mRNAs: Anti-reverse cap analogue (ARCA) mRNA encoding enhanced green fluorescent protein (EGFP), i.e., EGFP-mRNA (5-moUTP), ARCA Cy5-labeled EGFP-mRNA (5mo-UTP), and CleanCap mCherry-mRNA (5mo-UTP) were purchased from APExBIO Technology (Houston, TX, USA), and CleanCap Fluc-mRNA, i.e., luciferase-mRNA (5moU) from TriLink Biotechnologies (San Diego, CA, USA). The mRNA stock solutions (1 mg/mL) were maintained at −80°C.

PF14 (stearyl-AGYLLGKLLOOLAAAALOOLL-NH₂) (PepScan, Lelystad, Netherlands and PepMic, Shanghai, China). 1 mM peptide stock solution was prepared in the mixture of ethanol, DMSO, and trimethylene carbonate (TMC) (90:9.6:0.4 v:v:v) and stored at −20°C. Luciferase substrate XenoLight D-Luciferin – K⁺ salt (LH2) was obtained from PerkinElmer (Waltham, MA, USA), CaCl₂ and MgCl₂ from Thermo Fisher Scientific (Waltham, MA, USA), PS80 from VWR (Radnor, PA, USA) and chloroquine from Applichem (Darmstadt, Germany). The cell lysis buffer contained 1% (v:v) Triton X-100, 50 mM Tris (pH 7.5), 150 mM NaCl, and 1 mM EDTA. Luciferase substrate solution contained 1 mM LH2, 25 mM DTT, 1 mM ATP, 25 μM coenzyme A, 5 mM MgSO₄, 20 mM tricine, 1 mM EDTA, and 1 mM MgCO₃.

HeLa (Expacy. Cellosaurus HeLa (CVCL_0030), n.d.) and human immortalized keratinocyte cell line HaCaT (Expacy. Cellosaurus HaCaT (CVCL_0038), n.d.) were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Corning, NY, USA) containing 4.5 g/L of glucose, 10% (v:v) of fetal bovine serum (FBS) and 1% (v:v) of penicillin-streptomycin solution (100 U/mL penicillin and 100 μg/mL streptomycin). Cells were grown on 6 cm cell culture dishes (Corning) at 37°C in a humid atmosphere that contained 5% of CO₂ and were split every other day using trypsin-EDTA solution (Corning). For washing cells, Dulbecco’s Phosphate-Buffered Saline (DPBS) without calcium and magnesium was used. For experiments, cells were plated on 96-well plates (Thermo Fisher Scientific) at 1 × 10⁴ cells per well.

Pooled primary human epidermal keratinocytes (Promocell, Heidelberg, Germany) were cultured as previously described (Hermann et al., 2017). For the experiments, 4 × 10⁴ and 1 × 10⁴ cells were seeded in 24-well plates and 96-well plates, respectively, using a keratinocyte serum-free medium (KC-SFM) supplemented with recombinant epidermal growth factor (EGF), bovine pituitary extract (Gibco, Waltham, MA, USA), and antibiotic solution (5 U/mL penicillin and 5 μg/mL streptomycin).

For the formation of nanoparticles, PF14 was mixed with mRNA at a charge ratio (CR) 2:1 in Milli-Q (MQ) water in 1/10th of the final volume if not stated otherwise. The final concentration of mRNA was 0.8 ng/μL and PF14 was 1 μM. After 15 min incubation at room temperature (RT), CaCl₂, MgCl₂, PS80, chloroquine, or their combinations were added if indicated, followed by 15 more min of incubation. After formation, nanoparticle-containing solutions (1/10th of the final volume) were added to the cell medium (9/10th of the final volume) and cells were incubated with the solutions at 37°C for 24 h, if not stated otherwise. All of the transfections were performed with cells plated either in 96- or 24-well plates with density of 50%–70%. A growth medium containing 1/10th (v:v) of MQ water was used as a negative control solution in all in vitro experiments.

Before the transfection, cells were pre-incubated for 30 min at 37°C with growth medium (9/10th of the final volume) that contained inhibitors of endocytosis at the following concentrations: 10 μM chlorpromazine, 50 μM nystatin, and 20 μM 5-(N-ethyl-N-isopropyl)-amiloride (EIPA). After pre-incubation, nanoparticle-containing solutions (1/10th of the final volume) were added and cells were incubated with inhibitors and nanoparticles for 4 h at 37°C. After this, solutions were replaced with a fresh growth medium, and the effects were measured after an additional 20 h of incubation.

Cells were transfected with luciferase-mRNA for 24 h, then washed with PBS. To lyse the cells, 20 μL of cell lysis buffer was added and samples were incubated at −20°C until the solutions froze. Once thawed at RT, 70 μL of luciferase substrate solution was added per well, and the solutions were transferred to a white 96-well plate (Perkin Elmer, Waltham, MA, USA). The luminescence intensity was measured as previously described (Helmfors et al., 2015; Rocha et al., 2016) using the GloMAX 96 microplate luminometer (Promega, Madison, WI, USA).

The cells were transfected with nanoparticles of PF14 with mCherry-mRNA, EGFP-mRNA, or Cy5-mRNA for 24 h as in the protocol outlined in subsection 2.3. The cells were then detached by trypsinization, and the suspensions were diluted with a PBS buffer containing 0.5% BSA and 1 mM EDTA. Then, the plate was placed on ice. Next, cell nuclei were stained with 0.5 μg/mL of 6-diamidino-2-phenylindole (DAPI) in PBS buffer for 5 min to distinguish between live and dead cells. The flow cytometry analysis was conducted using an Attune NxT flow cytometer (Thermo Fisher Scientific) and 5000 events were analyzed for each well.

After 24 h of transfection with PF14-mRNA nanoparticles, cell viability was measured spectrophotometrically using the WST-8 assay kit (APExBIO Technologies) to determine the cytotoxicity of nanoparticles. Briefly, 10 µL of WST-8 solution per well was added after 24 h of transfection, and cells were incubated at 37°C for 3 h. Finally, the absorbance of the resultant solutions was measured at 450 nm, using an Infinite M200 PRO (Tecan, Männedorf, Switzerland) microplate reader. The percentage of viable cells was calculated using the following formula: c e l l   v i a b i l i t y   % = a b s o r b a n c e   o f   t e s t   c o m p o u n d − a b s o r b a n c e   o f   b l a n k a b s o r b a n c e   o f   c o n t r o l − a b s o r b a n c e   o f   b l a n k   ∗ 100

Nanoparticles of mRNA and PF14 were prepared as described above and diluted 1:100 in MQ water (with a final volume of 1 mL). The hydrodynamic diameter and zeta potential of the nanoparticles were measured using the ZetaSizer Nano ZSP instrument (Malvern Panalytical, Malvern, UK).

Cells were plated in a 24-well plate on cover glasses of 12 mm diameter (Menzel-Gläser, Braunschweig, Germany) 24 h prior to the transfection, which was carried out as described in subsection 2.3. After 24 h transfection, cells were washed with PBS and fixed by incubation with 4% paraformaldehyde (PFA) for 15 min at RT. Next, the cell nuclei were stained with 5 μg/mL of DAPI solution in PBS for 10 min at RT. Afterwards, the cells were washed with PBS and mounted onto glass slides with DAKO fluorescence mounting media (Dako Products, Agilent Technologies, CA, USA). For each condition, three specimens were prepared and at least three images per specimen were obtained. The images were taken using a confocal laser scanning microscope Olympus FV1200MPE (Olympus Corporation, Tokyo, Japan) with ×10 objective and ×60 oil objective at 1024 × 1024-pixel resolution. The images were captured at the wavelengths suitable for detection of DAPI (λex/em = 405/461 nm), EGFP (λex/em = 473/510 nm), and Cy5 (λex/em = 635/664 nm).

Mice were maintained in the animal facility at the Institute of Biomedicine and Translational Medicine, University of Tartu, according to the institute’s regulations. Six-to-ten-week-old C57BL/6J mice were used in the experiments. All animal experiments were performed according to permission by the Animal Ethics Committee at the Estonian Ministry of Agriculture (license No. 117, 2018; No. 158, 2020). For subcutaneous injection, nanoparticles were formed initially by combining 25.6 µM PF14 and 42.5 ng/μL of EGFP-mRNA or EGFP encoding mRNA labeled with Cy5 in MQ water for 15 min at RT. Subsequently, if indicated, 3 mM CaCl₂, or 0.05 mg/mL PS80 were added, followed by an additional 15 min incubation. Prior to injection, the resulting complexes were mixed with a 5% glucose solution. As a control, 20 µL of 5% glucose in PBS was used. To induce ICD, 20 μL of 0.2% phorbol 12-myristate 13-acetate (PMA) in acetone was applied topically. Ear thickness was measured with a Vernier digital caliper (Mitutoyo, Kawasaki, Japan) as previously described (Carreras-Badosa et al., 2020). For immunofluorescence analyses, mouse ears were fixed in 4% PFA for 24 h at 4°C. Afterwards, they were incubated in 30% sucrose for additional 24 h at 4°C before being embedded in Neg-50 cryo-medium tissue tek (Thermo Fisher Scientific). To perform confocal imaging, 10 μm thick sections of mouse ears were obtained from a cryotome. These sections were then permeabilized with 0.3% Triton X-100 in PBS for 10 min and incubated with 2% goat serum for 30 min at RT. The mouse ears were then stained with 10 μg/mL of EGFP polyclonal primary antibody (CAB4211, Thermo Fisher Scientific) followed by 10 μg/mL goat anti-rabbit IgG secondary antibody conjugated with Alexa Fluor 647 (A21245, Thermo Fisher Scientific). Specimens were also counterstained with DAPI and mounted with a fluorescent mounting medium (Dako Products). Finally, each slide was examined using an Olympus FV1200MPE confocal microscope (Olympus corporation).

Total RNA was purified using miRNeasy mini kit (Qiagen, Venlo, Netherland) following the manufacturer’s protocol. The RNA concentration and quality were evaluated with a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific). Subsequently, cDNA was synthesized from 400 to 600 ng of RNA using oligo-dT (TAG Copenhagen, Denmark), and RevertAid reverse transcriptase (Thermo Fisher Scientific). Quantitative PCR was performed on a Viia7 Real-time PCR system (Life Technologies, Carlsbad, California, USA) using 5× HOT FIREPol EvaGreen mix (Solis BioDyne, Tartu, Estonia). To normalize gene expression, HPRT was used, and ΔΔCt method was employed to calculate relative changes in gene expression.

Statistical analyses were performed using Prism 9.5.1 software (GraphPad Software, San Diego, CA, USA). Either one-way or two-way ANOVA test was used to compare mean values between studied groups or conditions, and the appropriate correction method was employed. In particular, Tukey’s multiple correction method was used when every mean was compared with every other mean, and Šidák multiple correction method was utilized for the comparison of selected sets of means, as indicated in the figure legends. The results were considered significant at p < 0.05.

As the transport efficiency and safety of nanoparticles largely depend on the physicochemical characteristics, it is essential to optimize their size, charge, and chemistry (Desai, 2012; Sabourian et al., 2020). In addition, knowing the impact of particle size and charge on the uptake process, nanoparticles can be redesigned to accumulate maximally in the target areas with minimal side effects. Therefore, we first tested the effect of different additives on the size and charge of the PF14-mRNA nanoparticles using DLS analysis. For that, we formed nanoparticles of PF14 with four different mRNAs: mCherry-mRNA, luciferase-mRNA, EGFP-mRNA, and Cy5-labeled EGFP-mRNA (hereafter referred as Cy5-mRNA), using a charge ratio of 2:1 PF14:mRNA. As additives, PS80, chloroquine, CaCl₂, and MgCl₂ were added to the nanoparticles.

DLS measurements revealed that the average hydrodynamic diameters of the particles formed by PF14 with mCherry, luciferase, EGFP and Cy5-mRNA were 300.1 nm, 146.6 nm, 123.1 nm, and 108.5 nm, respectively. This indicates that the different mRNAs, although having the same length, can have a slight influence on nanoparticle size (Figure 1A; Supplementary Table S1). Notably, PF14-EGFP-mRNA nanoparticles had a slightly higher zeta potential (40.3 mV) compared to those formed with mCherry (32.2 mV) and Cy5-mRNA (30.1 mV) as depicted in Figure 1B. Still, there were no significant changes in particle size when different additives were used for the formation of PF14-mRNA nanoparticles (Figure 1A). Furthermore, addition of CaCl₂ to PF14-luciferase-mRNA (25.1 mV) and PS80 to PF14-EGFP-mRNA (31.6 mV) resulted in reduced zeta potential, while the addition of MgCl₂ to Cy5-mRNA nanoparticles led to a mild increase in zeta potential (43.2 mV) (Figure 1B; Supplementary Table S1). Our results also showed that when PF14 was absent, all tested mRNAs exhibited a negative charge, with values ranging from −9.4 mV to −25.6 mV. However, when PF14 was added, all resulting nanoparticles had a positive charge, with values ranging from 25.1 mV to 43.2 mV, as shown in Figure 1B and summarized in Supplementary Table S1. Overall, these results demonstrate that all of the formed nanoparticles possessed suitable characteristics for further experiments.

Next, we assessed whether the PF14-mRNA nanoparticles formed with luciferase-mRNA and mCherry-mRNA in the presence and absence of additives induced the expression of reporter proteins in a cancer-derived cell line HeLa and immortalized keratinocyte cell line HaCaT. In addition, we investigated the impact of nanoparticles on the viability of these two cell lines. All of the nanoparticles induced protein expression to a certain extent, as seen in Figures 2A–D. However, among the tested compounds, only CaCl₂ significantly enhanced the expression of both luciferase (Figures 2A, B) and mCherry (Figures 2C, D) in both tested cell lines. Notably, the addition of PS80, chloroquine, and MgCl₂ to the nanoparticles did not affect the expression of the reporter protein in either cell line. While none of the nanoparticles were significantly toxic to HeLa cells (Supplementary Figure S1A), PS80 considerably affected the viability of HaCaT cells (Supplementary Figure S1B). In summary, PF14 effectively delivers mRNA into cell lines, and addition of CaCl₂ into the nanoparticles enhances their effect in HeLa and HaCaT cells without affecting cell viability.

We next examined the transport efficiency of PF14-Cy5-mRNA nanoparticles into primary human keratinocytes using confocal microscopy. As depicted in Figure 3A and Supplementary Figure S2, after 24 h, all nanoparticles were capable of transporting Cy5-mRNA into the cytoplasm of keratinocytes. Due to its reported ability to induce keratinocyte differentiation (Xie et al., 2005; Mahanty et al., 2019), CaCl₂ was excluded from experiments with keratinocytes. Quantitation analysis of images revealed that none of the additives significantly enhanced or reduced the delivery (Figure 3B). However, we observed a less efficient delivery when PS80 or MgCl₂ were added, as measured by the Cy5 signal alone. Furthermore, the EGFP signal was relatively weak in most of the conditions, indicating inefficient translational of Cy5-mRNA in the keratinocytes upon application of the PF14-Cy5-mRNA nanoparticles (Figure 3A). In conclusion, our results demonstrate that all tested combinations of additives exhibited similar levels of PF14-Cy5-mRNA transportation into primary human keratinocytes.

Next, we used EGFP-mRNA for formation of similar nanoparticles. Our results demonstrate that transfection of all PF14-based EGFP-mRNA nanoparticles led to the EGFP expression in keratinocytes (Figures 4A, B). Interestingly, among additives, PS80 significantly increased the expression of the reporter protein in keratinocytes (Figure 4B; Supplementary Figure S3), which was further confirmed with the use of luciferase-mRNA (Figure 4C). In addition, our flow cytometry data also demonstrated that PS80 greatly increased the EGFP expression after 24 h of transfection, as shown in Supplementary Figure S3B. Importantly, the addition of PS80 did not alter the fluorescence intensity of Cy5, indicating a specific effect on reporter protein expression after the delivery (Supplementary Figure S3B). However, the inclusion of MgCl₂ and chloroquine did not appear to enhance the efficiency of PF14-EGFP-mRNA nanoparticles (Figures 4B, C). It should be noted that the nanoparticles containing PS80 had a substantial impact on cell viability compared to the PF14-mRNA nanoparticles alone in both keratinocytes (Supplementary Figure S1C) and in HaCaT cells (Supplementary Figure S1B). In addition, we observed that naked Cy5-mRNA and PF14-mRNA nanoparticles reduced the viability of keratinocytes depending of the additives included, while HaCaT cells were less affected, and HeLa cells showed no influence. These results collectively demonstrate that primary human keratinocytes can be successfully transfected with PF14-mRNA nanoparticles. However, PF14-mRNA nanoparticles have a greater impact on the viability of primary human keratinocytes compared to cell lines.

Despite the uptake of CPPs has been reported for a wide range of cell types, the specific mechanism of internalization is not well understood and varies considerably depending on cell type, cargo, and the physicochemical features of nanoparticles (Ruseska and Zimmer, 2020). To investigate the internalization pathway of PF14-mRNA nanoparticles with or without additives, CaCl₂ or PS80 were chosen since they were effective in cell lines and keratinocytes, respectively. Then, HeLa, HaCaT, and keratinocytes were treated with endocytosis inhibitors, such as chlorpromazine, which inhibits clathrin-mediated endocytosis, EIPA, which inhibit macropinocytosis, and nystatin, which inhibits caveolin-mediated endocytosis (Ivanov, 2008; Rennick et al., 2021).

Our data demonstrate that EIPA significantly reduced the expression of a reporter protein in HeLa cells, confirming that PF14-mCherry-mRNA nanoparticles predominantly translocate into HeLa cells via macropinocytosis (Figure 5A). In the case of HaCaT cells, PF14-mCherry-mRNA nanoparticles containing CaCl₂ were internalized via macropinocytosis and clathrin-mediated endocytosis, as application of EIPA and chlorpromazine limited the expression of a reporter protein significantly (Figure 5C). In contrast, none of the blockers significantly altered the transport of the PF14-EGFP-mRNA nanoparticles in the absence or presence of PS80 in primary keratinocytes, although EIPA and chlorpromazine apparently caused a slight reduction of EGFP expression (Figure 5E). It should be noted, that even though the media was changed after 4 h of endocytosis inhibitor treatment, the viability of cells was reduced under certain conditions. Specifically, EIPA treatment decreased the viability of HeLa cells (Figure 5B), while also exhibiting a tendency to affect the viability of HaCaT cells (Figure 5D). Additionally, chlorpromazine had an impact on the viability of keratinocytes as shown in Figure 5F.

Next, we evaluated the efficacy of PF14 in transporting mRNA into mouse ears through subcutaneous injection. To mimic the changes in delivery conditions that occur during inflammation, where tight junctions between cells open, we induced ICD in a subgroup of mice. This was done by applying 0.2% PMA topically to the mouse ears, as previously described (Carreras-Badosa et al., 2020). We employed Cy5-mRNA to study the distribution of nanoparticles, and EGFP-mRNA to evaluate cellular uptake of administered nanoparticles and subsequent protein expression. The injection of nanoparticles was carried out 1 hour before the induction of ICD. As expected, the thickness of mouse ears treated with PMA increased significantly, indicating ongoing tissue inflammation (Figures 6A, B). It is important to note that application of PF14-mRNA nanoparticles (with both Cy5-mRNA and EGFP-mRNA) did not result in any ear swelling, as shown in Figures 6A, B. Additionally, there were no significant changes in cytokines related to inflammation such as Cxcl1, Tnfa, Il6, Il1b, and Ifng, as shown in Figure 6C, indicating that there was no provocation of inflammation. However, there was a slight tendency towards an increase in Il1b expression in the PF14-EGFP-mRNA injected group and an increase in Ifng in the PF14-Cy5-mRNA injected group (Figure 6C). Upon subcutaneous injection of the nanoparticles, Cy5-mRNA was visible in both untreated ears and PMA-treated ears, with apparently more spread distribution of Cy5 in the latter case (Figure 6D). Furthermore, we also detected EGFP fluorescence signal near the injection site and found more EGFP-positive cells in the PMA-treated ears compared to the control (Figure 6E). However, neither CaCl₂ nor PS80 improved the delivery of mRNA or the expression of a reporter protein in vivo (Supplementary Figure S4).