maintained at 36.5 ± 0.5 °C using a heating pad.

MI was induced via left anterior descending artery (LAD) occlusion.⁹ The beagles were randomly allocated to three groups using a random-number table: the phosphate-buffered saline (PBS) group (n = 6, LAD occlusion + PBS microinjection without NIR), the PC-CNTs group (n = 6, LAD occlusion + PC-CNTs microinjection without NIR) and the PC-CNTs + NIR group (n = 6, LAD occlusion + PC-CNTs microinjection with NIR). The surgical procedures were not blinded because of the distinct interventions, whereas data analysis was conducted in a blinded manner. A total volume of 0.1 mL of PBS or PC-CNTs suspension (30 µg mL⁻¹) was slowly microinjected into the LSG using a microinjection pump. The total volume was divided among four different injection sites to facilitate more homogeneous distribution within the ganglion. After injection at each site, the needle was kept stationary for 5 min before withdrawal to prevent backflow of the agents. In the PC-CNTs + NIR group, the LSG surface was then subjected to NIR laser irradiation at 1 W cm⁻² with a spot area of 1.0 cm² for 5 min. At the end of the study, animals were euthanized by intravenous overdose of sodium pentobarbital (150 mg kg⁻¹) under deep anesthesia, in accordance with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals (2020). Death was confirmed prior to tissue collection. All efforts were made to minimize animal suffering.

Heart Rate Variability (HRV) Analysis

HRV analysis was performed using commercially available LabChart 8.0 software (AD Instruments, New South Wales, Australia). For each analysis, a continuous 5-min ECG segment with minimal recording artifacts and stable ECG signals was selected. The analysis focused on frequency-domain parameters, including low frequency (LF, 0.04–0.15 Hz), high frequency (HF, 0.15–0.4 Hz) and the LF/HF ratio. LF and HF components were expressed in normalized units to facilitate comparative analysis, and the LF/HF ratio was calculated accordingly.

Ventricular Effective Refractory Period (ERP) Measurement

The heart was exposed through a left fourth-intercostal thoracotomy, and multielectrode catheters were secured to the left ventricular (LV) epicardium at the LV apex, LV base and mid-LV. ERPs at these sites were determined using programmed electrical stimulation at twice the diastolic threshold, defined as the minimum voltage required to capture the ventricle. Eight consecutive stimuli (S1–S1 = 330 ms cycle length) were delivered, followed by a premature stimulus (S2). The initial S1–S2 interval was set at 250 ms, shortened in 10-ms decrements, and subsequently refined to 2-ms steps upon approaching the refractory period, with the ERP defined as the longest S1–S2 interval that failed to achieve ventricular capture.

Measurement of LSG Neural Activity

Two tungsten-coated microelectrodes were inserted into the ganglion tissue, and a ground lead was connected to the chest wall to record LSG neural activity. Raw neural signals were continuously recorded for approximately 15 min using a PowerLab data acquisition system (8/35, ADInstruments, Australia) and amplified using an amplifier (DP-304, Warner Instruments, Hamden, CT, USA), with bandpass filters set at 300 Hz to 1 kHz and an amplification range of 30–50 times. After signal stabilization, a continuous 1-min segment from the stable recording period was selected for quantitative analysis using LabChart software. Valid neural firing was defined as deflections exhibiting a signal-to-noise ratio greater than 3:1, as described previously.^8,12

Assessment of Ventricular Arrhythmias

After MI induction, continuous surface ECG monitoring was performed for 1 hour to evaluate spontaneous VAs, as the first hour following coronary artery occlusion represents the acute arrhythmogenic phase of MI. VAs were classified as ventricular premature beats (VPBs), non-sustained ventricular tachycardia (VT), sustained VT and ventricular fibrillation (VF). Furthermore, the VF threshold (VFT) was tested via burst pacing (20 Hz, 0.1 ms duration, 10 s bursts) delivered to the right ventricular apex. The pacing voltage started at 2 V and was incrementally increased by 2 V, with the lowest voltage capable of inducing sustained VF recorded as the VF threshold.²⁷

Transcriptome Analysis

LSG tissues were rapidly harvested and snap-frozen in liquid nitrogen, after which total RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA, USA) and purified via the RNeasy mini kit. Following reverse transcription into cDNA, sequencing was performed on the BGSEQ-500 platform (BGI Shenzhen, China). Raw reads were filtered using SOAPnuke (v1.5.6) to remove adapter-contaminated reads, reads containing more than 5% unknown bases and low-quality reads in which bases with a quality score below 15 accounted for more than 20% of the total read length. Clean reads were aligned to the canine reference genome (Canis familiaris, Dog10K_Boxer_Tasha) using HISAT2 (v2.1.0) and to the reference gene set using Bowtie2 (v2.3.4.3). Gene expression levels were quantified using RSEM (v1.3.1), and differential expression analysis was performed using DESeq2 (v1.4.5), with its default independent filtering procedure applied to low-expression genes. P-values were adjusted using the false-discovery rate method, and genes with a Q value ≤ 0.05 were considered differentially expressed. GO and KEGG enrichment analyses were performed using the Phyper function in R based on the hypergeometric test, and enriched terms with a Q value ≤ 0.05 were considered statistically significant.¹⁴

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Total RNA from the adipose tissue surrounding the LSG was isolated using TRIzol reagent, quantified spectrophotometrically at 260 nm and reverse-transcribed into complementary DNA using a First Strand cDNA Synthesis Kit (Servicebio, Wuhan, China). Subsequent qRT-PCR was conducted on a 7500 Real-Time PCR System (Applied Biosystems, USA) utilizing a 2× Universal Blue SYBR Green qPCR Master Mix. Relative mRNA expression levels of target genes were determined using the 2^−∆∆Ct method after normalization against GAPDH expression, with all experimental procedures repeated in triplicate. Primer sequences used in this analysis are listed in the Supplementary Materials (Supplementary Table S1).

Histological Analysis

The LSG and peri-ganglionic adipose tissue were excised, fixed in 4% paraformaldehyde, paraffin-embedded and sectioned at a 5-μm thickness. Both tissues were subjected to H&E staining. The LSG was additionally subjected to double immunofluorescence staining with anti-TH (Abcam, Cambridge, United Kingdom) and anti-c-fos (Abcam) antibodies. The adipose tissue was subjected to immunofluorescence staining using anti-uncoupling protein 1 (UCP1) antibodies (Abcam) with DAPI nuclear counterstaining. All histological images were quantitatively analyzed using ImageJ software.

Statistical Analysis

The normality of all continuous variables was assessed using the Shapiro–Wilk test prior to statistical analysis. All continuous variables satisfied the assumption of normality and were presented as the mean ± standard error of the mean (SEM). Statistical comparisons between two independent groups were performed using an unpaired Student’s t-test, whereas comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) followed by Bonferroni’s post hoc test. Repeated measurements of VPBs over time were analyzed using two-way repeated-measures ANOVA followed by Bonferroni’s post hoc test. Data analysis and graphing were performed using GraphPad Prism 9.0 software, with differences considered statistically significant at P < 0.05.

Results and Discussion

Characterization and Photothermal Properties of PC-CNTs

Phospholipid-coated CNTs (PC-CNTs) were successfully synthesized using a bath-sonication method. TEM imaging revealed the morphology of both pristine CNTs and PC-CNTs, with PC-CNTs exhibiting a particle size of approximately 100–200 nm (Figure 1A). NTA analysis showed that PC-CNTs possessed a zeta potential of −11.4 mV and a median hydrodynamic diameter of 167.5 nm (Figure 1B). Because 808 nm light exhibits low water absorption and favorable tissue penetration, it was selected for subsequent photothermal experiments. CNTs have previously been shown to possess excellent photothermal conversion efficiency at this wavelength.²⁸ To investigate the photothermal effect of PC-CNTs in vitro, aqueous suspensions of PC-CNTs were irradiated with an 808 nm NIR laser at 1.0 W cm⁻² for 10 min, with temperature monitored every 30 s. Upon NIR laser irradiation at 1 W cm⁻², PC-CNTs suspensions exhibited a concentration-dependent increase in temperature at concentrations of 15, 30, 60, and 90 µg mL⁻¹, whereas the PBS control showed negligible temperature changes under the same conditions (Figure 1C). Representative infrared thermal images of PC-CNTs suspensions at different concentrations are presented in Figure 1D. Notably, PC-CNTs at a concentration of 30 µg mL⁻¹ exhibited a temperature increase within the range previously reported for photothermal neuromodulation, as the temperature exceeded 40 °C within 5 min, a range previously reported to suppress neuronal activity.²⁴ Furthermore, higher laser power densities resulted in greater temperature elevations in PC-CNTs suspensions, demonstrating a power density-dependent photothermal effect (Figure 1E). No appreciable decline in photothermal performance was observed during four consecutive heating–cooling cycles, indicating excellent photothermal stability of PC-CNTs (Figure 1F). These results demonstrated that PC-CNTs possessed robust and stable photothermal properties suitable for subsequent neuromodulation studies.

Figure 1.

Characterization and properties of PC-CNTs. (A) TEM images of CNTs and PC-CNTs. (B) Zeta potential and particle size distribution of PC-CNTs. (C) Temperature changes of PC-CNTs at various concentrations under 808-nm NIR laser irradiation at 1 W cm⁻² for 10 min. (D) Representative thermal images of PC-CNTs at different concentrations under 808-nm NIR laser irradiation at 1 W cm⁻² for 10 min. (E) Temperature changes of PC-CNTs at 30 µg mL⁻¹ with different irradiation power densities for 10 min. (F) Photothermal conversion curves of PC-CNTs (30 µg mL⁻¹) under four cycles of irradiation (1 W cm⁻²).

Biocompatibility of PC-CNTs in vitro and in vivo

Although the biocompatibility of PC-CNTs has been demonstrated previously,²⁰ the potential cytotoxicity associated with combined PC-CNTs and NIR treatment had not been evaluated. HT22 cells were incubated with PC-CNTs (0–240 µg mL⁻¹) for 24 h, followed by 5 min of 808 nm NIR laser irradiation (1.0 W cm⁻²) or a sham irradiation. As shown in Figure 2A, no significant difference in cell viability was observed between the irradiated and sham-treated groups at PC-CNTs concentrations ≤ 60 μg mL⁻¹ (P > 0.05). In contrast, significantly reduced cell viability was observed at concentrations exceeding 60 μg mL⁻¹ (P < 0.05). These results demonstrated that PC-CNTs exhibited favorable cytocompatibility and that short-term NIR irradiation produced negligible cytotoxic effects under the selected experimental conditions. Together, these preliminary photothermal characterization and biocompatibility results provided a rationale for selecting the NIR irradiation parameters (808 nm, 1.0 W cm⁻², 5 min) for subsequent in vivo experiments.

Figure 2.

In vitro photothermal cytotoxicity and preliminary in vivo biosafety assessment. (A) CCK-8 cell viability assay of HT22 cells incubated with various concentrations of PC-CNTs for 24 h, followed by NIR irradiation (1 W cm⁻², 5 min) or sham irradiation. (B) Representative anatomical location of the canine LSG and thermographic images of the LSG tissue treated with PC-CNTs (30 µg mL⁻¹, 0.1 mL) and NIR laser irradiation (808 nm, 1.0 W cm⁻²). (C) Study protocol in vivo. (D) Representative images of H&E staining of LSG tissues from the three groups. (E) Quantitative analysis of LSG neuron diameter among three groups. Data are shown as the mean ± SEM (n = 6). **P < 0.01 and ***P < 0.001 for one-way ANOVA test.

Abbreviations: LSG, left stellate ganglion; LAD, left anterior descending artery; HRV, heart rate variability; ERP, effective refractory period; VAs, ventricular arrhythmias.

To examine the neuromodulatory and photothermal effects of PC-CNTs in vivo, a canine MI model was established. Given that our previous research demonstrated the efficacy of PC-CNTs at 30 μg mL⁻¹ in inhibiting neural activity,²⁰ and that PC-CNTs at this concentration achieved the temperature range required for effective photothermal neuromodulation under NIR irradiation without inducing cytotoxicity, a concentration of 30 μg mL⁻¹ was selected for in vivo experiments. Figure 2B shows the representative anatomical location of the LSG and surrounding adipose tissues, which were administered with PC-CNTs and subjected to 808 nm NIR laser irradiation. The local temperature of LSG injected with PC-CNTs (0.1 mL, 30 µg mL⁻¹) increased from 35.6 °C to 43.6 °C within 5 min NIR irradiation, confirming effective local photothermal conversion. Figure 2C shows the beagle study flow diagram in vivo. Furthermore, H&E staining of the LSG across all groups revealed that neurons within the tissue maintained well-preserved morphological integrity with typical neuronal characteristics, including plump cell bodies, clearly visible nuclei and prominent nucleoli (Figure 2D). No significant difference in LSG neuron diameter was observed (Figure 2E), indicating that the combination of PC-CNTs and NIR irradiation exerted no apparent toxicity on the LSG in vivo.

Effect of PC-CNTs and NIR Photothermal Therapy on Cardiac Autonomic Function

HRV analysis was performed to assess cardiac autonomic function noninvasively. In frequency-domain analysis, the HF component reflects parasympathetic (vagal) activity, whereas the LF component and LF/HF ratio are commonly used as indices of sympathetic activity.⁸ ECG recordings and HRV analyses were conducted both at baseline and after the respective interventions. As shown in Figure 3A–C, no significant differences were observed in LF, HF or the LF/HF ratio among the three groups at baseline (all P > 0.05), indicating comparable autonomic status before treatment. Following the interventions, significant shifts in autonomic balance were observed (Figure 3D–F). Compared with the PBS group, PC-CNTs led to a significant decrease in LF (18.92 ± 1.44 nu vs 29.82 ± 1.19 nu, P < 0.001) and the LF/HF ratio (0.33 ± 0.04 vs 0.57 ± 0.03, P < 0.001), accompanied by a marked increase in HF (65.93 ± 3.23 nu vs 53.29 ± 2.74 nu, P < 0.05). These findings suggest that PC-CNTs treatment effectively suppressed sympathetic activity while enhancing parasympathetic tone. Furthermore, the combination of NIR irradiation significantly enhanced the autonomic modulatory effects of PC-CNTs. Compared with the PC-CNTs group, PC-CNTs + NIR group exhibited a further reduction in LF (12.77 ± 0.79 nu vs 18.92 ± 1.44 nu, P < 0.01) and the LF/HF ratio (0.17 ± 0.01 vs 0.33 ± 0.04, P < 0.01), alongside an additional increase in HF power (76.24 ± 2.18 nu vs 65.93 ± 3.23 nu, P < 0.05). These results suggest that NIR irradiation further enhanced the ability of PC-CNTs to restore autonomic balance, favoring a parasympathetic-dominant state.

Figure 3.

PC-CNTs alone and in combination with NIR photothermal therapy improved cardiac autonomic function assessed by HRV. (A–C) Baseline HRV parameters among three groups. (D–F) HRV parameters after the respective treatments among three groups. Data are shown as the mean ± SEM (n = 6). ^#P < 0.05, ^###P < 0.001 vs PBS group and *P < 0.05, **P < 0.01 vs PC-CNTs group for one-way ANOVA test.

Abbreviations: HRV, heart rate variability; LF, low frequency; HF, high frequency; LF/HF, the ratio of LF to HF.

Effect of PC-CNTs and NIR Photothermal Therapy on Ventricular Electrophysiological Stability

To evaluate the impact of PC-CNTs treatment alone and in combination with NIR irradiation on ventricular electrophysiological properties, ERP was measured at three sites of the left ventricle (LVB, LVM and LVA).²⁹ The schematic representation of electrode placement and representative traces of the ERP measurement procedure are shown in Figure 4A, where the ERP was defined as the longest S1–S2 interval that failed to capture the ventricle. At baseline, no significant differences in ERP were observed across the three groups at any of the measured sites (all P > 0.05; Figure 4B–D), indicating similar baseline ventricular electrophysiological properties among the three groups. Following the respective treatments, ERP values underwent significant changes (Figure 4E–G). Compared with the PBS group, PC-CNTs treatment alone significantly prolonged the ERP across all measured regions (LVB: 179.7 ± 2.3 ms vs 165.0 ± 2.8 ms; LVM: 178.7 ± 3.4 ms vs 162.3 ± 2.3 ms; LVA: 177.0 ± 2.6 ms vs 163.3 ± 3.7 ms; all P < 0.05), suggesting that PC-CNTs treatment improved ventricular electrophysiological stability. Notably, NIR irradiation further enhanced the electrophysiological benefits conferred by PC-CNTs treatment. Compared with the PC-CNTs group, the PC-CNTs + NIR group exhibited a more pronounced increase in ERP at the LVB (195.0 ± 4.1 ms vs 179.7 ± 2.3 ms, P < 0.01), LVM (196.0 ± 4.9 ms vs 178.7 ± 3.4 ms, P < 0.05) and LVA (196.0 ± 4.0 ms vs 177.0 ± 2.6 ms, P < 0.01). Taken together, these findings demonstrated that while PC-CNTs improved electrophysiological stability, the addition of NIR irradiation further enhanced this electrophysiological benefit.

Figure 4.

PC-CNTs alone and in combination with NIR photothermal therapy enhanced ventricular electrophysiological stability. (A) Schematic representation of the electrode placement in the left ventricle and representative traces of the ERP measurement. (B–D) Baseline ERPs at the LVB, LVM and LVA sites among three groups. (E–G) ERPs at the LVB, LVM and LVA sites after the respective treatments among three groups. Data are shown as the mean ± SEM (n = 6). ^#P < 0.05 vs PBS group and *P < 0.05, **P < 0.01 vs PC-CNTs group for one-way ANOVA test.

Abbreviations: ERP, effective refractory period; LVB, left ventricular base; LVM, left ventricular mid-region; LVA, left ventricular apex.

Effect of PC-CNTs and NIR Photothermal Therapy on Post-MI LSG Neural Hyperactivation and VAs

MI is known to induce significant LSG hyperactivation, characterized by increased neuronal firing frequency and amplitude.³⁰ Here, we evaluated the modulatory effects of PC-CNTs alone and in combination with NIR irradiation on post-MI LSG activity. Following LAD occlusion to induce MI, representative ECG tracings demonstrating the occurrence of acute ischemic changes are shown in Supplementary Figure S1. Representative LSG neuronal firing traces at baseline and post-MI for each group are presented in Figure 5A. There were no significant differences in the frequency or amplitude of LSG neuronal firing among the groups at baseline (Figure 5B and C). Compared with the PBS group, PC-CNTs treatment significantly inhibited the MI-induced excessive discharge of LSG neurons, with substantial reductions in both firing frequency (109.57 ± 4.2 impulses/min vs 130.7 ± 2.2 impulses/min, P < 0.01) and amplitude (0.086 ± 0.009 mV vs 0.145 ± 0.014 mV, P < 0.05). Notably, the combined application of PC-CNTs and NIR irradiation further enhanced this inhibitory effect, resulting in significantly lower frequency (50.33 ± 3.5 impulses/min vs 109.57 ± 4.2 impulses/min, P < 0.001) and amplitude (0.054 ± 0.003 mV vs 0.086 ± 0.009 mV, P < 0.05) compared with the PC-CNTs group (Figure 5D and E). These findings indicated that while PC-CNTs effectively suppressed post-MI LSG hyperactivation, the addition of NIR irradiation provided an enhanced neuromodulatory effect.

Figure 5.

PC-CNTs alone and in combination with NIR photothermal therapy suppressed post-MI LSG hyperactivation and reduced ventricular arrhythmia occurrence. (A) Representative traces of LSG neuronal firing recorded at baseline and post-MI in the three groups. (B and C) Quantitative analysis of LSG firing frequency (B) and amplitude (C) at baseline. (D and E) Quantitative analysis of LSG firing frequency (D) and amplitude (E) post-MI. (F) Representative ECG traces showing VPBs, VT and VF after LAD occlusion. (G) Time-course analysis of VPB numbers recorded every 15 min during the 1-h period following MI. (H) Comparison of VFT among the different groups after treatment. Data are presented as the mean ± SEM (n = 6). ^#P < 0.05, ^##P < 0.01 vs PBS group; *P < 0.05, **P < 0.01, ***P < 0.001 vs PC-CNTs group, analyzed by one-way or two-way ANOVA.

Abbreviations: LSG, left stellate ganglion; MI, myocardial infarction; LAD, left anterior descending artery; VPB, ventricular premature beat; VT, ventricular tachycardia; VF, ventricular fibrillation; VFT, ventricular fibrillation threshold. ns, not significant.

We subsequently monitored VA incidence during the 1-hour post-MI period. Representative ECG traces illustrating VPB, VT and VF are shown in Figure 5F. As demonstrated in the time-course analysis, the frequency of VPBs was significantly reduced in the PC-CNTs-treated group compared with the PBS group across all recorded intervals. This anti-arrhythmic effect was further enhanced by the combined NIR intervention, which resulted in a markedly lower total VPB count (Figure 5G). Similarly, the number of non-sustained VT episodes was significantly reduced following PC-CNTs treatment and was further decreased by the addition of NIR irradiation. The incidence of sustained VT/VF showed a decreasing trend among the three groups (Supplementary Table S2). Furthermore, PC-CNTs treatment significantly increased the VFT (34.33 ± 2.09 V vs 25.33 ± 1.91 V, P < 0.05), and this stabilizing effect on ventricular electrophysiology was further augmented by the addition of NIR irradiation, which led to a further increase in the VFT (44.67 ± 1.61 V vs 34.33 ± 2.09 V, P < 0.01) (Figure 5H). These results suggest that PC-CNTs treatment reduced post-MI VA susceptibility, whereas NIR irradiation further enhanced these anti-arrhythmic effects.

Effects of PC-CNTs and NIR Photothermal Therapy on Transcriptional Expression Profiles in the LSG

To delineate the potential molecular mechanisms underlying the cardioprotective effects of PC-CNTs and NIR irradiation, RNA sequencing (RNA-seq) was performed on LSG tissue. Inter-sample reproducibility was confirmed by Pearson correlation analysis (all coefficients >0.90; Supplementary Figure S2), and principal component analysis (PCA) revealed distinct clustering of the PBS, PC-CNTs and PC-CNTs + NIR groups (Supplementary Figure S3). DEGs were defined as absolute fold change >1.5 with an adjusted P-value ≤ 0.05. Volcano plots and Venn diagram analyses were presented in Figure 6A and B. A total of 647 DEGs were identified between the PC-CNTs and PBS groups (57 upregulated, 590 downregulated), and 300 DEGs were identified between the PC-CNTs + NIR and PC-CNTs groups (93 upregulated, 207 downregulated). Of these, 60 overlapping DEGs were identified between both comparisons (Figure 6B). Heat map analysis of shared downregulated genes revealed that 55 were suppressed by PC-CNTs alone and underwent further downregulation following NIR co-treatment (Figure 6C). GO enrichment analysis identified “cytokine activity” among the most enriched molecular function terms, with representative genes including CXCL14, CSF3, TIMP1 and CRLF1 (Figure 6D). Consistently, KEGG pathway analysis highlighted “cytokine-cytokine receptor interaction” as the most significantly enriched pathway (Figure S4). Quantitative analysis of the transcript levels (TPM) confirmed that PC-CNTs treatment significantly suppressed the expression of several inflammation-related genes compared with the PBS group (CXCL14: 108.2 ± 9.6 vs 299.0 ± 13.1, P < 0.001; CSF3: 38.4 ± 7.2 vs 66.0 ± 5.6, P < 0.001; TIMP1: 2740 ± 708 vs 7515 ± 608, P < 0.01; CRLF1: 6.8 ± 0.4 vs 16.9 ± 2.2, P < 0.01). Importantly, NIR irradiation further enhanced the suppression of these inflammation-related genes. Compared with the PC-CNTs group, the PC-CNTs + NIR group exhibited significantly lower transcript levels (CXCL14: 46.0 ± 17.8 vs 108.2 ± 9.6, P < 0.05; CSF3: 8.3 ± 0.9 vs 38.4 ± 7.2, P < 0.05; TIMP1: 790 ± 67 vs 2740 ± 708, P < 0.05; CRLF1: 2.1 ± 0.3 vs 6.8 ± 0.4, P < 0.05) (Figure 6E–H).

Figure 6.

PC-CNTs alone and in combination with NIR photothermal therapy altered neuroinflammation-related transcriptional profiles in the LSG post-MI. (A) Volcano plot of DEGs among PBS, PC-CNTs and PC-CNTs + NIR groups. (B) Venn map of DEGs among three groups. (C) Heatmap of shared downregulated DEGs among three groups. (D) GO enrichment analysis of the DEGs, showing the top enriched terms in BP, CC and MF categories. (E–H) Quantitative analysis of mRNA expression levels (TPM) for key inflammation-related genes: CXCL14 (E), CSF3 (F), TIMP1 (G), and CRLF1 (H) in the LSG. Data are shown as the mean ± SEM (n = 3). ^##P < 0.01, ^###P < 0.001 vs PBS group and *P < 0.05 vs PC-CNTs group for one-way ANOVA test.

Abbreviations: LSG, left stellate ganglion; MI, myocardial infarction; DEG, differentially expressed genes; TPM, transcripts per million; GO, Gene Ontology; BP, biological process; CC, cellular component; MF, molecular function. CXCL14, C-X-C motif chemokine ligand 14; CSF3, colony-stimulating factor 3; TIMP1, tissue inhibitor of metalloproteinases 1; CRLF1, cytokine receptor-like factor 1.

These findings suggest that the enhanced anti-arrhythmic effects of PC-CNTs and NIR irradiation were associated with altered inflammatory signaling within the LSG microenvironment. Neuroinflammatory processes have been implicated in the pathological progression of post-MI VAs. Beyond direct neuronal mechanisms, inflammatory infiltration of the LSG by macrophages, neutrophils and lymphocytes, together with the release of pro-inflammatory cytokines, may contribute to sustained sympathetic hyperactivity and arrhythmogenesis.^31,32 The inflammation-related molecules identified in the present study are predominantly involved in leukocyte recruitment, inflammatory amplification and immune-cell activation, all of which may contribute to neuroinflammatory remodeling of the stellate ganglion. CXCL14, a constitutively expressed CXC-family chemokine, has been implicated in inflammatory cell recruitment and, upon upregulation, facilitates inflammatory responses and cellular apoptosis.³³ CSF3 is a granulocyte colony-stimulating factor whose elevated expression in neuroinflammatory contexts may contribute to cardiovascular disease progression through immune cell recruitment and cytokine cascade amplification.³⁴ TIMP1, classically recognized as a matrix metalloproteinase inhibitor, also functions as a cytokine-like signaling molecule that drives pro-inflammatory responses and may participate in neuroinflammatory responses and extracellular matrix remodeling associated with arrhythmogenic substrates.³⁵ CRLF1, a cytokine receptor-like factor regulated by inflammatory cytokines such as TNF-α and IL-6, participates in the regulation of the immune response during systemic inflammation.³⁶ A recent study demonstrated that LSG ablation in canines following acute MI prevented M1 macrophage polarization and inflammatory cytokine release, thereby reducing VA vulnerability.⁹ Collectively, the altered expression of these inflammation-related genes suggests that PC-CNTs combined with NIR photothermal therapy may influence the inflammatory microenvironment surrounding the LSG, which could be associated with reduced sympathetic hyperactivity and arrhythmogenic susceptibility after MI. However, these transcriptomic findings represent associative molecular changes, and the specific functional contributions of individual mediators, such as CXCL14, CSF3, TIMP1 and CRLF1, require further validation using targeted approaches. The enhanced anti-inflammatory effects observed with combined PC-CNTs and NIR photothermal therapy relative to PC-CNTs alone may reflect contributions from multiple mechanisms. Well-characterized interactions between functionalized CNTs and cellular membranes can directly modulate neuronal excitability and may influence basal inflammatory signaling in ganglionic tissue, consistent with prior reports.²⁰ Furthermore, the photothermal effect elicited by NIR irradiation may influence macrophage polarization toward an anti-inflammatory phenotype,³⁷ thereby potentially reducing the primary cellular source of pro-inflammatory cytokines within the LSG. These findings provide a potential framework for understanding the enhanced modulation of inflammatory signaling observed in the PC-CNTs + NIR group. Together with the observed reduction in LSG neural activity and ventricular arrhythmia susceptibility, these findings support the potential involvement of neuroimmune remodeling in the neuromodulatory effects of PC-CNTs combined with NIR photothermal therapy.

Effect of NIR Photothermal Therapy on Peri-Ganglionic Adipose Tissue Browning

Anatomically, the LSG is surrounded by adipose tissue in vivo.¹³ Recent studies have reported that local hyperthermia can induce white adipose tissue browning and increase thermogenesis independent of sympathetic activation,^25,38,39 which promotes the upregulation of the thermogenic protein UCP1, alleviates oxidative stress, facilitates lipid catabolism and improves metabolism.^40,41 Adipose tissue browning has been shown to exert protective effects on cardiac function and arrhythmia susceptibility.^42,43 To investigate the impact of photothermal therapy on thermogenesis and browning-related gene expression, adipose tissue surrounding the LSG was subjected to molecular analysis. qRT-PCR revealed that NIR laser irradiation significantly upregulated the mRNA expression levels of browning-related genes, including UCP1 and PGC1α (markers of thermogenesis and mitochondrial biogenesis), as well as CIDEA, ELOVL3, CytC and ATPase β (Figure 7A). However, the mRNA levels of these genes did not differ significantly between the PC-CNTs and PBS groups, indicating that PC-CNTs alone do not significantly influence peri-ganglionic adipose browning in the absence of photothermal stimulation. Immunofluorescence staining confirmed the upregulation of UCP1 protein expression in the peri-ganglionic adipose tissue of the PC-CNTs + NIR group (Figure 7B). H&E staining further revealed that photothermal therapy significantly decreased the average adipocyte size in this depot (Figure 7C and D), a hallmark morphological feature of adipose browning. To further assess the effects of PC-CNTs on sympathetic remodeling, LSG neurons were subjected to double immunofluorescence staining for c-fos and tyrosine hydroxylase (TH, a sympathetic nerve marker) (Figure 7E). Quantitative analysis indicated that PC-CNTs significantly downregulated the levels of c-fos in the LSG, which was further reduced upon combined NIR laser irradiation, thereby supporting the enhanced therapeutic efficacy of this combined strategy (Figure 7F).

Figure 7.

Photothermal effect of PC-CNTs and NIR irradiation was associated with browning-related changes in peri-ganglionic adipose tissue. (A) The mRNA levels of thermogenic and mitochondrial genes in adipose tissue surrounding the LSG among three groups. (B) Representative immunofluorescence staining of UCP1 (green) in the adipose tissue around the LSG. (C and D) Representative H&E staining images (C) and quantitative analysis of adipocyte size (D) in the adipose tissue surrounding the LSG. (E) Representative immunofluorescence staining of c-fos and TH in the LSG among three groups. (F) Quantitative analysis of c-fos expression in TH⁺ neurons of the LSG. Data are shown as the mean ± SEM (n = 6). ^##P < 0.01 vs PBS group and *P < 0.05, **P < 0.01, ***P < 0.001 vs PC-CNTs group for one-way ANOVA test.

Abbreviations: LSG, left stellate ganglion; UCP1, uncoupling protein 1; PGC1α, PPARG coactivator 1 alpha; CIDEA, cell death inducing DFFA like effector A; ELOVL3, ELOVL fatty acid elongase 3; CytC, cytochrome C; ATPase β, ATPase Na⁺/K⁺ transporting subunit beta 1; DAPI, 4,6-diamidino-2-phenylindole. TH, tyrosine hydroxylase.

These findings suggest that the photothermal effect, rather than PC-CNTs per se, was associated with the induction of peri-ganglionic adipose tissue browning. Of particular mechanistic interest, this thermogenic response was associated with local hyperthermia and may occur independently of direct adrenergic stimulation, a distinction that is especially relevant in the post-MI setting, where sympathetic overactivation is already predominant.⁴⁴ Indeed, the PC-CNTs + NIR strategy simultaneously attenuated LSG neural hyperactivity while promoting adipose tissue browning through a non-adrenergic mechanism, distinguishing it from classical cold-induced thermogenesis.⁴⁵ The anatomical proximity of the LSG and its surrounding adipose tissue suggests that NIR-induced browning may represent a local tissue response associated with the photothermal intervention, which may contribute to the enhanced anti-arrhythmic effects observed in the PC-CNTs + NIR group compared with PC-CNTs treatment alone. Beyond direct neuromodulation and modulation of inflammatory signaling, peri-ganglionic adipose tissue may represent an additional component of the local microenvironment influenced by photothermal treatment. Browning-associated adipokines and thermogenic factors have been reported to exert beneficial effects on oxidative stress, energy metabolism, and cardiovascular remodeling.^46,47 Therefore, modulation of the local adipose microenvironment may represent a potential biological response associated with photothermal therapy. Compared with our previous study, the present work extends the application of PC-CNTs-based neuromodulation by incorporating NIR photothermal therapy. This combined strategy was associated with enhanced suppression of LSG hyperactivity and preclinical anti-arrhythmic effects, while revealing potential contributions of altered inflammatory signaling within the LSG and peri-ganglionic adipose browning. Nevertheless, the precise contribution of these mechanisms to the anti-arrhythmic effects remains to be fully established and requires further investigation.

Limitations

This study has several limitations. First, this was an exploratory preclinical study with a relatively small sample size. The sample size was selected based on previous canine MI studies and the principles of the 3Rs (Replacement, Reduction and Refinement). Although significant differences were observed in the primary outcomes, larger studies are needed to further validate these findings. Second, although transcriptomic analysis identified altered inflammatory signaling within the LSG and NIR photothermal therapy was associated with browning-related changes in peri-ganglionic adipose tissue, whether these molecular changes directly contribute to the anti-arrhythmic effects remains unclear. Further studies involving comprehensive mechanistic validation and targeted modulation approaches are needed to clarify the functional relevance of these molecular changes. Third, the present study was conducted in an acute MI model, limiting the evaluation of long-term cardiac remodeling and function. Future chronic MI studies with extended follow-up are needed to evaluate the long-term cardioprotective effects of PC-CNTs combined with NIR photothermal therapy and further assess PC-CNTs retention, biodistribution, safety and translational potential. Finally, the present study relied on direct surgical microinjection into the stellate ganglion, which may limit future clinical translation. Minimally invasive, ultrasound-guided injection approaches may provide a more clinically feasible strategy for precise stellate ganglion delivery.

Conclusion

In conclusion, the present study demonstrated the beneficial effects of PC-CNTs combined with NIR photothermal therapy in preventing VAs following MI. PC-CNTs treatment alone improved cardiac autonomic balance, prolonged ventricular ERP, suppressed MI-induced LSG hyperactivity and reduced VA occurrence. Additionally, the combination of NIR irradiation further enhanced the protective effect of PC-CNTs, which may be associated with altered inflammatory signaling within the LSG and photothermal-induced browning of peri-ganglionic adipose tissue. This study provides new insights into the potential application of PC-CNTs combined with NIR photothermal therapy as an enhanced neuromodulatory strategy for preventing cardiac VAs and highlights a promising preclinical approach for autonomic neuromodulation following MI.

Funding Statement

This work was supported by grants from the National Natural Science Foundation of China (82570458 and 82400282), National Key Research and Development Program of China (2023YFC2705700) and National Science and Technology Major Project of China (2025ZD0548203).

Ethics Statement

All animal experiments were approved by the Institutional Animal Care and Use Committee of Renmin Hospital of Wuhan University (Approval No. WDRM20231203A) and conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals. At the end of the study, animals were euthanized by intravenous overdose of sodium pentobarbital under deep anesthesia in accordance with the AVMA Guidelines for the Euthanasia of Animals (2020).

Disclosure

The author(s) report no conflicts of interest in this work.