B16-F10 murine melanoma cells (ATCC) were cultured in Advanced MEM medium (AMEM, Gibco), and 4T1 mammary carcinoma cells were cultured in Advanced RPMI 1640 medium (Gibco). Both media were supplemented with 5% fetal bovine serum (FBS, Gibco), 10 mM L-glutamine (GlutaMAX, Thermo Fisher Scientific), penicillin-streptomycin (100x, Sigma-Aldrich) and held in a 5% CO₂ humidified atmosphere at 37°C. All cells were routinely tested to be mycoplasma-free (MycoAlert™ Plus Mycoplasma Detection Kit, Lonza).

In experiments, eight- to nine-week-old (18-21 g) female C57Bl/6NCrl (C57Bl/6) and BALB/cAnNCrl (BALB/c) mice (Charles River Laboratories) were housed under specific pathogen-free conditions at a temperature of 20–24°C, a relative humidity of 55 ± 10%, a 12-h light–dark cycle and ad libitum food and water. All experimental procedures were performed in compliance with the guidelines for animal experiments of the EU directives (2010/63/EU), ARRIVE Guidelines, and the permission of the Ministry of Agriculture, Forestry, and Food of the Republic of Slovenia (Permission No. U34401-1/2015/43). Mice were randomly assigned to experimental groups ( Table 1 ; Supplementary Table 1 ), and the number of animals in each group is indicated in the Supplementary Tables 3, 4 .

Electroporation buffer (EPB) consisted of 136 mM NaCl, 5 mM KCl, 2 mM MgCl₂, 10 mM HEPES, and 10 mM glucose (pH 7.2, 300-310 mOsm/kg) (41). The calcium solution was prepared as CaCl₂×6H₂O in distilled water to 500 mM in the stock solution. For in vitro experiments, calcium solutions diluted in EPB to 5×10^-3 M, 4×10^-3 M, 3×10^-3 M, 2×10^-3 M, and 1×10^-3 M concentration in a final cell suspension were used. Bleomycin was diluted in EPB to 1.4×10^-6 M, 1.4×10^-7 M, 1.4×10^-8 M, 1.4×10^-9 M, and 1.4×10^-10 M concentration in a final cell suspension. For in vivo experiments, 50 mM, 168 mM, and 250 mM calcium solutions diluted in distilled water were used. Bleomycin was diluted in saline solution to a final concentration of 0.177 mM (10 μg/40 μl) for in vivo experiments.

The therapeutic plasmid encoding mouse IL-12 (pORF-mIL-12-ORT; pIL-12) (42) was isolated and purified using the EndoFree Plasmid Mega Kit (Qiagen) and diluted in endotoxin-free Milli-Q water. Plasmid quality and concentration were assessed by the 260/280 ratio (Epoch microplate spectrophotometer) and by agarose gel electrophoresis. For the experiments, the plasmid was diluted in saline to a final concentration of 0.25 μg/µl (p.t. application) and 0.5 μg/µl (i.t. application).

Cells were trypsinized and collected by centrifugation, washed in ice-cold EPB, and prepared for electroporation. Briefly, 50 μl of cell suspension (1×10⁶ cells) served as the treatment control, while the other 50 μl, containing calcium solution or bleomycin (concentrations described in 2.2), was pipetted between two custom-made stainless-steel parallel-plate electrodes (1.9 mm apart), and EP were applied (8 square-wave pulses, 1300 V/cm, 100 μs, 1 Hz) with an electric pulse generator GT-01 (Faculty of Electrical Engineering, University of Ljubljana, Ljubljana, Slovenia). Then, the cells were transferred to a 24-well ultralow attachment plate and incubated for 5 min at room temperature, followed by the addition of 1 ml of culture medium. To determine cell viability after treatment with calcium or bleomycin, alone or in combination with EP, a viability assay with Presto Blue (Thermo Fisher Scientific) was performed. Cells were diluted to 1×10⁴ cells/ml (control groups) or 2×10⁴ cells/ml (EP groups) and transferred to 96-well plates, with each well containing 1×10³ cells (exposed to the drug alone) or 2×10³ cells (exposed to combined treatment with EP). The plates were then incubated for 72 h in a 5% CO₂ humidified atmosphere at 37°C. Presto Blue reagent (10 μl/well) was added to cells, followed by 1 h incubation in a humidified incubator at 5% CO₂ and 37°C. The fluorescence emission was measured using a microplate reader Cytation 1 (BioTek) with a 530/590 nm excitation/emission filter. The measured fluorescence intensity of the treated groups was normalized to the control group or control EP group.

For induction of tumors, a suspension of 1×10⁶ B16-F10 and 0.5×10⁶ 4T1 cells in 100 μl of 0.9% NaCl was injected subcutaneously into the right flanks of mice. The used tumor models differ in immunogenicity, 4T1 breast carcinoma is more immunogenic than B16-F10 melanoma (43). Treatment was performed when tumors reached volumes of ~40 mm³. Throughout the treatment, mice were under inhalation anesthesia with 1.5 – 2.0% isoflurane delivered with anesthesia system connected to oxygen concentrator at a flow rate of 1 l/min (Supera Anesthesia Innovations, Clackamas). The study design and protocol are presented in Figure 1 .

In the first part, intratumoral (i.t.) therapy was performed with different calcium concentrations (50 mM, 168 mM, and 250 mM in dH₂O) and bleomycin solution (1.41 mM in saline). Injections were performed with insulin injection needle (29 gauge) and injection time was ~30 sec. In some cases, when injecting 4T1 tumor, due to its high solidity (compactness), the solution was leaking into peritumoral space, regardless of the injection time. EP were delivered immediately after i.t. injection of 40 μl of calcium solution or 2 min after injection of bleomycin. Control group tumors were injected intratumorally (i.t.) with a saline solution. The EP were generated by an ELECTRO cell B10 HVLV (LEROY Biotech, Betatech) with parallel stainless-steel electrodes in 2 sets of 4 pulses in a perpendicular direction (voltage-to-distance ratio 1300 V/cm, frequency 1 Hz, duration 100 μs). Good contact between the skin and electrodes was ensured by using a conductive water-based gel (Ultragel, Budapest, Hungary). The experimental groups are defined in Supplementary Table 1 .

In the second part, combined treatment consisted of peritumoral (p.t.) IL-12 GET followed by CaEP (250 mM calcium solution) or ECT. Additionally, a plasmid was mixed with either calcium (final concentration 250 mM calcium and 20 μg of plasmid DNA) or bleomycin solution (final concentration 1.4 mM bleomycin and 20 μg of plasmid DNA) immediately before the i.t. injection and subsequent application of EP with the same parameters as above ( Figure 1 ). First, p.t. GET was performed using a noninvasive multielectrode array (MEA) with a circular distribution of electrode pins (Iskra Medical). Electric pulses (12 pulses, 170 V/cm, duration 150 ms, frequency 2.82 Hz) were delivered with an electric pulse generator Cliniporator (IGEA) as previously described (43) after p.t. injection of 4×20 μl of pIL-12 (20 μg in total). After 5 min, i.t. treatment was performed. The experimental groups are defined in Table 1 .

After the therapy, tumor growth was monitored three times per week by Vernier caliper. The volumes were calculated using the equation V=a×b×c×π/6, where a, b and c represent perpendicular diameters of the tumor. The mice that were tumor-free for 100 days (complete response) were rechallenged with a subcutaneous injection of the same tumor cells in the opposite (left) flank and monitored for tumor outgrowth.

On day three after the treatment, three mice from groups 1, 6, 7, 11, 13, and 26 ( Supplementary Table 4 ) were sacrificed, and tumors with skin were excised. Tumors were first submerged in 4% paraformaldehyde for 16 h and then incubated in 30% sucrose for 24 h before embedding in OCT medium and snap-frozen in liquid nitrogen. Blocks were stored at -20°C until consecutive frozen sections were cut using a Leica CM1850 cryostat at a thickness of 14 μm in a direction perpendicular to the skin. Slides were dried at 37°C for 10 min, washed for 5 min in phosphate buffered saline (PBS), and then antigen retrieval was performed by immersing the slides in hot citrate buffer (approx. 95°C; 10 mM sodium citrate, 0,05% Tween 20, pH 6.0) and allowed to stand for 1 h at the bench to cool. Slides were then rinsed in two changes of PBS; all subsequent steps were performed in a light-protected humidified chamber. Sections were incubated in permeabilization blocking buffer (5% donkey serum, 0.5% Triton X-100, 300 mM glycine in PBS) for 30 min at room temperature and then rinsed by immersing in PBS for 5 min, followed by 1 h incubation with blocking buffer (5% donkey serum, 300 mM glycine in PBS). Slides were incubated overnight with primary antibodies ( Supplementary Table 2 ) in blocking buffer (2% donkey serum, 300 mM glycine in 1x PBS) at 4°C. The next day, the sections were washed 3 times for 10 min with PBS and then incubated with secondary antibodies ( Supplementary Table 2 ) for 1 h at room temperature in blocking buffer and washed 3 times with PBS. Nuclear counterstaining was performed by incubation with Hoechst 33342 solution (3 μg/ml in PBS) for 10 min, washed 3 times with PBS, and mounted with Prolong™ Glass Antifade Mountant (Thermo Fisher Scientific).

Three tumor samples per group were imaged with an LSM 800 confocal microscope (Zeiss) with a 20x objective (NA 0.8). Lasers with excitation wavelengths of 405 nm, 488 nm, 561 nm, and 640 nm were used to excite Hoechst 33342, Alexa Fluor 488, Cy3, and Alexa Fluor 647, respectively. For the capture of the emitted light, a gallium arsenide phosphide (GaAsP) detector was used combined with a variable dichroic and filters at channel-specific wavelengths: 410 – 545 nm (Hoechst 33342), 488 – 545 nm (Alexa Fluor 488), 565 – 620 nm (Cy3) and 645 – 700 nm (Alexa Fluor 647). The obtained images were visualized and analyzed with Imaris software (Bitplane). Based on the negative control, cutoff values were determined for each channel.

Statistical analysis was performed using GraphPad Prism 9 (GraphPad Software). The significance of viability data was determined by two-way ANOVA with a Holm Sidak multiple comparisons test (vs. control or EP cells). IC₅₀ values were calculated by nonlinear regression analysis. Experiments in vitro were performed in 3 biological and 8 technical replicates, and data are represented as the means ± SEM unless otherwise stated. The significance of Kaplan Meier survival curves was determined by the Log-rank test. Growth delay was presented with AM ± SE and complete responses were labeled with 100 days of growth delay. The significance of growth delay data was determined using Two-way ANOVA with a Holm Sidak multiple comparisons test. The significance of immunofluorescence data of frozen tissue sections was determined by one-way ANOVA with Dunnett’s multiple comparisons post hoc test. The mode of action (additivity, synergism, and antagonism) of treatments was evaluated by the method developed by Spector et al. (44). A P value of <0.05 was considered to be statistically significant (ns P≥0.05; *P 0.01 – 0.05; **P 0.001 – 0.01; ***P 0.0001 – 0.001; ****P <0.0001).

The cytotoxicity of CaEP or ECT/BLM on B16-F10 and 4T1 cells was first evaluated to determine the intrinsic sensitivity of the two cell lines to the treatments ( Figure 2 ). The exposure of cells to the increased extracellular calcium without EP did not affect the survival of cells ( Figure 2A ). In contrast, exposure to BLM without EP reduced cell survival at the highest concentrations used compared to nontreated cells ( Figure 2B ). When cells were exposed to EP in the presence of calcium, a dose-dependent decrease in cell viability was observed, with IC₅₀ values of 3.6×10^-3 M and 3.4×10^-3 M calcium for B16-F10 and 4T1 cells, respectively ( Figure 2A ). The same effect was observed after ECT with bleomycin, with IC₅₀ values of 4.5×10^-10 M and 3.7×10^-10 M bleomycin for B16-F10 and 4T1 cells, respectively ( Figure 2B ). There were no statistically relevant differences between the two cell lines.

The antitumor effectiveness of CaEP with 50 mM, 168 mM, and 250 mM calcium was tested in B16-F10 melanoma in C57Bl/6 mice and 4T1 mammary carcinoma in Balb/c mice. ECT with bleomycin was used as a positive control.

Treatment of B16-F10 melanoma with CaEP resulted in significantly prolonged survival of animals (P<0.05) in comparison to control or controls with calcium injection alone ( Figure 3C ). The GD of B16-F10 tumors treated with 50 mM, 168 mM, and 250 mM CaEP was 3.6 ± 0.8 days, 3.8 ± 0.7 days, and 5.9 ± 1.1 days, respectively ( Figure 3A ). Injection of 250 mM calcium solution alone resulted in a growth delay of 3.0 ± 0.8 days. The survival of mice treated with ECT with bleomycin was significantly higher, 20.7 ± 1.0 days, compared to Ca/EP ( Figure 3C ; Supplementary Table 3 ).

When 4T1 mammary carcinoma tumors were treated, somewhat different results were obtained. Survival of mice and GD was not much altered after injection of 50 mM and 168 mM calcium alone ( Figure 3D ); however, when 250 mM calcium solution was injected alone, it resulted in significantly prolonged GD of 27.6 ± 12.2 days, whereas after CaEP, it was 32.0 ± 10.0 days. Rather high variability in response to the highest calcium solution could be due to the leakage of the solution due to the injections into the very solid 4T1 tumors. CaEP with 50 mM and 168 mM calcium solutions caused GDs of 5.1 ± 0.9 days and 13.0 ± 4.0 days, respectively. ECT with bleomycin resulted in a GD of 30.9 ± 11.0 days, not significantly higher in comparison to Ca/EP with 250 mM calcium solutions ( Figure 3B, D ; Supplementary Table 3 ).

All treatments were well tolerated by the animals, as no treatment-related mortality or reduced body weight were observed. Morphologically, in both tumor models, edema of the treated area arose in all groups with EP and calcium with or without EP, which resolved until day 3 after the treatment. Concurrently, visible necrosis of tumors developed with the formation of a scab. Nonetheless, most tumors continued to grow from the rim or beneath the scab. Cured Balb/c mice were not resistant to secondary challenge ( Supplementary Table 4 ).

The contribution of IL-12 GET to either CaEP or ECT with bleomycin in two different settings was tested in both tumor models, B16-F10 melanoma and 4T1 carcinoma. The concentration of 250 mM calcium solution was chosen for this experiment, where the best response was obtained in the previous experiments evaluating different doses of calcium solutions. Here, IL-12 p.t. GET was performed before intratumoral treatment with CaEP or ECT. Next, the effectiveness of an intratumorally injected solution of mixed pIL-12 and calcium or BLM followed by EP was tested, and finally, IL-12 p.t. GET was added to this setting.

In B16-F10 tumors IL-12 p.t. GET alone did not affect tumor growth. However, IL-12 i.t. EP significantly improved the survival of mice, as up to 83% complete responses were obtained ( Figure 4A ). Similarly, after combinatorial p.t. and i.t. IL-12 GET up to 75% complete responses were achieved despite the double quantity of plasmid being used ( Figure 4A ; Supplementary Figure 1 , Supplementary Table 4). The survival of animals in combinatorial therapy of GET and CaEP or ECT with bleomycin was significantly lower ( Figure 4 ). When p.t. GET was performed before injection of calcium alone, some synergistic effect on tumor GD was observed ( Figure 4B ). A significant tumor growth delay of 6.0 ± 0.7 days was obtained after CaEP, which was further potentiated after p.t. GET, where growth delay was 27.0 ± 9.1 days and 18% complete responses were achieved ( Figure 4C ; Supplementary Table 4 ). CaEP with IL-12 i.t. had an additive effect as opposed to CaEP alone, with 8.9 ± 0.9 days of GD; however, IL-12 p.t. GET did not further enhance antitumoral effectiveness ( Figure 4D ). BLM/EP was more effective on B16-F10 tumors than CaEP, with 15.2 ± 0.8 days of tumor GD ( Figure 4F ). An additive effect was observed when IL-12 p.t. was injected before ECT/BLM with 8% complete responses; however, no tumor cures were achieved after IL-12 p.t. GET ( Figure 4F ; Supplementary Table 4 ). When ECT/BLM with IL-12 i.t. was performed, 17% complete responses were obtained, and additional IL-12 p.t. GET resulted in up to 42% tumor cure ( Figure 4G ; Supplementary Table 4 ).

To determine the effect of adjuvant IL-12 GET on 4T1 tumors, which are moderately immunogenic compared to poorly immunogenic B16-F10 tumors (43), the same treatment protocol was used. In this model, IL-12 i.t. GET significantly delayed tumor growth by 4.0 ± 1.5 days, whereas additional IL-12 p.t. GET achieved 42.9 ± 13.8 days of tumor GD and up to 42% complete responses ( Figure 5A ; Supplementary Figure 2 , Supplementary Table 4 ). In contrast to B16-F10 tumors, calcium injection alone resulted in 50% cures in 4T1 tumors ( Figure 5B ). When EP was added, no further potentiation was observed. Additionally, IL-12 p.t. GET have some additive effect on CaEP alone ( Figures 5C, D ). BLM/EP resulted in 24.2 ± 6.8 days of GD and an 8% survival rate ( Figures 5E, F ; Supplementary Table 4 ). When IL-12 p.t. GET was added to the treatment, efficiency was potentiated to 56.2 ± 12.3 days of GD and 50% tumor cures. BLM/EP with IL-12 i.t. resulted in no complete responses, with 21.4 ± 7.0 days of GD; however, IL-12 p.t. alone or with GET pulses resulted in 36% and 25% complete responses, respectively ( Figure 5G ; Supplementary Table 4 ). Again, cured mice were not resistant to secondary challenge ( Supplementary Table 4 ).

To determine if GET of a plasmid encoding IL-12 in combination with either CaEP or BLM/EP leads to the infiltration of immune cells in tumors or their proximity, frozen sections of tumors with surrounding skin were excised three days after the therapy and immunofluorescently stained for macrophages (surface expression of F4/80) ( Figures 6A , 7A ), CD4+ helper T lymphocytes ( Figures 6A , 7A ), NK cells (surface expression of NKp46) ( Figures 6B , 7B ), CD8+ T lymphocytes ( Figures 6B , 7B ), proliferation (Ki-67 nuclear expression) ( Figures 6C , 7C ) and blood vessels (expression of CD31) ( Figures 6C , 7C ).

The percentage of proliferating cells in the control groups was similar in both tumor models; however, B16-F10 tumors had greater proliferation potential. There was a significant decline in Ki-67+ cells after CaEP with IL-12 p.t. GET (group 13) in the B16-F10 tumor model, but no significant changes were observed in the other groups ( Figures 6C , 8A ). In 4T1 tumors ( Figure 7C ), there was marginally decreased Ki-67 in tumors after CaEP with IL-12 p.t. GET (group 13) and (BLM+IL-12) i.t. + EP + IL-12 p.t. GET (group 26) ( Figure 8C ). There was an insignificant increase in Ki-67+ cells after CaEP alone, as a pool of highly proliferating cells was located next to the necrotic area ( Figures 6C , 7C ). This is in accordance with observations of tumor growth after therapy, where tumors continue to grow from under or from the rim of the scab, which forms after treatment due to tumor necrosis.

There was a significant difference between the two tumor models in their vascularization, where the percentage of tumor vessel area was 7.7% for B16-F10 tumors ( Figure 8B ) and 19% for 4T1 tumors ( Figure 8D ). There are also morphological differences in the vasculature of the two tumor models; in B16-F10 tumors, the vessels are sparse and larger in diameter, whereas in 4T1 tumors, they are denser, smaller in diameter, and more evenly arranged ( Figures 6C , 7C ). In both tumor models, staining of vessels was predominantly in the tumor margin, while very little staining occurred in the tumor core. There was a significant increase in the percentage of vessel area in B16-F10 tumors after treatment with i.t. and p.t. IL-12 GET alone, but no changes after other treatments ( Figure 8B ).

The presence of all four stained immune cell populations was confirmed in both B16-F10 ( Figure 6 ) and 4T1 tumors ( Figure 7 ); however, in control tumors, populations of CD8+ cells and NKp46+ cells were very sparse ( Figures 6B , 7B ). In both tumor models, the F4/80+ area was mostly present in the tumor edge and the nearby skin, whereas CD4+ cells were predominantly found at the tumor edge ( Figures 6A , 7A ). After different treatments, there were no significant changes in the F4/80+ area or CD4+ cell count in both tumor models ( Figures 9A, D, E, H ).

In the B16-F10 tumor model, there was a significant increase in the CD8+ cell count after p.t. and i.t. IL-12 GET (group 6) ( Figure 9B ), whereas in the 4T1 model, only after CaEP with IL-12 p.t. GET ( Figure 9F ). The most significant modifications were found when interrogating populations of NK cells, where there was a significant increase in NKp46-positive area in both tumor models after CaEP with p.t. IL-12 GET (group 13) and ECT/BLM with i.t. and p.t. IL-12 GET (group 26) and was concentrated in the proximity of the necrotic area ( Figures 6B , 7B , 9C, G ).