Focused shock waves and inertial cavitation release tumor-associated antigens from renal cell carcinoma

Abstract

Tumor biomarkers play an essential role in immunotherapeutic strategies in cancer treatment, contributing to early diagnosis, patient selection, treatment monitoring, and personalized treatment plans. Despite their importance in cancer care, circulating biomarkers may not always be detectable or sufficiently elevated to provide reliable test results. Due to the pressing need for innovative approaches to enhance biomarker levels, this study explored the potential use of focused shock waves and cavitation for non-invasively releasing tumor-associated antigens. Renal carcinoma cell lines ACHN and TOS-1 were used in an in vitro study to analyze the impact of shock waves on two membrane glycosphingolipid antigens, MSGG and G1, respectively. Focused shock waves were generated using a partial spherical piezoceramic dish. Flow-cytometric analysis of treated cells immediately after 1,000 focused shock waves at 16 MPa overpressure showed a 29.4 % and 17.6 % decrease in MSGG and G1 antigens on the cell surfaces. In the immunostaining of glycosphingolipid fractions on thin-layer chromatography (TLC), both tumor markers were reduced by an average of 49.30 % (MSGG) and 57.08 % (G1). Immunoelectron microscopy images confirmed decrease in the cell membrane intensity immediately after shock waves because of the release of antigens into the extracellular spaces. The released antigens were primarily found on cell debris formed by shock waves and cavitation induced damage to the cell membrane. Theoretical analyses were performed to understand antigen release mechanisms. Moreover, the biophysical events that occurred following the interaction of a shock wave with a suspended cell were modeled and clarified. A novel model was used to calculate the tensile stresses following shock waves and to explain the deformations observed in scanning electron microscopy images. The release of tumor antigens by focused shock waves and inertial cavitation represents exciting prospects for advancing cancer care strategies.

1. Introduction

Identification and integration of tumor biomarkers into cancer management strategies has transformed the field of cancer treatment, leading to more precise and patient-centric approaches in immunotherapy, [1]. As our knowledge of tumor immunology and immunotherapy advances, the role of immunogenic tumor markers in cancer care continues to grow. As a result, tumor biomarkers have become indispensable tools in early detection, diagnosis, prognosis, treatment selection, and monitoring. Through the years, understanding molecular structure and functionality of tumor markers have also allowed for a wider variety of tumor markers to be considered for diagnostic and therapeutic applications. Among the growing number of tumor markers investigated for these purposes are tumor-associated glycosphingolipids (GSLs), a well-defined and structurally characterized group of highly diverse glycolipid molecules located in the cell membrane of tumor cells [2], [3].

Generally, a higher than normal level of tumor-associated GSL antigens with an aberrant phenotype can be found in some cancerous cells. In this context, a correlation between cancerous transformation and expression of tumor-associated GSLs has been demonstrated for multiple tumors, such as human melanoma, glioma, and gastrointestinal cancers [4]. A number of renal carcinoma cell lines (RCC) are also examples of tumor cells characterized by higher expression of tumor-associated-GSL antigens [5], [6]. In some cases, GSLs can trigger the immune system of the patients to produce monoclonal antibodies (mAb) against them. These mAbs have been utilized for cancer diagnosis and monitoring by measuring the presence or changes in the levels of GSLs. Moreover, analyzing GSLs in blood or other body fluids, known as liquid biopsy (LB), are being investigated for their potential to detect cancer and evaluate treatment response [7]. In addition to diagnostic applications, a number of GSLs tumor markers have also been studied in a variety of therapeutic settings [8], [9]. Preclinical studies have shown that tumor-associated GSLs could be potential immunotherapeutic targets for suppressing tumor growth; e.g., growth inhibition of T cells lymphoma, and human melanoma by L5178, and anti-Gg3/anti-GD3, respectively [10], [11]. In addition, some GSLs tumor markers have been considered as potential targets for cytotoxic T lymphocytes (CTLs) during ex-vivo vaccine development [12].

While these strategies have shown promising results in experimental model systems, clinical trials have not always produced uniformly satisfactory results, and there have been instances of unexpected tumor relapse and distant metastases. In many of these cases, the primary reason for the failure is an insufficient immune response due to the low level of detectable tumor markers by the immune system. The results of several studies have further showed that immunogenetic markers could effectively stimulate immune responses only when they were expressed at levels exceeding a certain threshold [13]. It was therefore essential for the success of tumor marker-based treatment approaches to use techniques that enhance expression of immunogenic tumor markers or boost their release into the extracellular space. To achieve this goal, experimental and preclinical studies have examined using genetically engineered tumor cells for editing and enhancing the expression of tumor markers [14], [15]. In spite of their advances, however, these methods were limited in their application due to technical difficulties associated with their implementation, unwanted side effects, the limited number of suitable tumor candidates, and the high cost.

Since impulsive release of energy, such as shock waves and cavitation, can trigger various cellular responses in cells, including changes in gene expression, we proposed using shock waves as a non/less invasive method for modifying antigenic expression on tumor cells. Focused shock waves and cavitation used in our study could damage the tumor microenvironment, thus increasing the possibility of immune cell infiltration within the tumor, and potentially enhancing immune responses. The results showed that focused shock waves could in fact enhance the expression of hidden intracellular tumor-associated GSL antigens on the surface of renal cell carcinoma cells [16]. Shock impulse stress disrupted the intracellular cell cytoskeleton and damaged the cell membrane in the treated cells, resulting in increased expression. The immediate but transient nature of the elevated GSL expression on the cell surfaces was a novel finding, showed for the first time that shock pulsed stress alone could initiate a chain of events leading to antigenic enhancement, which was independent of genes-stimulated pathways [16].

This study aims to investigate the direct and immediate effects of focused shock waves and cavitation on the expression of tumor-associated antigens in the cell membrane and the release of tumor biomarkers into the extracellular space. The mechanism of cell damage leading to the release of tumor antigens is then explained by the physics governing cavitation and the dynamics of shock waves propagation in a single cell during cell-shock wave interactions. The findings of this study provide novel bases for applications of focused shock waves as a non–/less invasive method of manipulating tumor biomarkers. Exploring the use of releasing tumor antigens physically in combination with other cancer care modalities, such as immunotherapy or targeted therapy, may offer synergistic benefits for boosting the effectiveness of existing treatment techniques.

2. Materials and methods

2.1. Cell lines

Two human renal carcinoma cell lines used in this study were TOS-1 and ACHN cells. TOS-1 cell line derived from a metastatic tissue of a patient with RCC was developed by the Department of Urology, Tohoku University [5]. ACHN cell line obtained from malignant plural effusions in RCC patients was purchased from American Tissue Culture Collection (ATCC). Both cells were cultured in Dulbecco’s modified Eagle’s medium (MEM) supplemented with 10 % fetal bovine serum (FBS), 1 % penicillin/streptomycin and 1 % L-glutamine (all from Gibco, NY, USA), in humidified atmosphere of 5 % CO2 at 37 °C.

2.2. Monoclonal antibodies

Two primary monoclonal antibodies (mAbs), mAb-RM1 and mAb-RM2, were kindly provided by Dr. M. Satoh (Department of Urology, Tohoku University School of Medicine, Sendai, Japan [17]. Primary mAb-RM1 is used to measure the expression of monosialogalactosyl-globoside antigen (MSGG) on ACHN cells [18], while mAb-RM2 measures the expression level of disialoganglioside GalNAcDSLc4 antigen or G1 on TOS-1 cells [6]. Anti-mouse IgM antibody produced in goat (Sigma-Aldrich, Darmstadt, Germany) was used as a negative control.

2.3. Shock wave apparatus

Shock waves were generated using an in-house designed apparatus, as previously described [16]. In brief, it is composed of a partial spherical piezoceramic dish (125 mm inner diameter and 16 mm depth) placed in a water bath filled with degassed water at 30 °C. Fig. 1A illustrates the experimental setup. Fig. 1B shows enlarged views of the pressure–time history at 0 μs and 103 μs measured at the focal point. The pressure was measured by a needle hydrophone (Muller-Platte Needle Probe, Germany) with 0.5 mm diameter sensitive area and 50 ns rise time. A negative pressure of −2.5 MPa preceded the shock front, which was created due to shrinkage of the piezoceramic element. It was followed by a positive shock wave of 16 MPa (P_max) with 0.6 µs pulse duration and 0.04 mJ mm⁻² positive energy flux density. The shock wave was succeeded by a tensile wave of −2 MPa negative pressure, 1.8 μs duration, and 2 μJ.mm⁻² negative-energy flux density [16]. The tensile wave behind the shock wave was induced by the expansion wave at the opening of the piezoceramic dish, which was intensified by the wave convergence, and caused inertial cavitation in the focal area [19]. The collapse of the cavitation bubble induced a positive pressure pulse of 1.5 MPa at 103.8 μs. Fig. 1C illustrates the radial distribution of pressure at the focus perpendicular to the axis. Around the focus, the focal zone outline (50 % of shock wave maximum overpressure ratio P/P_max = 1/2 or −6 dB) had a symmetrical oval shape with a radius of 1.5 mm in the radial direction and a major axis length of 13 mm in the axial direction. Therefore, the volume of the focal zone was about 60 μl. The shock wave pressures were highly repeatable (±5% of peak pressure), thus the cells were exposed to the same wave profile during each exposure.

Fig. 1.

A schematic diagram of the shock wave experimental setup. (A) Suspended human renal carcinoma cells, TOS-1 or ACHN, were placed in the focal area of a piezoceramic generator located in a degassed water bath at 37 °C. Cells were exposed to 1,000 shock waves at 5 Hz repetition rate. (B) Pressure-time history measured at the focal point shows a shock wave with peak pressure of 16 MPa at 0 μs and cavitation-induced pressure of 1.5 MPa at 103 μs. (C) Radial distribution of shock wave pressure at the focus perpendicular to the axis.

2.4. Shock wave treatment

Cultured cells were treated with 0.02 % EDTA solution (Gibco, Thermo Fisher Scientific Inc. Grand Island, NY), washed with phosphate-buffered saline (PBS) and collected to 1x10⁶ cells/ml in MEM medium. A total of 2 ml of the cell suspension was transferred into a polyethylene tube with an inner diameter of 13 mm for shock wave treatment. To ensure maximum shock wave propagation with minimal reflections, the bottom of the tube was cut and sealed with a thin layer of 25 µm polyethylene film having the same acoustic impedance as the medium. By using an XYZ micrometer positioning stage (Sigmakoki Co., Japan), the tube containing the cells was precisely positioned in the focal area of the shock wave generator. Fig. 1A illustrates the position of the suspended cells. The number of shock waves used in the experiment was previously determined based on the amount of shock waves required to result in 50 % loss of cell viability (LD50) for the treated cells [16]. Therefore, suspended cells were exposed to 1,000 shock waves (+SW) with 5 Hz repetition frequency at 16 MPa over pressure [16]. The sham control cells (−SW) were similarly treated, without the shock wave exposures.

2.5. Flow cytometry

Double staining flow cytometry with a primary monoclonal antibody (mAb) and propidium iodide (PI) was utilized to measure the expression of the targeted antigens exclusively on the surface of viable cells. The level of MSGG expression on ACHN cells was determined with FITC-labeled primary mAb-RM1. FITC-labeled primary mAb-RM2 was used for measuring the expression level of G1 on TOS-1 cells [5]. In brief, 500 µl of the cell suspension (ACHN or TOS-1) were incubated with 50 µl of the respected primary mAb (mAb-RM1 or mAb-RM2) for one hour at 4 °C, washed twice with PBS washing buffer (1 %BSA/PBS and 0.05 %NaN₃), treated with 50 µl fluorescein (FITC)-conjugated anti-mouse IgG + IgM (Sigma Chemical Co., Japan) for one hour on ice in the dark, washed twice with PBS, incubated with 20 µl PI (Sigma Chemical Co., Japan) for 30 min, suspended in 300 µl Isotone solution (Becton Dickinson, Sunnyvale, CA, USA) and finally measured by FACScan cellQuest flow cytometer (Becton Dickinson, San Jose, Ca, USA). Auto-fluorescence signals measured in a group of suspended cells in PBS, without labeling them with any antibody were subtracted from the final results. Data were analyzed with CELLQuestTM software with minimum 10,000 events per sample. The expression levels of the antigens were measured immediately after the shock wave exposures. Sham-exposed samples were processed in the same way, except for the shock waves.

For the negative control, samples were prepared as described above without incubation with the respected primary monoclonal antibody.

2.6. Extraction and purification of gangliosides

Glycolipids were extracted from the cells as described previously [5]. Briefly, about 500 mg of packed cells (1×10⁸ cells) were washed with PBS. Crude glycolipid was extracted from the cell pellets by homogenization and filtration with 15 volumes of isopropyl alcohol/hexane/water (55:25:20, v/v/v) and chloroform methanol (2:1,1:1, and 1:2 v/v). The extracts were evaporated and separated into the upper and the lower phases by Folch partitioning method [20]. The upper phase was then dialyzed against distilled water for 2 days and fractionated by DEAE-Sephadex-A25 (Pharmacia, Uppsala, Sweden) column chromatography into upper neutral glycolipids and acidic gangliosides fractions [21]. Acidic fractions containing the gangliosides extracted from the same wet weight of cells along with standard gangliosides were placed on high performance thin-layer chromatography (HPTLC) plates (Baker, NJ, USA) using micro-syringe, developed in a solvent system of chloroform–methanol-0.5 % aqueous CaCl2 (50:40:9 v/v/v), and visualized by spraying with 0.5 % Orcinol in sulfuric acid (Orcinol-H2S04).

2.7. Immunostaining of thin-layer chromatography (TLC)

TLC-immunostaining was performed as previously described [22]. HPTLC plates were immunostained according to a modified version of Magnani‘s procedure. Briefly, total ganglioside fractions were applied on HPTLC plates using a solvent system of chloro-form methanol water (CMW) 50:40:10 containing 0.05 % CaCl₂. Plates were air-dried and immersed in high molecular weight 0.5 %-polyisobutylmetracrylate (PITM) (Sigma-Aldrich Chemical Co., WI, USA) in ether for 1 min, and blocked with 5 % BSA/PBS for 1 h at room temperature, then reacted overnight with primary mono-clonal antibodies, mAb-RM1 (anti-MSGG), or mAb-RM2 (anti-G1), and nonimmune mouse IgM at 4 °C. Plates were then washed 3 times with cold PBS and incubated with biotinylated anti-mouse IgM (Sigma-Aldrich Chemical Co., WI, USA), as the secondary antibody, for 1 h at room temperature. The plates were then stained with vector avidin–biotin solution for 30 min and finally stained with 3‘,3-diaminobenzidine. The data from TLC immunostaining was analyzed by ImageJ image processing program for quantitative measurements [23].

2.8. Scanning electron microscopy

Shock wave treated cells were immediately fixed in 2.5 % glutaraldehyde plus 2 % paraformaldehyde for 2 h and then post fixed in 1 % osmium-tetroxide solution for 1 h. After dehydration in an ascending ethanol series, cells were dried with liquid CO₂ with a critical point dryer (Hitachi, Tokyo, Japan) and sputtered with platinum-palladium (Pt-Pd). Morphology of the shock wave treated cells was compared with the control under LEM1200EX scanning electron microscope (JEOL, Tokyo).

2.9. Immunoelectron microscopy

The membrane GSL antigens were localized in the subcellular spaces by pre-embedding indirect immuno-peroxidase method [24]. In brief, cell pellet was fixed in 4 % paraformaldehyde/0.05 M phosphate buffer for 2 h. After a serial washing in 10 %, 15 % and 20 % sucrose/PBS, cell pellet was embedded in OCT compound, frozen in dry ice-ethanol, and stored at –80 °C. Frozen sections of the cell pellet (5 µm thick) mounted on albumin-coated glass slides were immersed in 1 % BSA/PBS and incubated with the first mAb-RM1, for 12 h at 4 °C, followed by overnight incubation with peroxidase-conjugated F(ab')2 fragments of anti-mouse IgM diluted at 1:100 with PBS containing 10 % normal human serum, as the second antibody (Histofine Kit SAB-PO; Nichirei Co. Tokyo, Japan). Diaminobenzidine DAB (Dojin, Kumamoto, Japan) was used as chromogen at a concentration of 30 mg/100 ml Tris-HCL buffer with 0.006 % H₂O₂. To minimize the endogenous peroxidase activity, 65 mg/100 ml sodium azide was added to DAB. The reaction with DAB was visualized later by incubating the samples with 1 % osmium tetroxide for 20 min. After dehydration in ethanol, ultra-thin sections embedded in Epon were coated with lead citrate and observed with LEM1200EX scanning electron microscope (JEOL, Tokyo). The images were analyzed by ImageJ for quantitative measurement of the intensity of the expressed antigens in the cell membranes [23].

2.10. Statistical analysis

Data were analyzed using unpaired t-test including Welch’s correction. Data represent mean ± SD of three independent experiments. Results were considered to be significant when the corrected P-value was less than 0.05, indicated as P<0.05 in the figure legends.

3. Results

3.1. Flow cytometric measurement

Membrane expression of MSGG antigen on ACHN and G1 antigen on TOS-1 cells were monitored by double staining flowcytometry using PI and their respective primary monoclonal antibodies, mAb-RM1 and mAb-RM2. A dot plot and histogram analysis of the antigenic expressions in control ACHN (Fig. 2A, a) and TOS-1 cells (Fig. 2C, c) are shown in comparison to the expression on shock wave-treated ACHN (Fig. 2B, b) and TOS-1 cells (Fig. 2D, d).

Fig. 2.

Double staining flowcytometric analysis of GSLs expressions on ACHN and TOS-1 renal cell carcinoma cells affected by shock waves. (A) Dot plot of the sham control ACHN cells. (a) Expression level of MSGG antigen exclusively on viable control ACNH cells compare to negative control (NC). (B) Dot plot of the shock wave-treated ACHN cells. (b) Expression level of MSGG antigen exclusively on viable ACNH cells following shock waves. (C) Dot plot of the sham control TOS-1cells. (c) Expression level of G1 antigen exclusively on viable control TOS-1 cells. (D) Dot plot of the shock wave treated TOS-1 cells. (d) Expression level of G1 antigen exclusively on viable TOS-1 cells following shock waves.

The total population of the cell in each sample was divided into PI positive (R3) and PI negative (R1 and R2) regions based upon the intensity of PI (FL2-Hight) (Fig. 2A – D). Eliminating the dead cells from the antigen analysis, expression of MSGG or G1 antigens (FL1-Hight) was then measured using only the subset of cells gated in R1 and R2 regions, (Fig. 2a – d). For each sample group, ΔM was calculated as follow:

M1_(-SW) – M1_(+SW) = ΔM (1)

where M1_(-SW) represents the antigen expression on the cell samples in the sham control group, while M1_(+SW) represents the expression after shock wave treatment.

As shown in Fig. 2b and d, antigenic expressions of MSGG and G1 on the surfaces of the cells decreased immediately following the shock wave. MSGG expression was reduced by an average of one third in ACHN cells (29.41 % ± 0.45 %) as compared to the control cells (Fig. 2a and b). Similarly, the expression levels of G1 on TOS-1 cells were decreased by about 17.57 % ± 0.45 % (Fig. 2c and d).

3.2. TLC analysis

Immunostaining in the context of TLC analysis typically involves the use of antibodies to detect specific compounds separated on the TLC plate.

Here, after extraction of GSLs from the cell membranes of ACHN and TOS-1 cells and TLC separation of the GSLs compounds, the density of the antigen of interest in the samples were quantitatively measured by immunostaining of the antigen with the respected primary mAb-RM1 or mAb-RM2. When compared with the controls, TLC analysis of the samples revealed a drop of 49.30 % and 57.08 % in MSGG and G1 fractions in the shock wave treated ACHN and TOS-1 cells, respectively (Fig. 3A – D).

Fig. 3.

Effects of shock waves on glycosphingolipid (GSL) fractions in renal cell carcinoma. (A) TLC immunostaining of MSGG fraction extracted from ACHN. (B) TLC immunostaining of G1 fraction extracted from TOS-1. (C) Shock wave effect on the intensity of MSGG fraction extracted from ACHN. (D) Shock wave effect on the intensity of G1 fraction extracted from TOS-1 cells. Each data point represents mean ± SD (n = 3), *P<0.05.

3.3. Immunoelectron microscopic analysis

Immunoelectron microscopic image of a control TOS-1 cell shows G1 antigens as electron-dense deposits in the cell membrane (Fig. 4A). An enlarged image of the control cell revealed highly immune-stained vesicular structure (50–100 nm) in close proximity to the cell membrane (Fig. 4a). However, the antigens were almost absent from the cytoplasm of the control cells.

Fig. 4.

Immunoelectron microscopic images of TOS-1 cells immune-stained for membrane G1 antigen. (A) Control TOS-1 cell. (a) An enlarged view of the control TOS-1 cell shows that G1 antigen is mainly found in the cell membrane (short arrows) and in vesicles next to the membrane (long arrows). The control cells shed limited number of membranous structures containing G1 antigen (asterisk). (B) Shock wave treated TOS-1 cell. (b) An enlarged view of a shock wave treated cell shows a significantly reduction in the expression of G1 antigen in the cell membrane (short arrow), as well as in the adjacent cytoplasmic vesicles (long arrow). By contrast, a strong expression of G1 is found in the debris detached from the cell (arrowheads) into the extracellular space. ICS represents intra-cellular space, while ECS indicates extra-cellular space.

The images of the treated cells immediately after the shock waves and cavitation revealed development of various morphological deformations based on severity of the damage, ranging from reversible membrane damages to complete rupture of the cells. Bleb formation and release of membrane debris from the damaged cell membranes. In addition, the shock waves treated cells exhibited vacuolization of the cytoplasm, where cytosolic membranous structures joined each other and formed larger structures in the cytoplasm (Fig. 4B, b).

Following the shock waves, the density of the G1 antigen detected in the cell membrane and the vesicular structures decreased noticeably (Fig. 4B, b). The shock waves exposures have also resulted in the vesicular structures joining the cell membrane. In contrast, G1 was found highly expressed in the blebs and debris that separated from the cell membrane of the treated cells and released into the extracellular space. In addition, the images did not show any introduction of antigens into the cytoplasm of the treated cells.

3.4. Morphological analysis

Scanning electron microscopic images of the control TOS-1 cells in suspension showed the spherical cells covered with microvilli abundant on their surfaces (Fig. 5A).

Fig. 5.

Scanning electron microscopy of TOS-1 renal cell carcinoma. (A) Control cell. (B, C) Shock waves treated cells in suspension.

After the shock wave and cavitation exposures, the treated cells showed various morphological deformations based on the severity of the damage; such as the cell membrane depression (Fig. 5B), elongated cells and bleb formation (Fig. 5C). The impingement of focused and cavitation shock waves on the cell membrane caused the membrane to dent in the direction of the shock wave, shown in Fig. 5B and C. A decrease in the number of microvilli on the surface of treated cells can be seen due to damage to the underlying cell cytoskeleton.

3.5. Quantifying focused shock wave parameters

In this study, focused shock waves were applied at 16 MPa pressure with 0.04 mJ mm⁻² energy flux density. This is comparable to a pressure of 7 to 80 MPa with 0.004 to 0.6 mJ mm⁻² energy flux density that is commonly used in clinical extracorporeal shock wave therapy [25], [26]. The cell suspensions were subjected to total energy per volume of 0.43 J ml⁻¹ after 1,000 focused shock wave pulses. This amount of energy would cause about 0.1 °C of temperature rise based on the cell suspensions' specific heat capacity of 4.217 J.g⁻¹.K⁻¹ [27]. In our measurements, there was no noticeable bulk temperature increase in the samples following exposure to shock waves. The bulk thermal effect was therefore small and insignificant in the current study.

The shock waves had a focal volume of 60 μl, while the suspended cells had a volume of 2 ml. There was also no restriction on the movement of cells in suspension. As a result, only a small percentage of the cells (3 %) were exposed to focused shock wave during one exposure. In other words, individual suspended cells were subjected to an average of 30 focal pressures during 1,000 shock waves exposure. This number is comparable to that of 10 to 30 shock waves typically applied to cultured cells [26], [28].

The average shock wave pressure in focal region was 8 MPa (Fig. 1C), which according to the cell medium’s shock Hugoniot equation [16], [29], [30] would generate a particle velocity or microstreaming of 5.3 m.s⁻¹. During the full width at half maximum (FWHM) time duration of the shock wave pulse of 0.32 μs (Fig. 1B), the microstreaming caused an average particle displacement of 1.7 μm [31], [32]. In part, this particle movement may have contributed to the separation of debris from cells, as shown in Fig. 4B-b.

3.6. Dynamics of shock waves propagation in cells

From a physical standpoint, one of the significant parameters associated with effects of shock waves, whether direct or cavitation induced, on cells is their primary impact on the geometry and nonhomogeneous structures [33], [34], [35]. The shock front thickness in the cell medium is about 1 to 2 nm, approximately equal to a few mean free paths of water molecules [36], which is less than the cell membrane thickness of 5 to 10 nm (including lipid bilayer and embedded proteins) [36]. When the shock front impinges on the cell, it reflects, refracts, and diffracts over the cell membrane and subcellular cell structures. The cell membrane, containing lipid, has an acoustic impedance of about 1.38 MRayl (density 952 kg.m⁻³ and speed of sound 1,459 m.s⁻¹) [37]. The cytoplasm’s acoustic impedance is 1.60 MRayl (density 1,032 kg.m⁻³ and speed of sound 1,550 m.s⁻¹) [38], whereas the cell’s surrounding medium has an impedance of 1.499 MRayl (density 1,007 kg.m⁻³ and speed of sound 1,489 m.s⁻¹) [29], [39]. Therefore, when the transmitted shock wave inside the cell interacts with the membrane, it reflects as an expansion (rarefaction) wave. Fig. 6 illustrates shock wave/cell interaction dynamics model based on the descriptions provided above.

Fig. 6.

A schematic diagram modeling the interaction between an incident shock wave (ISW), either direct or cavitation induced, with a suspended cell (R_cell = 5 μm). (A) At t = 2.2 ns after the interaction, a transmitted shock wave (TSW) propagates inside the cell’s cytoplasm, a reflected wave (RW) propagates upstream, and the cell membrane (CM) moves with the particle velocity behind the TSW. (B) At t = 5 ns, the TSW further propagates and reflects as a reflected expansion wave (REW). (C) At t = 6.6 ns, after interaction of the TSW with the downstream membrane (DSM) it transmits to the medium and reflects as a converging REW. (D) At later time of t = 200 μs, the upstream membrane (USM) flattens, the DSM elongates, and the interfacial instability (II) grows.

During shock wave focusing, the shock front evolves from a converging (concave) configuration to a nearly straight shape at the focus, and then propagates with a diverging (convex) shape [19]. Accordingly, the radius of curvature of shock waves is of the millimeter order (1.5 mm focal zone radius, Fig. 1A), two orders of magnitude higher than the diameter of cells (R_cell = 5 μm). Therefore, the radius of curvature of the incident shock wave is ignored in the model of Fig. 6, when compared with the cell dimensions. The model is also valid for point source shock waves, such as laser or cavitation, interacting with cells, where the change in shock wave curvature, speed, and pressure with propagation complicates the calculations.

The reflected and transmitted pressures can be obtained from the pressure reflection relations [32], [37]:

$$\frac{P_r}{P_i}=\frac{\left(Z_2/\cos {\theta }_t\right)-\left(Z_1/\cos {\theta }_i\right)}{\left(Z_2/\cos {\theta }_t\right)+\left(Z_1/\cos {\theta }_i\right)},\frac{P_t}{P_i}=\frac{2Z_2/\cos {\theta }_t}{\left(Z_2/\cos {\theta }_t\right)+\left(Z_1/\cos {\theta }_i\right)},$$ (2)

where P_r, P_t, and P_i are reflected, transmitted, and incident pressures, Z₂ and Z₁ are acoustic impedances of transmitted and incident mediums, θ_i and θ_t are the angles of incident and transmitted waves, respectively. The transmitted shock wave inside the cytoplasm (Fig. 6A and B) reflects from the membrane (Fig. 6C) with a negative pressure of −0.60 MPa for the 8 MPa average incident pressure, assuming a normal incident (θ_i = θ_t = 0). This tensile stress causes a lateral tension τ (τ = RP, where R is the cells radius and P is the pressure) equal to 3.0 J.m⁻² in the membrane of TOS-1 and ACHN cells (R_cell = 5 μm) [40]. A typical cell plasma membrane (red cells [40]) ruptures at a lateral tension of 0.01–0.023 J.m⁻² on laboratory time scales, based on the membrane’s critical areal strain of 2–5 % [31], [40]. While the lateral tension due to the average tensile wave was two orders of magnitude higher than the typical cell membrane rupture tension, the 0.32 μs FWHM time scale (Fig. 1B) of the tensile wave loading was six orders of magnitude less than the laboratory time scale [40]. The short time duration of the stress loading may explain why the tensile stress of the reflected expansion wave modified (Fig. 4) but did not rupture the cell’s membrane (Fig. 5).

The reflected expansion wave has a concave shape, caused by the nearly spherical geometry of the cell membrane (Fig. 6C); therefore, it converges and increases its pressure amplitude and energy density as it propagates inside the cell cytoplasm affecting the intracellular organelles. The shock wave deforms the cell shape in the direction of the interaction, where it flattens the upstream membrane and stretches the downstream membrane as shown in Fig. 6D. The distortions are the result of particle movement behind the shock wave [30], estimated to be 1.7 μm from shock wave microstreaming. This may explain mechanism of the deformations observed in the scanning electron microscopy images of the treated cells in Fig. 5B and C.

The shock wave passage impulsively accelerates the interface between the cell membrane and the cytoplasm, which induces interfacial (Richtmyer–Meshkov) instability [41], [42]. The instability results from the coupling between the pressure gradient across the shock and the density gradient across the membrane; so called the induction of baroclinic vorticity. The instability perturbs the cell membrane and generates bubbles and spikes of fluids penetrating the membrane, which in turn causes mixing of the cell’s cytoplasm and extracellular fluids.

Furthermore, the mechanisms of shock wave effects for cells adhering to rigid surfaces (e.g. slides, bone, or stone) [28], [43], [44] differ from those for suspended cells or tissues with cell–cell adhesion. While the mechanisms of effects of shock waves, either direct or cavitation induced, on cells in medium (or tissue or tumor) are explained here, for cells adhering to a rigid boundary, strong shock wave reflection due to significant acoustic impedance mismatch between the solid and the fluid, as well as shock/boundary layer interaction (over the rigid surface) are main mechanisms of the cells damage.

The negative pressure rarefaction wave behind the focused shock wave may have caused inertial cavitation, which also impacted the cells. During inertial cavitation, the cavitation bubbles grow, over-expand, and then shrink. The concentration of energy during the final stage of the bubble collapse generates plasma, leading to local shock waves (Fig. 1B at 103 μs), ultraviolet (UV) light, free radicals and reactive oxygen species (ROS), all affecting cells near the focal extension [26], [44]. With the presence of a rigid or viscoelastic boundary, the bubble collapse generates a strong liquid jet that pierces the cells. This causes irreversible poration and cell death [43]. However, in the current suspended cells setup (absence of a rigid boundary) liquid jet effects are minimal and negligible [31].

4. Discussion and conclusion

The growing understanding of the effects of external stresses on human tissue has led to the development of a variety of medical approaches, including ultrasound and shockwave treatments [36]. Clinical applications of extracorporeal shock waves provide examples of mechanical stresses that can be harnessed for a variety of medical purposes [44]. During past decades, considerable progress has been made in understanding the interactions between shock waves and cells, including tumor cells [28]. However, a deeper understanding of the underlying mechanisms by which shock waves interact with cells and their subcellular structures is needed to facilitate the development of innovative medical procedures that exploit specific cellular responses. A shock wave, in principle, is characterized by a single surge of high-pressure wave with extremely fast rise time and short pulse duration, that travels through the body and exerts a significant impulsive mechanical stress on the targeted area. Studies have shown that, in addition to extensive damage in the focal area, the cavitation and the dissipating energy of the diverging shock waves can also damage and disrupt the sub-cellular structures of tissues in close proximity to the targeted site [28]. Consequently, at the cellular level, shock waves, direct or cavitation induced, can have a variety of effects depending on the amount of energy transmitted to cells, including permanently damaging the cell membrane, leading to cell death, or causing reversible changes to the cell membrane [36].

In this study, quantitative analysis of the expression levels of two targeted tumor-associated antigens, MSGG and G1, showed that the tumor markers significantly decreased immediately following the shock wave exposures (Fig. 2). The reduced intensity of immune-stained GSL bands in TLC samples of the treated cells compared with the control (Fig. 3) confirmed the flowcytometric results, showing a similar level of reduction for the treated cells. Although cells were exposed to the same stresses, the difference in the reduced expression levels of MSGG and G1, which were respectively about 30 % and 18 % in ACHN and TOS-1 cells, is believed to be due to the differences in the cell type and their responses to the mechanical stress. Hence, the findings indicate that the responses of cells to shock waves are influenced by both physical and biological factors [45].

Furthermore, to determine if shock waves have a comparable effect on the expression of other molecules, or if the effect is specific to GSLs, we also measured the effect of focused shock waves on E-cadherin, a transmembrane protein molecule. Interestingly, E-cadherin was similarly reduced by focused shock waves as GSLs (Data not shown). This result suggested that shock waves may also be effective at targeting other molecular structures.

Detailed immunoelectron microscopy results confirm that antigens were absent within the cells after shock wave treatment, with antigens primarily found in the extracellular space, as discussed in Section 3.3. The decrease in antigen levels on the cell membrane, as quantitatively measured by flow cytometry and TLC, combined with their absence from the cytoplasm, indicates that the antigens were exclusively relocated to the extracellular space.

Application of focused shock waves is shown here to cause lateral stress, generating tension in the cell membrane that initiates a cascade of events. These events begin with membrane damage, leading to an influx of extracellular Ca²⁺, cytoskeletal damage, increased intracellular pressure, cytosolic flow, lipid aggregation in the cytoplasm forming cytosolic vesicle, bleb formation on the cell surface, and the eventual shedding of surface antigens with the cell membrane debris, as shown in Fig. 7. These cellular responses align with the concept of mechanotransmission, where mechanical forces applied to the cell membrane propagate to the cell interior, driving biochemical changes. The morphological modifications observed in treated cells, such as intracellular vesicular growth, bleb formation, and fusion of GSL antigen-rich vesicles with the damaged membrane, seen in Fig. 4, Fig. 5, support this mechanistic link.

Fig. 7.

Schematic diagram of shock wave effects on expression level of tumor-associated GSL antigens in the cell membrane of renal cell carcinoma. Application of shock wave causes a cascade of events that starts with the mechanical damage to the cell membrane (1), follows by damages to the cytoskeleton (2), and in-flux of extracellular Ca²⁺ into cells (3), which increases intra-cellular pressure, cytosolic flow, and lipid aggregation in the cytoplasm. The stresses generate blebs (4), fusion of the vesicles to join the cell membrane to prevent the rupture of the cells (5), and finally shedding of the damaged cell membrane in form of debris containing high level of GSL antigens into extracellular space (6).

These chains of events on the cells (Fig. 7) can be supported by the observed morphological modification in the treated cells, such as the growth of vesicular structures in the cell cytoplasm, formation of blebs, fusion of the vesicular structures rich in GSLs antigens with the damaged cell membrane, and release of cell membrane debris into the extracellular space (Fig. 4, Fig. 5). In general, cells whose membrane is partially damaged by the shock waves can survive through a membrane repair system that is triggered with influx of Ca²⁺ into the cytoplasm [46], [47], [48]. The influx leads to multiple rapid exocytosis events and patching of lipid vesicles with the damaged site to block the flow into/out of the plasma membrane. This cellular defense system can further explain the movement of the vesicles under the disrupted cell membrane when they fused rapidly with the adjacent plasma membrane, seen in this study. Although it was expected that the fusion of vesicles containing high level of GSL antigens should have increased the intensity of the antigens measured in the cell membrane, the level of antigens eventually decreased in the treated cells as a result of the total detachment of the cell membrane in form of debris.

In conclusion, we demonstrated that focused shock waves and inertial cavitation can manipulate tumor biomarkers and release them into extracellular spaces. The findings provide promise for improving the effectiveness of antigen-based cancer care strategies currently being used in the clinical setting. While this noninvasive method offers novel outcomes, it is still in an early stage of development. As we continue to gain a better understanding of the interactions between shock waves and cavitation with biomarkers, additional research would be necessary to determine their effect on other cancer biomarkers. The optimization of shock wave energy or its waveform to alter inertial cavitation would also contribute to the desired outcomes based on tumor cell types and biomarkers.

CRediT authorship contribution statement

Nushin Hosano: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Zahra Moosavi-Nejad: Writing – review & editing, Writing – original draft, Validation, Project administration, Formal analysis. Takuichiro Hide: Writing – review & editing, Validation, Project administration, Formal analysis. Hamid Hosano: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.