Growth environment influences B16.F10 mouse melanoma cell response to gene electrotransfer

Abstract

We developed and characterized a 3D collagen hydrogel model for B16.F10 melanoma tumors. Cells in this 3D environment exhibited lower proliferation than cells in the conventional 2D culture environment. Interestingly, the basal expression levels of several genes varied when compared to conventionally grown cells. In each growth environment, a significant number of melanoma cells were transfected by plasmid electroporation (electrotransfer), although expression could only be ascertained on the surface of the 3D constructs. Cellular responses to plasmid entry as demonstrated by pro-inflammatory cytokine and chemokine upregulation varied based on the growth environment, as did the mRNA levels of several putative DNA-specific pattern recognition receptors (DNA sensors). Unexpectedly, when plasmid DNA was delivered while cells where attached in the 2D or 3D environments, the mRNAs of the DNA sensor p204 and the inflammatory mediator TNFα were regulated in cells receiving pulses only. However, we were unable to confirm coordinate upregulation of TNFα and p204 proteins. This study confirms that cell responses differ significantly based on their environment, and demonstrates the difficulty of extending experimental observations between cell environments.

1. Introduction

Non-viral nucleic acid delivery via electroporation (electrotransfer) has been confirmed in primary cells and cell lines in many variations of in vitro three-dimensional (3D) growth. For example, plasmid DNA electrotransfer has been performed to porcine granulosa cells in barium alginate capsules [1], human embryonic kidney cells in multicellular tumor spheroids [2], human colorectal carcinoma and mouse sarcoma by hanging drop method [3], Chinese hamster ovary cells in collagen hydrogels [4], primary human dermal fibroblasts in sheets [5], mouse melanoma cells in hydroxypropyl methylcellulose [6], and human breast cancer cell lines in crosslinked self-assembling peptide [7]. In one study, plasmid DNA delivery was demonstrated to human keratinocytes and mouse melanoma cells combined in microgravity spheroids [8]. RNA delivery has also been performed to mouse neural stem cells in neurospheres [9].

As with many studies in 3D models where cells are embedded in complex environments, the purpose of this study was to bridge in vitro studies, which are often performed in suspended cells, and in vivo studies. The use of 2D or 3D in vitro models aids animal welfare by replacing animal use. One goal of this study was to compare baseline characteristics of cells grown in different environments. In vitro, cells are traditionally electroporated in suspension, primarily in cuvettes. We compared this this method of electroporation with electroporation of cells in environments potentially more relevant to in vivo studies, specifically cells grown attached in two dimensions (2D) and cells grown embedded in 3D collagen hydrogels.

Transfection of different cell types with plasmid DNA is known to modulate endogenous intracellular nucleic-acid specific pattern recognition receptors, also known as DNA sensors, and to active their coordinate signaling cascades [6, 10–13]. These signaling proteins include pro-inflammatory cytokines and chemokines [14] which induce inflammation in vivo [15, 16]. Depending on the cell type and the level of stimulus, cell death can also be an outcome [17, 18]. Therefore, a second goal of this study was to test if these endogenous responses varied based on the cell’s environment during electrotransfer.

2. Materials and Methods

2.1. Cells

B16.F10 mouse melanoma cells (CR6475, American Type Culture Collection, Manassas, VA, USA) were cultured in McCoy’s medium (Corning, Thermo Fisher Scientific, USA) supplemented with 10% fetal bovine serum (Atlanta Biologicals, R&D Systems, Minneapolis, MN, USA) and 1% gentamycin (Atlanta Biologicals) in a 5% CO₂ humidified incubator at 37°C. The cells were authenticated by STR profiling and lack of interspecies contamination confirmed (2017, IDEXX Bioanalytics, Columbia, MO, USA) and routinely tested negative for mycoplasma infection (e-myco Valid Mycoplasma PCR Detection kit, Intron Biotechnology, Burlington, MA, USA).

2.2. Collagen hydrogel (3D) constructs

For each construct, 425 μl rat tail type I collagen and 50 μl neutralization buffer (Advanced BioMatrix, Carlsbad, CA) were combined at 4°C. This suspension was immediately mixed with 25 μl 4×10⁷ B16.F10 cells in media by gentle pipetting. This cell concentration was optimized in a preliminary study. The combination of collagen, neutralization buffer and cells were immediately transferred to a 12 mm diameter transwell with a 0.4 μm pore size polycarbonate membrane insert in a well of a 6 well plate (Costar 3407, Thermo Fisher Scientific, USA) and allowed to solidify 5–10 minutes at 37°C. Once solidified, 1 ml of medium was layered over the construct, while 6 ml of medium was pipetted into the well below. Constructs were incubated 48 hours with a daily medium change. The medium was then removed and each construct was gently dislodged from the transwell with a sterile scalpel, forceps, and spatula and transferred temporarily to a sterile petri dish containing medium before experiments were performed.

2.3. Immunohistochemistry (IHC)

Collagen hydrogels were fixed in 4% paraformaldehyde, embedded in paraffin, and blocks were sectioned (IDEXX). Cell proliferation was quantified using Ki67 immunofluorescent staining as previously described [19]. Images were captured with an Olympus BX51 microscope using Camera Software for DP80 (Olympus America Inc., Center Valley, PA, USA). 4’,6-Diamidino-2-phenylindole (DAPI, Akoya Biosciences, Marlborough, MA) and ifi204 (p204, Invitrogen, Thermo Fisher Scientific) staining was performed as previously described [20]. Images were captured with a fluorescence microscope (BZ-X700E, Keyence Corp., Itasca, IL, USA).

2.4. Plasmid DNA (pDNA) transfection

Plasmids encoding firefly luciferase (gWiz Luc, Aldevron, Fargo, ND, USA) or Katushka fluorescent protein (pTurboFP635-C, Evrogen, Moscow, Russia) driven by the CMV-IE promoter/enhancer were commercially prepared (Aldevron, Fargo, ND, USA or GeneWiz, South Plainfield, NJ, USA) and diluted to 2 mg/ml in physiological saline. Endotoxin levels were confirmed to be <100 EU/mg. The application of eight 5 ms pulses at a voltage to distance ratio of 600 V/cm and a frequency of 1 hertz immediately disintegrated the 3D constructs during pulsing. The constructs maintained structural integrity during the application of six 100 μs pulses at a voltage to distance ratio of 1300 V/cm and a frequency of 4 hertz. To maintain experimental consistency, this pulse protocol was used in all cell environments. Pulses were delivered using a legacy model ECM 380 (BTX Harvard Apparatus, Holliston, MA, USA) with variations of a classic plate electrode, cuvettes, a custom plate electrode, and a commercial tissue slice electrode. The pDNA concentration in each experiment was varied to normalize the ratio pDNA to the number of cells present. Suspension experiments were based on previous results, while the assays of attached cells were created de novo. In suspension, each 100 μl aliquot contained 2 × 10⁶ cells in 40 μg pDNA, creating a ratio of 20 pg pDNA/cell. The mixture was transferred into cuvettes, pulses applied, and the cells transferred to the wells of a 48 well plate. This ratio was increased for cells in the attached environments because we anticipated transfection to be less efficient. For experiments on attached cells, cells were seeded into 48 well plates and cultured for 48 hours. Medium was replaced with 250 μl medium containing 6.25 μg DNA layered over a nearly confluent well in a 48 well plate containing approximately 1×10⁵ cells. This created a ratio of 62.5 pg pDNA/cell. A propriety electrode [21] was inserted into each well and pulses applied. This electrode consisted of four perpendicular plates with a nonconductive stopper that fit into a well of a 48 well plate, allowing pulses to be applied to attached cells. For 3D constructs, after incubation, each construct containing approximately 2×10⁶ cells was transferred to a platinum electrode chamber within a petri dish containing 100 μl 0.9% saline containing 62.5 μg pDNA. A second layer of pDNA suspended in saline was applied, creating a sandwich and repeating the ratio of 62.5 pg/cell. Pulses were applied using a platinum electrode wand (Figure 1, BTX Harvard Apparatus, Holliston, MA).

Figure 1.

Petri dish platinum electrodes and chamber for tissue slices used for electroporation of 3D constructs. Photo Courtesy of BTX Harvard Bioscience.

2.5. Transfection efficiency

Twenty-four hours after transfection, intracellular Katushka fluorescent protein was quantified by flow cytometry (BD LSR II with FACSDiva software, BD Biosciences, USA). In brief, single cells were gated out using the doublets discrimination analysis technique based on FCS and/or SSC scatter plots. Katushka positive cells among all singlets were then identified and analyzed for their frequency.

2.6. Cytoplasmic reducing power as an indicator of viability

For the time course in Figure 2B, cells or 3D constructs were transferred to the wells of a 48-well plate in medium containing PrestoBlue (Invitrogen, Thermo Fisher Scientific). Plates were incubated 16 hours with hourly monitoring (excitation: 544 nm, emission: 620 nm, Fluostar Omega, BMG Labtech, Cary, NC, USA) and compared to uninoculated control wells. In Figures 3A and 3B, 24 hours after transfection, medium was replaced with medium containing PrestoBlue and assayed per the manufacturer’s protocol.

Figure 2.

B16.F10 cell growth and proliferation in 2D and 3D environments. (A) Image of collagen hydrogel. (B) Relative cytoplasmic reducing power with respect to background over 20 h in 2D (open circles) and 3D (closed circles) environments, *, p < 0.05; hours 0, 6, 7, 8, 9; n = 5 per group. (C) Percentage of proliferating cells expressing Ki67, ***, p < 0.001; Representative images of (D) 2D and (E) 3D models. Cell nuclei (DAPI) are shown in blue and Ki67 staining in green. Cells in monolayer are on the same plane while cells in 3D are dispersed three dimensionally, so differing numbers of cells are visible between the fields.

Figure 3. Relative baseline mRNA levels of cells in different environments.

Suspension, cells 4 hours after the preparation of the suspension; 2D, cells grown 48 hours in a 2D environment; 3D, cells grown 48 hours in a 3D environment. (A) DDX60, (B) IFNβ, (C) TNFα mRNA levels relative to samples grown in 3D. *, p<0.05; **, p<0.01 with respect to 3D cells. n=4–9 per group.

2.7. Viability and transfection efficiency in 3D constructs

Live/Dead staining was performed in 3D constructs since cells were held in place by the matrix and therefore would not be washed away with medium changes. To establish viability in 3D constructs, live cells were stained blue while dead cells were stained green using a ReadyProbes Cell Viability Kit (Molecular Probes, Thermo Fisher Scientific). Katushka expression was normalized to live cells to establish transfection efficiency. Images were captured with a BZ-X700E fluorescence microscope (Keyence Corp.). Cells were quantified using the Hybrid Cell Count analysis application.

2.8. Reverse transcription real time PCR

The baseline levels of five reference genes, β-actin (Actβ), β2 microglobulin (β2M), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), β-glucuronidase (Gusβ), and hypoxanthine phosphoribosyltransferase 1 (HPRT1) were quantified and compared between suspension cells and cells grown in collagen hydrogels. β2M and HPRT1 mRNA levels varied slightly but significantly, so GAPDH and Gusβ were used as reference genes for the studies described here. Messenger RNAs were quantified four hours after transfection as previously described [6, 10] and detailed in Supplementary Table 1.

2.9. ELISAs

ELISAs were performed on medium collected 4 and 24 hours after transfection per manufacturer’s instructions (Invitrogen, Thermo Fisher Scientific and R&D Systems, Minneapolis, MN, USA).

2.10. Statistical analysis

The differences between the experimental groups were statistically evaluated by two-way analysis of variance followed by Dunnett’s post-test (GraphPad Software, La Jolla, CA, USA). Student’s T test was used to evaluate experiments with two groups. A p-value of less than 0.05 was considered to be statistically significant. A sample size (n) for each experiment represents biological replicates unless otherwise stated.

3. Results and Discussion

3.1. 3D collagen hydrogel constructs

A representative 3D construct is shown in Figure 2A. Initially, the growth characteristics of cells grown in 2D and 3D environments were compared. The rate of cytoplasmic reduction was slower in the 3D compared to the 2D environment (Figure 2B), but eventually reached the same level. Cell proliferation as measured by Ki67 staining was also significantly lower in the 3D environment when compared to the 2D environment (Figure 2C). Representative images of Ki67 stained cells are shown for 2D and 3D models (Figure 2D and 2E, respectively). Differing numbers of cells are visible within each image because 2D cells are visualized in a monolayer while the cells in the constructs are dispersed three dimensionally so fewer cells are visible in each section.

3.2. Baseline mRNA expression

To determine if baseline mRNA expression in cells varied based on differences in the cell’s environment, the levels of several mRNAs were compared between naïve cells in the 2D and 3D environments and in cells that had been suspended 4 hours previously. The mRNA levels of a putative DNA sensor, DEAD box helicase 60 (DDX60), were increased more than 15-fold in recently suspended cells than in cells attached in either 2D or 3D environments (Figure 3A). This may implicate some kind of activation in response to manipulation. Interferon β (IFNβ) levels were increased in 2D attached cells by approximately 20-fold over 3D or suspension cells (Figure 3B). It is difficult to hypothesize as to why growth in the 2D environment induced IFNβ mRNA. Finally, the mRNA levels of the pro-inflammatory mediator tumor necrosis factor α (TNFα) were increased in both suspension and 2D when compared to cells in the 3D environment (Figure 3C), which may more closely simulate the in vivo environment.

In an oligonucleotide microarray study, growth of a human metastatic melanoma cell line in multicellular tumor spheroids resulted in an increase in expression of >100 transcripts and a decrease of 73 transcripts [22]. We tested the levels of two of these upregulated transcripts, interferon regulatory factor 7 (IRF7, 3.07-fold) and superoxide dismutase 2 (SOD2, 5.11–8.11-fold). In our 3D model, IRF7 was not upregulated, while SOD2 was slightly, but significantly, downregulated (data not shown). Thioredoxin interacting protein (TXNIP) was among a number of mRNAs upregulated in a 3D model using either a human lung carcinoma or a squamous cell carcinoma cell line [23]. We did not detect TXNIP mRNA regulation in this mouse melanoma model. The differences between these studies may be accounted for by the difference in species, cell line, and methods of 3D growth.

3.3. Transfection efficiency and viability

We next demonstrated that cells were efficiently transfected whether pulses were applied to cells in suspension, while attached in 2D, or during 3D growth (Figure 4). This demonstration confirmed previous work in many cell types and multiple 3D models [1–8]. After transfection of cells suspended and immersed in pDNA in solution, the transfection efficiency was approximately 54% (Figure 4A). When cells were attached during transfection, only the upper surface of the cells was exposed to pDNA; a transfection efficiency of approximately 20% was observed (Figure 4B). No significant differences in cytoplasmic reducing power were seen between the groups (Figures 4C and 4D).

Figure 4. Katushka reporter expression and cytoplasmic reducing power as an indicator of viability 24 hours after transfection.

Transfection efficiency by flow cytometry of cells transfected with pTurboFP635-C (A) in suspension; (B) while attached in 2D. Cytoplasmic reducing power of cells transfected (C) in suspension and (D) while attached in 2D. **, p<0.01 with respect to control cells. n=3–6 per group.

Microscopy was used to assess viability and transfection efficiency in 3D collagen hydrogels (Figure 5). While pDNA application did not significantly affect viability, pulse delivery alone significantly decreased viability. In this format, pDNA was “sandwiched” around each construct so, while all cells were pulsed, pDNA may not have permeated the construct. Each construct was approximately 4 mm thick, limiting the imaging to the lower surface. These factors may be responsible for the significant but minimal cell transfection efficiency of approximately 10%.

Figure 5. Cell viability and reporter expression in 3D constructs.

Representative images of A, Control constructs; B, Pulse application only; C, pTurboFP635-C only; D, pTurboFP635-C combined with pulse application. Live cells are shown in blue, dead cells are shown in green, Katushka reporter expression is shown in red. Scale bar, 1000 μm. E, Quantification of viability (Dead, green; Live, blue); F, Transfection efficiency. **, p<0.01 with respect to control cells.

3.4. Expression of proinflammatory cytokines and chemokines

We previously observed the upregulation of the mRNAs and proteins of several proinflammatory cytokines and chemokines after transfection of B16.F10 cells and other cells types [6, 10–13] immediately after pDNA transfection. Some aspects of this upregulation were confirmed here in each environment (Table 1). IFNβ mRNA was upregulated in suspension cells and cells in a 3D environment. Interestingly, upregulation in the 2D environment was not observed, but this may be due to the high 2D baseline levels (Figure 3). TNFα mRNA was upregulated after pDNA electrotransfer in all environments. This represented an additional increase over the high baseline levels observed in cells in suspension and in 2D attached cells (Figure 3). Surprisingly, we found that TNFα mRNA was significantly upregulated by pulse application alone in the 2D and 3D samples. It was not necessary to introduce exogenous pDNA for this effect. C-X-C motif chemokine ligand 10 (CXCL10) is an IFN-inducible chemokine and therefore upregulation amplified by IFN signaling. We detected significant CXCL10 mRNA upregulation in all groups receiving pDNA with pulse application.

Table 1.

Fold changes in cytokine and chemokine mRNA levels 4 hours after pDNA electrotransfer. pDNA, gWiz Luc; EP, electrotransfer. n=4 per group.

	Transfected in suspension	Transfected attached	Transfected in 3D constructs
IFNβ
Control	1.0 ± 0.0	1.0 ± 0.0	1.0 ± 0.0
EP	2.3 ± 1.0	1.1 ± 0.4	2.9 ± 1.4
pDNA	1.5 ± 1.1	1.6 ± 0.6	1.3 ± 1.0
pDNA+EP	83.2 ± 41.5**	1.8 ± 0.6	5.7 ± 1.8**
TNFα
Control	1.0 ± 0.0	1.0 ± 0.0	1.0 ± 0.0
EP	1.5 ± 0.2	2.7 ± 0.3**	8.4 ± 1.7**
pDNA	1.0 ± 0.0	1.1 ± 0.2	2.9 ± 1.6
pDNA+EP	2.1 ± 0.5*	2.9 ± 0.1**	7.8 ± 1.1**
CXCL10
Control	1.0 ± 0.0	1.0 ± 0.0	1.0 ± 0.0
EP	2.1 ± 1.1	0.7 ± 0.0	1.0 ± 0.1
pDNA	1.1 ± 0.1	1.2 ± 0.1	1.1 ± 0.1
pDNA+EP	191.6 ± 61.9**	15.8 ± 0.9**	8.6 ± 1.5**

CXCL10 protein secretion reflected mRNA upregulation in each environment four and 24 hours after pDNA electrotransfer (Figure 6). However, these levels varied significantly based on both the time point and the cell environment. CXCL10 secretion by cells transfected in suspension was higher at the four hour than the 24 hour time point. Cells in this environment responded to pDNA electrotransfer with the highest secretion levels. Cells transfected while attached, whether in a 2D or 3D environment, also secreted significant levels of CXCL10. Secretion by cells in the 3D environment was much lower than the other environments. This could be potentially due to lower relative transfection efficiency (Figures 4 and 5). TNFα protein secretion at four and 24 hours remained below the ELISA assay detection limit at both time points (data not shown). This is not unexpected; due to processes occurring after mRNA is synthesized, protein levels may be dissimilar [24, 25].

Figure 6. Quantification of CXCL10 in medium by ELISA four and 24 hours after transfection with gWizLuc.

Cells were transfected (A) in suspension; (B) while attached 2D; (C) while attached 3D. Control, naïve cells; EP, Pulse application only; pDNA, pTurboFP635-C exposure only; pDNA+EP, pTurboFP635-C combined with pulse application. *, p<0.05; **, p<0.01 with respect to control cells. n=3–4 per group.

3.5. Expression of DNA-specific pattern recognition receptors

The mRNAs of the DNA sensors DNA-dependent activator of interferon regulatory factors/Z-DNA binding protein 1 (DAI/ZPB1), DDX60, and interferon activated gene 204 protein (p204) were upregulated in response to pDNA electrotransfer in several cell types [6, 10–13]. This upregulation was observed in each cell environment at varying levels (Table 2), implying that these proteins may have an active role in sensing pDNA as it enters the cell. Since DDX60 mRNA levels were significantly higher at baseline in suspension cells than in the other cell environments (Figure 2), this represented a further upregulation in this group. Although we previously observed that the mRNA upregulation of DAI/ZBP1 and p204 did not parallel transfection efficiency [10], in the experiments described here there appears to be some correlation.

Table 2.

Fold changes in DNA sensor mRNA levels 4 hours after pDNA electrotransfer. pDNA, gWiz Luc; EP, electrotransfer. n=4 per group.

	Transfected in suspension	Transfected attached	Transfected in 3D constructs
DAI/ZBP1
Control	1.0 ± 0.0	1.0 ± 0.0	1.0 ± 0.0
EP	1.5 ± 0.4	0.5 ± 0.1	1.0 ± 0.2
pDNA	1.1 ± 0.1	1.1 ± 0.4	0.8 ± 0.2
pDNA+EP	136.8 ± 24.4**	2.0 ± 0.7*	2.5 ± 0.6*
DDX60
Control	1.0 ± 0.0	1.0 ± 0.0	1.0 ± 0.0
EP	2.29 ± 1.4	0.5 ± 0.1	1.7 ± 1.0
pDNA	1.0 ± 0.7	2.0 ± 0.5	2.1 ± 0.8
pDNA+EP	215.9 ± 68.5**	33.9 ± 6.9**	8.0 ± 2.0**
p204
Control	1.0 ± 0.0	1.0 ± 0.0	1.0 ± 0.0
EP	2.2 ± 0.4	3.3 ± 0.3**	5.2 ± 1.0**
pDNA	0.9 ± 0.1	1.32 ± 0.0	1.7 ± 0.5
pDNA+EP	24.8 ± 7.8**	4.6 ± 0.3**	3.5 ± 1.0**

While upregulation of DNA sensors may be expected in response to the introduction of exogenous DNA, the mRNA of the DNA sensor p204, the mouse ortholog of human IFI16, was upregulated after pulse application alone in some groups without the introduction of exogenous pDNA. In suspended cells, p204 was upregulated only in the presence of exogenous pDNA. In attached cells, whether in a 2D or 3D environment, p204 mRNA was upregulated by pulses alone. However, this mRNA upregulation was minimal. The p204 protein is expressed at a basal level (Figure 7A) and we were unable to confirm coordinate protein upregulation of in any group by IHC (Figure 7B–D). Protein levels do not necessarily mirror mRNA levels [24, 25].

Figure 7. Representative images of p204 immunohistochemistry in 3D constructs.

(A) Control (B) Pulse application only (C) gWizLuc exposure only (D) gWizLuc combined with pulse application. Cell nuclei are shown in blue and ifi204/p204 staining in red.

TNFα and p204 mRNAs were upregulated in pulsed cells while attached but not in suspension. This effect may be due to the cell environment during pulse application. Cells in suspension are in constant movement due to Brownian or thermal motion and gravitational setting. Therefore, they are not in a fixed position between pulses. The potential for repeated pulse applications at a single point is minimal, so pulse effects on organelles are not likely and no DNA sensor activity is detected. This is confirmed by the fact that no indications of DNA sensor activity are observed after pulse application to hydroxymethylcellulose spheroids in suspension [6]. However, whether in a 2D or 3D environment, repeated pulses applied to cells held stationary with respect to the electric field may target the same point on the plasma membrane. Repeat pulsing to the same site may permeabilize the cell’s organelles [26], including the nucleus and mitochondria, which contain DNA. Therefore, organelle damage may cause the release of genomic or mitochondrial DNA, making this DNA available to DNA sensing, and p204 may be the most sensitive of the DNA sensors tested to this effect. Although TNFα is primarily expressed by immune cells, this protein can be secreted at varying levels by B16 cells [27]. Increased TNFα expression can be associated with DNA sensing [28]. However, we were unable to confirm the coordinate upregulation of TNFα and p204 mRNAs by TNFα and p204 proteins, which emphasizes the importance of confirming mRNA quantification with protein quantification.

4. Conclusions

We previously demonstrated that different pulse protocols induce varying responses to DNA entry after electrotransfer [6, 10, 12–14, 29, 30]. We also demonstrated cell and tissue responses vary based on plasmid properties [11, 29]. DNA specific PRRs are upregulated in both tumor [6, 10–12] and non-tumor [13] cells in response to DNA electrotransfer. Here, we developed a structurally stable 3D collagen hydrogel in which B16.F10 mouse melanoma cells were embedded in an extracellular matrix, potentially a better model of the in vivo cell environment than suspended cells or cells in traditional 2D culture. We compared cells in these environments and demonstrated differences in proliferation, transfection levels, baseline and induced gene expression, and protein expression. Overall, this study demonstrates that there are detectable differences between cells in different environments and in their biological responses, making generalizations from experiments in vitro to the in vivo environment more uncertain.

A nucleic acid is an obvious requirement for gene therapies. The inflammation associated with both viral [31] and non-viral [32] gene therapies may be caused by the activation of nucleic acid-PRRs. This outcome is generally a benefit in inducing vaccine immunogenicity or augment cancer therapies; however, it may be detrimental to other therapies, for example protein replacement or gene correction therapies. The results of this study may be relevant to better gene therapy design.

Highlights.

• A 3D B16F10 mouse melanoma cell collagen hydrogel model was developed

• Reporter protein expression was detected after plasmid electrotransfer

• Changes in mRNA and protein expression were detected after transfection

• mRNA and protein expression varied based on cell environment during transfection ss