Effect of Near-Infrared Pre-Irradiation on Irreversible Electroporation Treatment of Rat Gastric Tissues

Abstract

Irreversible electroporation (IRE) is a recognized ablation technique that induces apoptosis via potent electric fields. Nonetheless, the heterogeneity of biological tissues often results in inconsistent treatment outcomes, leaving residual viable cells and leading to potential relapse. To address this, previous strategies incorporated chemical enhancers to IRE, but these faced limitations such as limited tissue diffusion and hyperpigmentation. In this study, we explore the synergistic application of near-infrared (NIR) irradiation with IRE. Using an in vivo rat gastric tissue model, we pre-irradiated samples with NIR at 3 J/cm² prior to IRE. The combined treatment, termed NIRE, produced a change in tissue impedance of 13.5 Ohm compared to IRE alone, indicating NIR’s potential in modulating tissue electrical properties. Subsequent histopathological and molecular assessments revealed a 1.12-fold increase in apoptosis for NIRE over IRE. Notably, the apoptosis-related proteins BCL and p21 exhibited a 1.24-fold and 1.29-fold overexpression following NIRE treatment, respectively, emphasizing NIRE’s enhanced apoptotic activation. In essence, our findings underscore the augmented therapeutic efficacy of IRE when complemented with NIR, presenting a promising avenue for bolstering treatment outcomes in tissue ablation.

1. Introduction

Irreversible electroporation (IRE) stands at the forefront of modern ablation techniques, driving apoptosis through the irreversible disruption of cell membrane integrity. Achieved through the delivery of short, high-intensity electrical pulses via electrodes to the targeted tissue, the interaction between the electric field and the cell is governed by the Maxwell equation. This interaction is inherently tethered to the cellular membrane potential, influenced by the diverse ions residing both within and outside the membrane. Consequently, due to variations in transmembrane ions, the effect of the electric field is not uniform across all cancer cells. Further complicating matters is the inherent heterogeneity of the target tissue, discerned through its cellular distribution, extracellular matrix, and vascular architecture [1]. If the field-cell interaction is insufficient, cancer cells may survive, posing risks of relapse [2]. To address this limitation, the integration of IRE with anticancer drugs birthed a novel therapeutic approach: electrochemotherapy (ECT). ECT’s primary function is to facilitate direct access for anticancer drugs to the cellular DNA and cytoplasm [3,4]. Additionally, ECT has proven adept at inducing death in endothelial cells of tumor-feeding vessels, culminating in tumor ischemia [5]. Clinically, ECT has already exhibited prowess against various malignancies, including melanoma, breast, renal cell, and basal cell carcinomas, as well as pancreatic cancer [6]. Nevertheless, ECT is not without its challenges, with limited drug dispersion in target tissues due to heterogeneity [7], sometimes resulting in skin hyperpigmentation [8,9].

A recent avenue of exploration involves the union of calcium ions with IRE, projecting this as a prospective anticancer modality [10,11,12,13]. In this technique, the IRE platform introduces high concentrations of calcium ions directly into the cell’s cytoplasm. This surge in calcium concentration gives rise to calcium ion-adenosine triphosphatase (Ca²⁺-ATPases) formations, escalating ATP consumption for the extrusion of calcium ions from the cell [14]. A rise in intracellular calcium also jeopardizes the mitochondrial membrane potential, causing dwindling ATP production [15]. Further consequences include the generation of reactive oxygen species (ROS) and the activation of lipases and proteases [16], collectively orchestrating apoptosis through ATP exhaustion [12,17]. Unlike ECT, the coupling of calcium ions with IRE does not produce pigmentation, but it mandates further validation and exploration [18].

The chemical augmentation of IRE-based treatments encounters barriers, chiefly the requirement for uniform molecule distribution within the target tissue. A novel approach using near-infrared (NIR) as a photon booster for IRE treatment has been introduced [19]. With its wavenumber spectrum spanning 4000 to 12,500 cm⁻¹, NIR light, boasting longer wavelengths, offers superior tissue penetration compared to visible light [20]. NIR light-mediated photobiomodulation instigates signaling cascades corresponding to intracellular Ca²⁺ fluctuations. Factors influencing these changes include exogenous Ca²⁺ influx and endoplasmic reticulum-driven Ca²⁺ release, suggesting ROS-induced transmembrane depolarization as a plausible light-activated signaling route [21]. However, existing research posits that NIR’s effect on the membrane potential is multifaceted; while some studies highlight hyperpolarization in stem and cancer cells [19,22], others indicate an energy density-dependent depolarization or hyperpolarization [19]. A noteworthy observation is that hyperpolarization, especially at 2–4 J/cm², significantly bolsters electroporation efficacy compared to depolarization [23,24]. The inclusion of cationic peptides also influences this dynamic, accentuating the membrane’s negative charge, thereby modulating the electrostatic potential [25]. The ramifications of these changes extend to the efficiency of molecular transport across the membrane after electroporation [26,27]. Given these findings, we postulate that priming the tissue with NIR irradiation, culminating in cytoplasmic membrane hyperpolarization, might bolster IRE-driven apoptosis or necrosis. Capitalizing on the uniform irradiation capability of NIR, we hypothesize that pre-regulating tissues with NIR could synergistically elevate IRE’s electroporation efficacy. To validate this proposition, we turned to rat gastric tissues for our investigations, considering their accessibility for electroporation and NIR via endoscopic means. To further elucidate their combined effects at the protein level, we introduced and assessed apoptosis-related proteins such as B-cell lymphoma-extra-large (Bcl-xl), cyclin-dependent kinase inhibitor (p21), and superoxide dismutase 2 (SOD2).

2. Materials and Methods

2.1. Numerical Analysis for Electric Field Distribution

Initially, we performed a numerical simulation to predict the electric field distribution in IRE, referencing a previous study [28]. The simulation employed a disk-type electrode (Tweezertrode (45-0487), BTX, Holliston, MA, USA) with an exposure diameter of 3 mm and a distance of 2 mm (Figure 1A,B). We used a commercial finite element package (COMSOL Multiphysics, version 6.1; COMSOL, Stockholm, Sweden) for the task. The simulation.

Parameters were set with an applied voltage of 250 V around the electrodes, exhibiting a conductivity of 1.37 × 10⁶ S/m [29]. The governing equation used in the simulation was the Laplace equation:

$$-\overrightarrow{\nabla }·\left(\mathsf{\sigma }\overrightarrow{\nabla }\varnothing \right)=0$$

(1)

where $\varnothing$denotes the electric potential and $\mathsf{\sigma }$ signifies the consistent tissue conductivity. For the purpose of the simulation, the rat stomach tissue was modeled as a flat model, measuring 15 mm vertically and 10 mm horizontally, and with a thickness of 2 mm. This model comprised 4,488,676 elements. The electrically exposed surface adhered to the boundary conditions of $\varnothing$ = V (source) and $\varnothing$ = 0 (sink). Conversely, the remaining surfaces were treated as electrically insulated.

2.2. Numerical Analysis for Temperature Distribution

Temperature distribution within the tissue was modeled using Pennes’ bioheat equation [30]:

$$\rho C_p\frac{\partial T}{\partial t}+\overline{\nabla }·\overline{q}=Q+Q_{bio}, \overline{q}=-\kappa \nabla T, Q_{bio}={\rho }_b{C}_b{\omega }_b\left(T_b-T\right)$$

(2)

where $\rho$ is the tissue density (1088 kg/m³), $C_p$is the specific heat capacity of the tissue (3690 J/kgK), T is the temperature (K), $\overline{q}$ is the heat flux by conduction (W/m²), $Q$ is the heat source (W/m³), ${\rho }_b$is the blood density (1000 kg/m³), $C_b$ is the specific heat capacity of blood (4180 J/kgK), ${\omega }_b$ is the blood perfusion rate (0.00715 /s), $T_b$ is the arterial blood temperature (310.15 K), and $\kappa$ is the thermal conductivity (0.53 W/mK). Both the tissue and the electrode commenced with an initial temperature of 37 °C. Detailed electrical conditions are elaborated in

Table 1. Experimental conditions.

No.	Control	Group 1 (IRE)	Group 2 (NIR)	Group 3 (NIRE)
1	No treatment	Applied voltage: 250 V	λ: 808 nm	NIR + IRE
2		Pulsing: 20	Power density: 695 mW
3		Pulse duration: 100 µs	Energy density: 3 J/cm2
4		Pulse interval: 1 s	-
5		Electrode distance: 2 mm	-

. The electric field exhibited uniform distribution between the electrodes ensuring that the temperature difference between them did not exceed 32 °C (Figure 1C).

2.3. Animals

For this study, 15 female Sprague Dawley rats, aged 4 weeks and weighing between 300 and 400 g, were procured from Orient Co., Ltd. (Seoul, Republic of Korea) in the Republic of Korea. The entire study, including the experimental procedures, data acquisition, interpretation, and analysis, was performed following the guidelines set by the Animal Research: Reporting of In Vivo Experiments guidelines.

2.4. IRE Procedure

In this study, we employed an NIR laser system (λ: 808 nm, LVI-LSR808-1000, Laser Lab Co., Ltd., Gyeonggi-do, Republic of Korea) that had the capability of emitting a 1000 mW output in both continuous wave (CW) and pulsed wave modes. This system was equipped with a 1 mm diameter multimode optical fiber core that allowed incident light delivery at a 30° angle from the core axis and positioned the target 3 mm away from the tissue (Figure 1A). Additionally, an electrical generator (BTX, Holliston, MA, USA) was employed, which could produce a maximum output of 3000 V (50 A at 100 µs pulse width) for the IRE procedure. Following NIR irradiation, electric pulses were delivered via the disk electrode based on the above simulation. The experimental group was divided into four groups: control (n = 6), NIR irradiation (n = 3), IRE (n = 3), and NIR irradiation followed by IRE (NIRE) (n = 3). In the control group, the electrode was contacted to the tissue and no electric pulsing was applied. The NIR-irradiation group was exposed to NIR rays of wavelength 808 nm in the CW mode at 695 mW to deliver a dose of 3 J/cm². In the IRE group, an external 250 V was applied as 20 pulses between electrodes placed 2 mm apart, each with a pulse width of 100 µs. The NIRE group was subjected to both NIR irradiation, as in the NIR group, and IRE, as in the IRE group. Tissue impedance was analyzed by recording both pre- and post-pulsing impedance values that were displayed on the generator.

For the experiment procedure, rats were first administered enrofloxacin (5 mg/kg) and ketoprofen (5 mg/kg). Anesthesia was induced using alfaxalone (10 mg/kg) and xylazine (10 mg/kg). After shaving and decontamination of the skin using iodine surgical scrub and 70% ethanol, the abdominal wall was dissected to expose the stomach. A 5 mm length of the stomach’s body part was incised (Figure 1B(2)), and electrodes were inserted on either side of the targeted tissue (Figure 1B(3)).

During the electrode placement in the control group, only mere electrode contact was ensured. For the NIR irradiation group, the tissue with a 4 mm diameter target size received NIR irradiation aiming for a 3 J/cm² dose, without electrode intervention. The IRE group underwent IRE between the disk electrodes, maintaining a 2 mm gap by means of pressing the electrodes. In the NIRE group, the target tissue was initially subjected to NIR irradiation without pressing the electrodes, ensuring accurate targeting. This was followed by an IRE application, pressing the electrodes in the manner adopted by the IRE group. Impedance readings from the generator were recorded for each step of the experimental process. Following IRE treatment, the rats were euthanized 12 h later for ensuing analysis.

2.5. Hematoxylin and Eosin Staining

Histological examinations were performed using hematoxylin and eosin (H & E) staining. The excised tissues, taken from the stomachs, were halved and spread out, with a few corners pinned for stability (Figure 1A). They underwent fixation in a 10% neutral buffered formalin, followed by dehydration through an ethanol series. They were then embedded in paraffin, sectioned into 4 µm thick slices, and placed on slides for H & E staining. Digital imaging of the slides was facilitated using the Pannoramic 250 FLASH III scanner (3DHISTECH Ltd., Budapest, Hungary). An analysis of these images, including the assessment of ablation, was conducted using the CaseViewer software (Version 2.4, 3DHISTECH Ltd.).

2.6. TdT-Mediated dUTP-Biotin Nick End-Labeling Staining

The ApopTag Peroxidase In Situ Apoptosis Detection Kit S7100 (Qbiogene, SA, Germany) was employed to detect apoptotic cells, following the manufacturer’s instructions. Briefly, after deparaffinization, the previously sectioned samples were treated with 3.0% hydrogen peroxide for 5 min. This was followed by two rinses in phosphate-buffered saline, each lasting 5 min. A working-strength TdT enzyme was then applied to the samples. After application of anti-digoxigenin for counterstaining, the samples were mounted onto slides. Digital images of these samples were acquired using the Pannoramic 250 Flash III scanner (3DHISTECH Ltd.).

2.7. Western Blotting Analysis

To perform a Western blot analysis, cells from the treated stomach were isolated and lysed using RIPA lysis buffer (pH 7.8), comprising 50 mM Tris-HCl, 150 mM NaCl, 1% IGEPAL, 10 mM NaF, 0.1 mM EDTA, and a protease inhibitor cocktail (Sigma Aldrich, Missouri, USA). Proteins were separated by performing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred to a polyvinylidene difluoride (PVDF) membrane. The membrane was then incubated with primary antibodies against B-cell lymphoma-extra-large (Bcl-xl) (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA), cyclin-dependent kinase inhibitor (p21) (Invitrogen), and superoxide dismutase 2 (SOD2) (Invitrogen) at a 1:1000 dilution overnight at 4 °C. The following day, the membrane was incubated with secondary antibodies (m-IgGk BP-HRP; IgG-HRP; chicken anti-goat igG-HRP, Santa Cruz, Santa Cruz Biotechnology Inc., Dallas, TX, USA) at a dilution of 1:5000 for 1.5 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) substrate (SuperSignal, ThermoFisher Scientific, Waltham, MA, USA). Densitometric values of the proteins were quantified, normalized to β-actin (Santa Cruz), and represented as histograms.

2.8. Statistical Analyses

Statistical analyses were performed on datasets collected from the tumor tissues. The results are presented as the mean ± standard deviation. Statistically significant differences between the experimental groups were identified using a two-tailed paired Student’s t-test (significance levels, * p < 0.05; ** p < 0.01) or a one-way analysis of variance (significance level, ^# p < 0.05; ^## p < 0.01; ^### p < 0.001). When significant differences were found in ANOVA, Tukey’s post hoc analysis was conducted for pairwise comparisons. All statistical analyses were carried out using R version 4.2.1 software (2022-06-23 ucrt).

3. Results

3.1. Impedance Analysis

The initial analysis revealed a significant impedance difference between the IRE and NIRE groups (Figure 1D). The NIRE group showed an impedance variation of 246 Ohm before and after pulsing (p < 0.001), whereas the IRE group exhibited a change of 233 Ohm (p < 0.001). Significantly, the impedance in the NIRE group was greater by 13.5 Ohm compared to the IRE group (p < 0.01).

3.2. Histopathological Analysis

The results of histopathological analysis are depicted in Figure 2. H & E staining showed that all samples retained the typical four-layered structure, consisting of mucosa, submucosa, muscularis, and serosa (Figure 2). The TdT-mediated dUTP-biotin nick end-labeling (TUNEL) assay indicated that the NIR-treated samples were unaltered, mirroring the control samples and showing no signs of apoptosis. In contrast, the IRE and NIRE treatments resulted in evident tissue destruction in the mucosal and submucosal areas (Figure 2A(3,4)). The TUNEL assay confirmed the induction of apoptosis by the IRE and NIRE treatments (Figure 2B(3,4)). Quantifying the apoptosis revealed that the comparison among groups (p < 0.001) was significantly significant (Figure 2C). Both IRE and NIRE differed significantly from the control group (16.7-fold, p < 0.01; 18.7-fold, p < 0.05), with the NIRE group displaying significantly higher values (1.12-fold, p < 0.05) compared to the IRE group. Specifically, the NIR-treated samples maintained intact mucosa, submucosa, and muscularis layers, akin to the control samples (Figure 3(1–6)). However, the NIRE-treated samples (Figure 3(10–12)) had fewer cells across all layers compared to the NIR-treated samples (Figure 3(7–9)). Both the IRE and NIRE treatments resulted in samples showing decellularization, a residual loose extracellular matrix, and condensed nuclei, alongside minor vascular hemorrhaging and preserved vasculature. Notably, fibrin deposition was observed interspersed between the vascular layers in both the mucosal and muscularis, although no differences were identified in the submucosa samples.

Furthermore, the vascular structural integrity within the mucosal and submucosal layers was scrutinized (Figure 4A). Capillary-like structures in the mucosal and submucosal tissues of the control and NIR-treated samples were intact with surrounding cells. In comparison, the IRE-treated samples showcased fewer surrounding cells but maintained vessel integrity. Vessels from the NIRE-treated samples paralleled those from the IRE samples. The TUNEL assay demonstrated apoptotic activity in the mucosal vessels of the IRE and NIRE treatments, whereas the control and NIR-treated samples displayed viable cells (Figure 4B). Nonetheless, the TUNEL assay did not specifically target perivascular cells in the submucosa (Figure 4B(6,8)).

3.3. Protein Expression Analysis Using Western Blotting

Sample apoptosis was assessed at the protein level, as depicted in Figure 5, by specifically examining the expression of apoptosis-related proteins: Bcl-xl, p21 and SOD2. These proteins were selected for Western blot evaluation due to their roles in apoptosis. Bcl-xl, a mitochondrial transmembrane protein from the Bcl-2 family, is pivotal in regulating the intrinsic apoptosis pathway. P21 functions in cell adhesion molecule binding and integrin binding. SOD2 plays a role in electron transfer activity pathways. The protein expression of these proteins in each experimental groups was visualized as bands on the Western blot (Figure 5) and quantitatively measured as fold changes relative to β-actin.

By performing ANOVA analysis, significant differential expression was observed for P21 (p < 0.05) and SOD2 (p < 0.01), while the expression of the BCL protein was not significantly different among groups. Furthermore, the interaction among the proteins was evident; BCL expression was significantly correlated with both P21 (p < 0.01) and SOD2 (p < 0.001), while P21 also displayed a significant relationship with SOD2 (p < 0.001). In a detailed analysis, the NIRE group showed a significantly higher Bcl-xl protein expression than the IRE group (1.24-fold, p < 0.05). Compared to the control group, P21 protein expression was downregulated in both the NIR and IRE groups (0.78-fold, p < 0.05 for NIR; 0.58-fold, p < 0.05 for IRE). Yet, the NIRE group exhibited an elevated p21 expression compared to the IRE group (1.29-fold, p < 0.05). SOD2 expression in the NIR group was noticeably lower than in the control group (0.6-fold, p < 0.05), but its expression was not significantly different in other groups compared to the controls. The NIRE group, however, displayed a reduced SOD2 protein expression relative to the IRE group (0.93-fold, p < 0.05). Taken together, when compared to the IRE group, the NIRE group demonstrated activated expression of p21 and SOD2, both regulators of apoptosis. Moreover, the NIRE group exhibited activated Bcl-xl expression, indicative of its role in apoptosis inhibition.

4. Discussion

We investigated the combined effect of NIR with IRE treatment, so-called NIRE, to enhance apoptosis in murine gastric tissue. Utilizing an energy density of 3 J/cm² for NIR, based on prior research [19], we observed that NIR influenced the electrostatic properties of the tissue. This was evident from the altered impedance after IRE and NIRE treatment (Figure 1B). While, it is established that NIR modifies the optical properties of tissue [31], limited studies address its impact on tissue’s electrical characteristics. One explanation for the observed impedance shift could be the temperature-dependent resistivity principle. As the tissue absorbs NIR, it may undergo temperature changes, potentially altering its electrical attributes. Despite the thermal effects suggested by another study [32], temperature may play a minor role in influencing electrical impedance. Simultaneously, NIR’s potential to change the tissue’s dielectric properties could subsequently affect its electrical impedance. Supporting this, our prior in vitro research demonstrated that NIR irradiation can alter electrical capacitance and conductivity [19]. Additionally, our findings reveal that NIR’s effects on electrical properties are mediated through NIRE. Histopathological analysis showed that NIRE treatment led to notably more apoptosis than IRE alone (Figure 2C), suggesting that NIR could enhance tissue’s sensitivity to IRE by altering conductivity. While the TUNEL assay offers insights into DNA fragmentation during apoptosis [33], our H & E analysis revealed consistent outcomes across various tissue layers with IRE and NIRE treatments. However, the WB method highlighted a difference in protein expression linked to apoptosis. WB, a widely recognized molecular biology technique, was employed to identify protein variations induced by IRE or NIRE, especially focusing on apoptotic signals.

Apoptosis, a programmed cell death, typically manifests through unique signaling pathways [34]. Bcl-xl, associated with the BCL2-like 1 gene, acts as an anti-apoptotic protein by preventing mitochondrial content release, such as cytochrome c [35]. The p21 protein regulates cell cycle progression [36], while SOD2 facilitates electron transport activity, converting oxidative phosphorylation byproducts to hydrogen peroxide and diatomic oxygen [37]. Among these proteins, SOD2 expression was observed to increase after IRE treatment [38,39]. In addition, p21 protein expression was enhanced by IRE treatment [40] by using Western blotting. However, Bcl-xl expression was reduced by IRE treatment [41]. In our experiment, Bcl-xl expression decreased with IRE but increased further with NIRE treatment compared to IRE. Bcl-xl, a member of the Bcl-2 family, is vital in apoptosis regulation [42], and its altered expression in many cancers helps evade apoptosis [43]. Conversely, the activation of the BCL2-associated x protein (BAX) and BCL2-antagonist-killer (BAK) can initiate apoptotic events through cytochrome c release [44]. The enhanced Bcl-xl expression in NIRE-treated samples, compared to controls, suggests a potential increase in apoptotic signals. Further studies on the expression of Bcl-xl due to NIRE treatment are needed. Moreover, p21 activation was more prominent in NIRE- than IRE-treated samples, aligning with previous findings that link p21 activation to DNA damage and subsequent apoptosis [45,46,47]. In addition, diminished SOD2 expression across experimental groups relative to controls implies that ROS production might be influenced directly by either treatment. Given SOD2’s role in safeguarding against cell death [48], our findings suggest NIRE could potentiate apoptosis alongside IRE. Investigating Bcl-xl, p21, and SOD2 expressions in depth, especially concerning their roles in apoptosis, remains crucial.

Furthermore, IRE treatments have been effective for liver, pancreatic, and prostate cancer due to electrode accessibility. However, gastric cancers present challenges, as electrodes are difficult to position. Nevertheless, our prior study affirmed the viability of IRE application using catheter electrodes combined with endoscopy for gastric and esophageal regions [49]. This minimally invasive method promotes patient recovery after IRE for gastric cancer treatment and offers the potential of pairing NIR and IRE via an optical fiber for photon energy delivery in an endoscopic setting. Research has shown that unaffected cells remain after resection [49], emphasizing the need for more refined electrode designs and the potential benefits of integrating NIR with IRE for eradicating residual cells.

This study had some limitations. Firstly, normal and tumor gastric tissue differ, leaving NIRE’s full efficacy in tumor ablation uncertain. Secondly, the significance of hyperpolarization in rat gastric tissue by NIRE warrants validation through immunohistochemistry or WB in subsequent research. Lastly, while our gene enrichment analysis was based on microarray data, practical WB evaluations may not entirely capture all proteins linked to apoptosis, necessitating further study on apoptosis-related proteins.

5. Conclusions

Our findings demonstrate that NIRE treatment resulted in a tissue impedance change of 13.5 Ohm relative to IRE treatment. In terms of apoptosis induction, NIRE led to an increase of 1.12 times in comparison to IRE. Furthermore, the expression of the apoptosis-associated protein BCL was amplified by 1.24 times and p21 protein by 1.29 times, and SOD2 exhibited a 0.93-fold expression in NIRE-treated tissues as compared to those treated using IRE. This enhanced response can be attributed to the interactions between NIR and cells, which seems to modulate cellular metabolism prior to IRE treatment. Importantly, we observed no adverse or toxic effects from NIR exposure on the tissue. These findings hold promise for augmenting the apoptotic efficacy of IRE in therapeutic settings.