**Low-Intensity Sonoporation-Induced Intracellular Signalling of Pancreatic Cancer Cells, Fibroblasts and Endothelial Cells**

**Ragnhild Haugse 1,2,3 , Anika Langer <sup>3</sup> , Elisa Thodesen Murvold 4,5, Daniela Elena Costea 3,5 , Bjørn Tore Gjertsen 3,6, Odd Helge Gilja 5,7, Spiros Kotopoulis 5,7,8,**† **, Gorka Ruiz de Garibay 3,**† **and Emmet McCormack 1,2,3,4,9,\* ,**†


Received: 29 September 2020; Accepted: 3 November 2020; Published: 6 November 2020

**Abstract:** The use of ultrasound (US) and microbubbles (MB), usually referred to as sonoporation, has great potential to increase the efficacy of chemotherapy. However, the molecular mechanisms that mediate sonoporation response are not well-known, and recent research suggests that cell stress induced by US + MBs may contribute to the treatment benefit. Furthermore, there is a growing understanding that the effects of US + MBs are beyond only the cancer cells and involves the tumour vasculature and microenvironment. We treated pancreatic cancer cells (MIA PaCa-2) and stromal cells, fibroblasts (BJ) and human umbilical vein endothelial cells (HUVECs), with US ± MB, and investigated the extent of uptake of cell impermeable dye (calcein, by flow cytometry), viability (cell count, Annexin/PI and WST-1 assays) and activation of a number of key proteins in important intracellular signalling pathways immediately and 2 h after sonoporation (phospho flow cytometry). Different cell types responded differently to US ± MBs in all these aspects. In general, sonoporation induces immediate, transient activation of MAP-kinases (p38, ERK1/2), and an increase in phosphorylation of ribosomal protein S6 together with dephosphorylation of 4E-BP1. The sonoporation stress-response resembles cellular responses to electroporation and pore-forming toxins in membrane repair and restoring cellular homeostasis, and may be exploited therapeutically. The stromal cells were more sensitive to sonoporation than tumoural cells, and further efforts in optimising sonoporation-enhanced therapy should be targeted at the microenvironment.

**Keywords:** sonoporation; microbubbles; ultrasound; intracellular signaling; phosphorylation; ultrasound contrast agents; drug delivery; cellular stress; pancreatic cancer; tumour microenvironment

#### **1. Introduction**

The use of ultrasound (US) and microbubbles (MB) in combination with chemotherapy to increase the efficacy of cancer therapy has gained interest in the last 20 years. The term "sonoporation", is often used to describe this phenomenon. The term describes the formation of pores that occur when cells come into contact with MBs oscillating in an US field [1], hypothesised to enhance uptake for co-administered chemotherapeutics [2]. However, there is still no consensus on what the exact mechanisms underlying US + MBs enhanced cancer therapy are. Furthermore, the cellular stress induced by US + MBs themselves has been proposed to contribute to the anti-cancer effects [3,4].

Despite the insufficient mechanistic understanding, substantial in vitro [5–13] and in vivo research [5–9,14–17] has shown that US + MBs-enhanced cancer therapy is beneficial for cancer therapy. In 2016, results from the first Phase 1 human clinical trial using US + MBs in combination with chemotherapy to treat pancreatic ductal adenocarcinoma (PDAC) were published [18]. PDAC is a deadly cancer, with less than an 8% five-year overall survival [19], which requires better treatment options. The clinical trial demonstrated that the use of US + MBs is safe, and a secondary endpoint indicated that sonoporation + chemotherapy (gemcitabine) may increase survival of patients. PDAC is characterised by extensive desmoplastic stroma, thought to originate from cancer-associated fibroblasts (CAFs), and a complex hypoxic tumour microenvironment, which are major contributors to resistance to chemotherapy [19,20]. Some hypothesised effects of sonoporation on the tumour microenvironment are increased drug extravasation from the blood vessels or destruction of tumour vasculature [2], but the relevance of microenvironmental effects for the clinical efficacy of sonoporation remains largely unknown.

In vitro sonoporation research has typically focused on cancer cell lines, evaluating the sonoporation efficacy either by uptake of cell impermeable dyes [5,8,9,21] and/or by evaluation of the viability of cells exposed to sonoporation [21] or sonoporation in combination with drugs [5–9,22]. Based on the fact that MBs are usually injected into the vasculature, in vitro studies have also addressed the effects of sonoporation on endothelial cells, showing increased permeability both in the cellular membrane [23,24] and interendothelial openings between cells [25–28].

Our previous study [4] using leukemic cells as a model system of cancer compared to healthy peripheral blood cells demonstrated that different cell types have different sensitivities to sonoporation in terms of molecular uptake, effect on viability and intracellular signalling. This was a simplified system compared to solid tumours, which consist of the complex tumour microenvironment with stromal cells, immune cells, fibroblasts, blood vessels and extracellular matrix (ECM) in addition to, and supporting, the cancer cells [29]. The observed difference in sonoporation sensitivity raises the question of whether the cell types within a solid tumour may also respond differently, meaning that treatment of the cancer cells might not be the most important sonoporation effect in the enhancement of chemotherapeutic efficacy. Investigations on the cellular responses to US + MBs should, therefore, be carried out in cell types relevant for the tumour microenvironment.

In this study, we aim to improve our understanding of how sonoporation may affect pancreatic cancer by evaluating the effects on the uptake of cell impermeable dye, viability and intracellular signalling response in PDAC, endothelial and fibroblast cell types. The results from this study will help expand our understanding of how the different cell types respond to sonoporation, which is needed for the development of better in vitro and in vivo models for sonoporation, the optimisation of sonoporation parameters and the choice of drugs. In fact, the results indicate that the cancer cells are not the most sensitive to sonoporation in terms of uptake of cell-impermeable molecule, reduction in viability or intracellular signalling response (phosphorylation of p38, ERK1/2, ribosomal protein S6 and 4E-BP1), further suggesting that cells in the tumour microenvironment may be relevant for sonoporation efficacy.

#### **2. Materials and Methods**

#### *2.1. Chemicals*

All chemicals were purchased from Merck KGaA (Darmstadt, Germany) unless otherwise stated.

#### *2.2. Maintenance Cell Culture*

Pancreatic ductal adenocarcinoma cell line MIA PaCa-2 (ATCC® CRM-CRL-1420TM, kindly donated by Professor Anders Molven, University of Bergen, Bergen, Norway) was cultured in high-glucose Dulbecco's modified Eagle's medium (DMEM #5671) supplemented with 10% foetal bovine serum (FBS), 2% L-glutamine, 1 mM sodium pyruvate and 2.5% horse serum. Human foreskin fibroblasts (BJ, ATCC® CRL-2522TM, kindly donated by professor Donald Gullberg, University of Bergen, Bergen, Norway) were cultured in high-glucose DMEM #5671 supplemented with 10% FBS, 1% L-glutamine and 50 U/mL Penicillin/50 U/mL Streptomycin. Single-donor lot human umbilical vein endothelial cells (HUVECs) (Cat#CC-2517, Lonza, Basel, Switzerland, kindly donated by Prof. Jim Lorens, University of Bergen, Bergen, Norway) were cultured in EGMTM-2 medium supplemented with EGMTM-2 Endothelial Cell Growth Medium-2 BulletKitTM (Lonza). Cell culture medium was changed on the HUVECs every second day according to suppliers' recommendations, except when cultured in Petaka G3 LOT® (Celartia Ltd., Powell, OH, USA) [c.f. Section 2.4]. HUV-EC-Cs (ATCC® CRL-1730TM) were cultured in F12-K medium (ATCC, Manassas, VA, USA) supplemented with endothelial growth supplement from bovine neural tissue (#E2759) and 10% FBS. All cells were cultured in a 5% CO<sup>2</sup> humidified atmosphere at 37 ◦C.

#### *2.3. Isolation of Cancer-Associated Fibroblasts (CAFs)*

CAFs were isolated from cancer tissue biopsies of PDAC patients with primary lesions after informed consent, and in accordance with the Declaration of Helsinki and approval by the Ethics Committee (REK 2013/1772). In brief, tissues were collected and washed in Dulbecco's modified Eagle's medium (DMEM #6429) supplemented with 2% antibiotic-antimycotic (AB/AM), 100 U/mL penicillin, 100 µg/mL Streptomycin and 25 ng/mL Amphotericin B (Thermo Fischer Scientific, Waltham, MA, USA). Bleeding and necrotic areas of the tissues were cut out using a sterile scalpel and washed thoroughly. Tumour tissues were then cut into approximately 2–4 mm<sup>2</sup> tissue bits and then transferred onto a 10 cm culture dish, slightly air-dried (approximately 2 min) to allow explants to attach on the growth surface of the dish, and incubated in FAD medium (DMEM/Nutrient Mixture F-12 Ham supplemented with 10% FBS, 0.4 mg/mL hydrocortisone, 1% Insulin-Transferrin-Selenium, 50 mg/mL L-ascorbic acid, 10 ng/mL epidermal growth factor and 1% AB/AM). Explants with an outgrowth of cells with fibroblast morphology were trypsinized in a clonal ring placed around the respective explant and fixed on the bottom of the dish with sterile vaseline and placed in 6-well plates. Isolated fibroblasts were further characterised for lineage-specific markers: ESA, CD31, CD45 and CD140b for epithelial, endothelial, blood-borne and mesenchymal origin, respectively. CAFs were further cultured in DMEM (#6429) supplemented with 10% FBS and without AB/AM.

#### *2.4. Cell Culture for Experiments*

Three days prior to experiments, 26 mL of cell suspension was injected in Petaka G3 LOT® (low oxygen transport) cell culture chambers (Figure 1a referred to as Petaka herein). A suitable number of cells were injected to achieve confluency at the time of US/MB exposure. To avoid the influx of air, the air valve on the Petakas was sealed with tape, and they were cultured in the horizontal position for a minimum of 24 h to allow for cells to adhere to the plastic surface. The Petaka chambers are designed for cell culture within a limited gas exchange and a gradually decreasing oxygen concentration [30], which is closer to pO<sup>2</sup> ("physioxia") found in living tissue compared to the "normoxic" conditions at atmospheric O<sup>2</sup> pressure commonly used in vitro [31].

acoustic focus.

cultured in Petakas.

SonazoidTM (GE Healthcare, Little Chalfont, UK) was reconstituted by adding 2 mL NaCl 9 mg/mL (B Braun, Melsungen, Germany) and gently agitated for 30 s. MBs were aspirated via a 19 G needle and transferred to an eppendorf tube. A 19 G venting needle was used to avoid a pressure drop in the vial. To ensure that the reconstituted bubbles were stable, Sonazoid™ bubbles were used within 1 h of reconstitution. A 60 µL volume of reconstituted Sonazoid™ (1.2 × 109 bubbles/mL) was diluted in NaCl 9 mg/mL to a total volume of 1 mL and injected into the Petaka immediately prior to US exposure, giving a concentration of 2.8 × 106 bubbles/mL in the Petaka. In the untreated sample, and US alone controls, 1 mL of NaCl 9 mg/mL was added to Petaka. The Petakas were gently rolled

The cells were exposed to US using a custom-made US treatment chamber (Figure 1), based on a previous design [32] and previously used in [22], suited for US exposure of adherent cells when

The US system consisted of 128, 9 × 6 mm PZ26 elements firing upwards as a plane-wave into the Petaka, ensuring US treatment of the entire cell-covered surface in the Petaka. The US transducers were driven by a custom Open Ultrasound system (Lecoeur Electronique, Chuelles, France). The

in all directions to ensure a homogenous distribution of MBs and NaCl.

*2.6. In Vitro Treatment with Ultrasound and Microbubbles* 

**Figure 1.** (**a**) Drawing of cell culture bioreactor (Petaka) used for culturing of cells prior to ultrasound (US) treatment; (**b**) cutaway of custom-made US treatment chamber used for US treatment of cells (adapted from [22], Pharmaceutics, 2020). **Figure 1.** (**a**) Drawing of cell culture bioreactor (Petaka) used for culturing of cells prior to ultrasound (US) treatment; (**b**) cutaway of custom-made US treatment chamber used for US treatment of cells (adapted from [22], Pharmaceutics, 2020).

#### Before US treatment, air pockets were removed from the Petaka. To ensure that the floating MBs *2.5. Microbubbles*

would come into contact with the cells, the Petaka was placed in the US treatment chamber (Figure 1) with the cell-covered side on top, closest to the US absorber. A low-Mechanical Index (MI) (<0.01) B-mode scan was performed before treatment with US ± MB to detect air pockets in the US bath between the water medium and the Petaka. Air pockets, if present, were removed before US treatment. The US conditions used were based on a previous study [22] and referred to as "Medium US" or "High US" (see Table 1 for details). As untreated control, a Petaka, containing cells but no MBs, was placed in the US treatment chamber for 5 min without application of US. All experiments were performed in triplicate at minimum, except CAFs, where sonoporation experiments were only performed once and only treated with "High US", due to limited material availability. Imaging of SonazoidTM (GE Healthcare, Little Chalfont, UK) was reconstituted by adding 2 mL NaCl 9 mg/mL (B Braun, Melsungen, Germany) and gently agitated for 30 s. MBs were aspirated via a 19 G needle and transferred to an eppendorf tube. A 19 G venting needle was used to avoid a pressure drop in the vial. To ensure that the reconstituted bubbles were stable, Sonazoid™ bubbles were used within 1 h of reconstitution. A 60 <sup>µ</sup>L volume of reconstituted Sonazoid™ (1.2 <sup>×</sup> <sup>10</sup><sup>9</sup> bubbles/mL) was diluted in NaCl 9 mg/mL to a total volume of 1 mL and injected into the Petaka immediately prior to US exposure, giving a concentration of 2.8 <sup>×</sup> <sup>10</sup><sup>6</sup> bubbles/mL in the Petaka. In the untreated sample, and US alone controls, 1 mL of NaCl 9 mg/mL was added to Petaka. The Petakas were gently rolled in all directions to ensure a homogenous distribution of MBs and NaCl.

#### cells was performed immediately after sonoporation with 10× magnification using a Nikon Eclipse E200 microscope equipped with a Lumenera Infinity 1 camera. *2.6. In Vitro Treatment with Ultrasound and Microbubbles*

The cells were exposed to US using a custom-made US treatment chamber (Figure 1), based on a previous design [32] and previously used in [22], suited for US exposure of adherent cells when cultured in Petakas.

The US system consisted of 128, 9 × 6 mm PZ26 elements firing upwards as a plane-wave into the Petaka, ensuring US treatment of the entire cell-covered surface in the Petaka. The US transducers were driven by a custom Open Ultrasound system (Lecoeur Electronique, Chuelles, France). The acoustic field had been calibrated in three axes using a 200 µm needle hydrophone (Precision acoustics Ltd., Dorset, UK) in the fully assembled US chamber, and the Petaka was placed at the acoustic focus.

Before US treatment, air pockets were removed from the Petaka. To ensure that the floating MBs would come into contact with the cells, the Petaka was placed in the US treatment chamber (Figure 1) with the cell-covered side on top, closest to the US absorber. A low-Mechanical Index (MI) (<0.01) B-mode scan was performed before treatment with US ± MB to detect air pockets in the US bath between the water medium and the Petaka. Air pockets, if present, were removed before US treatment. The US conditions used were based on a previous study [22] and referred to as "Medium US" or "High US" (see Table 1 for details). As untreated control, a Petaka, containing cells but no MBs, was placed in the US treatment chamber for 5 min without application of US. All experiments were performed in triplicate at minimum, except CAFs, where sonoporation experiments were only performed once and only treated with "High US", due to limited material availability. Imaging of cells

was performed immediately after sonoporation with 10× magnification using a Nikon Eclipse E200 microscope equipped with a Lumenera Infinity 1 camera.


**Table 1.** US parameters (5 min treatment with US ± microbubbles (MBs)).
