key: cord-0768088-f5navgev
authors: Nasr, Sarah S.; Lee, Sangeun; Thiyagarajan, Durairaj; Boese, Annette; Loretz, Brigitta; Lehr, Claus-Michael
title: Co-Delivery of mRNA and pDNA Using Thermally Stabilized Coacervate-Based Core-Shell Nanosystems
date: 2021-11-13
journal: Pharmaceutics
DOI: 10.3390/pharmaceutics13111924
sha: e1791c266b397f9d0d3ee620a747daa70e38397f
doc_id: 768088
cord_uid: f5navgev

Co-delivery of different species of protein-encoding polynucleotides, e.g., messenger RNA (mRNA) and plasmid DNA (pDNA), using the same nanocarrier is an interesting topic that remains scarcely researched in the field of nucleic acid delivery. The current study hence aims to explore the possibility of the simultaneous delivery of mRNA (mCherry) and pDNA (pAmCyan) using a single nanocarrier. The latter is based on gelatin type A, a biocompatible, and biodegradable biopolymer of broad pharmaceutical application. A core-shell nanostructure is designed with a thermally stabilized gelatin–pDNA coacervate in its center. Thermal stabilization enhances the core’s colloidal stability and pDNA shielding effect against nucleases as confirmed by nanoparticle tracking analysis and gel electrophoresis, respectively. The stabilized, pDNA-loaded core is coated with the cationic peptide protamine sulfate to enable additional surface-loading with mRNA. The dual-loaded core-shell system transfects murine dendritic cell line DC2.4 with both fluorescent reporter mRNA and pDNA simultaneously, showing a transfection efficiency of 61.4 ± 21.6% for mRNA and 37.6 ± 19.45% for pDNA, 48 h post-treatment, whereas established commercial, experimental, and clinical transfection reagents fail. Hence, the unique co-transfectional capacity and the negligible cytotoxicity of the reported system may hold prospects for vaccination among other downstream applications.

Nucleic acid-based therapies are currently moving with vast strides towards increasingly broader clinical application. DNA and various forms of RNA (siRNA, miRNA, mRNA, and saRNA), as well as antisense oligonucleotides [1, 2] , have shown promise in alleviating various genetic disorders previously uncatered for by conventional therapeutics. Investigating nucleic acids (NA) as vaccination tools has been for years one of the most advanced fields of research for nucleic acid-based therapies. Many ongoing clinical trials investigate mRNA-based vaccines for rabies, influenza H7N9, influenza H10N8, cytomegalovirus, human metapneumovirus, parainfluenza virus 3, respiratory syncytial virus, and Zika, among others [3] . Moreover, several veterinary DNA vaccines have already been approved [4] . Nevertheless, until recently, no NA-based vaccines have been approved for human use. This situation has been rapidly changing following the outbreak of SARS-CoV-2 in 2019. In December 2020, the mRNA vaccines of Pfizer/BioNTech (BNT162b2) [5] and Moderna (mRNA-1273) [6] were the first vaccines to receive approval for emergency use in humans against SARS-CoV-2.

Gelatin-pDNA coacervates (CoAc) were prepared using the complex coacervation technique (Figure 1) , where all coacervates were assembled in MQ water. First, the optimum mixing temperature of gelatin and pDNA was assessed by mixing gelatin type A solution (10 mg·mL −1 ) with pAmCyan (100 µg·mL −1 ) in a ratio of 1:1 v/v under three different mixing temperatures (23, 37 and 55 • C). For further optimization of the mixing conditions, the appropriate mass ratio between gelatin and pDNA was investigated. A solution of pAmCyan in MQ water (100 µg·mL −1 ) was added to gelatin type A solutions in MQ water with different concentrations (10 mg·mL −1 , 7 mg·mL −1 , 5 mg·mL −1 , 3 mg·mL −1 , 1 mg·mL −1 and 0.1 mg·mL −1 ) at 37 • C ( Table 1 ). The two solutions were then mixed via vortexing at maximum speed for 1 min. Gelatin-pDNA coacervates (CoAc) were thus assembled at gelatin to pDNA mass ratios of 100:1, 70:1, 50:1, 30:1, 20:1 and 1:1, respectively ( Table 1) . Assessments of the optimum mixing ratio and temperature were performed using dynamic light scattering measurements of particle size and zeta potential.

CoAc 30 was selected given its small size, low PDI, and negative zeta potential for further experiments. Thermal stabilization was performed in four subsequent heatingcooling cycles per sample. Each cycle started with CoAc incubation at 55 ± 0.5 • C for 30 min, followed by rapid cooling to 0 ± 0.5 • C for 5 min. The resulting gelated particle is referred to as thermally stabilized coacervate (TS-CoAc), while particles that were not subjected to thermal cycling (unstabilized) are referred to simply as coacervate (CoAc).

Protamine sulfate solution (0.3 mg·mL −1 ) was used to coat the preformed CoAc 30 or TS-CoAc 30 , to the final protamine sulfate to gelatin mass ratio of 1:5 w/w (taking into consideration the two-fold dilution of gelatin's mass contribution during the coacervation step). Protamine sulfate solution was mixed with the preformed particle cores under laminar flow conditions in a microfluidic setting, using a meander chip with two inlets, one for each component at a total flow rate of (2 mL·min −1 ). The formed core-shell system was left to stand for at least 24 h at 4 • C. The core-shell particle was then surface-loaded with 1 µg mCherry per 170 µg of particles, and mixed by simple pipetting to a final pDNA: mRNA mass ratio of 5:1 per particle. The particles were further allowed to stand for 15 more minutes before use. 

CoAc30 was selected given its small size, low PDI, and negative zeta potential for further experiments. Thermal stabilization was performed in four subsequent heatingcooling cycles per sample. Each cycle started with CoAc incubation at 55 ± 0.5 °C for 30 min, followed by rapid cooling to 0 ± 0.5 °C for 5 min. The resulting gelated particle is Figure 1 . The general preparation procedure of CoAc, TS-CoAc, and P-TS-CoAc, and the proposed mechanism of thermal stabilization of CoAc into TS-CoAc. P-CoAc was prepared by the introduction of CoAc and protamine sulfate into a microfluidic coating system while skipping the thermal stabilization step of CoAc. 

Samples were characterized for particle size, PDI, and zeta potential using Zetasizer Nano-ZS (Malvern Instruments, Worcestershire, U.K.), utilizing 4 mW He−Ne laser at a wavelength of 633 nm and a backscattering angle of 173 • at 25 • C. Samples were measured in concentrations of (1550 µg·mL −1 ) of non-coated particles and (925 µg·mL −1 ) of coated particles. Particle size is given as intensity-based z-average. Moreover, the colloidal stability of P-CoAc and P-TS-CoAcs was studied for three weeks at 4 • C storage temperature, where dynamic light scattering was used to assess changes in particle size or PDI over time.

NTA (NanoSight LM10, Malvern Instruments, Worcestershire, UK) was used to assess the effect of thermal stabilization on the colloidal stability of the gelatin-pDNA core in a cell culture medium; this is where both non-stabilized CoAc and thermally stabilized TS-CoAc were incubated in RPMI-1640, in a ratio of (1:10 v/v) at 37 • C for 4 h. For each sample, particle count (particle·mL −1 ) was recorded at zero time and after 4 h of incubation in RPMI-1640. Data was collected from three videos, 30 s each, where the camera level was manually set to 14 during all captures. NanoSight 3.3 software was used to process the videos at a detection threshold of 5.

The number of nanoparticles·mL −1 was also determined in MQ water using the same settings. The information was used along with the previous knowledge of the number of either pDNA or mRNA molecules per 1 mL of particle suspension to determine the numbers of pDNA and mRNA molecules per nanocarrier, based on the method previously described by Zagato et al. [41] .

Gelatin solution (3 mg·mL −1 ) was prepared by dissolving gelatin in MQ water at 55 • C. From the gelatin stock, CoAc was prepared as previously described. A portion of the resultant CoAc was then thermally cycled to prepare TS-CoAc, as previously described. CD spectra of all samples were recorded at 37 • C on a Jasco 810 spectropolarimeter (Jasco, Tokyo, Japan) in a 0.1-cm path-length quartz cell. Samples were scanned in and blanked to MQ water. Each sample was scanned 15 rounds per measurement, and eventually, the spectra for each were obtained after subtracting the contribution of the MQ water blank.

Unstained CoAc and TS-CoAc particles, with or without protamine sulfate coating, were visualized using TEM (JEM 2011, JEOL, St Andrews, UK). Either non-coated particles (775 µg·mL −1 ) or coated particles (925 µg·mL −1 ) were used, where 10 µL samples were mounted on a copper grid (S160-4, Plano GmbH, Wetzlar, Germany), and allowed to dry overnight before visualization. The TEM measurement was performed at an accelerating voltage of 200 kV.

Investigating the effect of thermal stabilization on pDNA protection was performed by Dnase I challenge, comparing P-CoAc without and with thermal stabilization. In the same assay, the protection of the surface-bound mRNA against RnaseA was also monitored. mRNA-loaded P-TS-CoAc or P-CoAc (NA load of 1.5 µg mRNA and 7.5 µg pDNA) were incubated with 0.008 U·mL −1 Dnase I and 0.027 µg·mL −1 Rnase A (both Sigma-Aldrich, Darmstadt, Germany) in Dnase I working buffer at 37 • C for 30 and 60 min. Particle samples were compared to samples of equivalent masses of naked mRNA and pDNA subjected to the same treatments. Following the pre-stated incubation periods, 20 µL samples or controls were drawn from the reaction mixture and the reaction was quenched using 3 µL 50 mM EDTA and 1 µL Ribolock (both Sigma-Aldrich, Darmstadt, Germany), then the nucleic acids were released from the particles via digestion with Trypsin (30 µL, 1.17 mg·mL −1 ) for 30 min at 37 • C, followed by addition of high molecular weight Heparin (10 µL, 300 mg·mL −1 ). Samples were loaded onto 1.3% w/v agarose gel (Serva, Heidelberg, Germany) containing ethidium bromide (0.3 µg·mL −1 ) (Sigma-Aldrich, Darmstadt, Germany), in TBE buffer (1×) and run for 90 min at 90 mV. The gel was visualized under UV-light (Fusion FX7 imaging system, Peqlab, Erlangen, Germany).

As further confirmation of the difference in pDNA shielding effects between thermally stabilized and unstabilized cores in a more physiologically relevant medium, sample volumes from uncoated CoAc and TS-CoAc equivalent to 3 µg pDNA were incubated with 10% FCS in HBSS for 3 h at 37 • C. The activity of serum nucleases was then quenched using EDTA (150 µL, 50 mM). pDNA was then released from the sample using subsequent treatment with Trypsin (30 µL, 1.17 mg·mL −1 ) for 150 min at 37 • C, then in a high molecular weight Heparin solution (30 µL, 30 mg·mL −1 ). Samples were loaded onto a 0.7% w/v agarose gel and run for 60 min at 60 mV. Data were normalized to intact supercoiled pDNA as a control.

To assess the entrapment efficiency of pDNA and mRNA, CoAc 30 or TS-CoAc 30 were ultracentrifuged at 58,000× g, 4 • C for 2 h. The supernatant was then analyzed for pAmCyan1 content using PicoGreen assay or for mCherry using RiboGreen assay, according to the manufacturer's protocols. Data were normalized to free pAmCyan or mCherry subjected to the same treatment as the samples.

Both transfection efficiency and cytotoxicity of the system were assessed in murine dendritic cell line DC2.4. Transfection efficiency of pAmCyan and mCherry was assessed in DC2.4 murine dendritic cell line, passages 6 to 8. Briefly, cells were seeded at a density of 50,000 cells per well in 24 well plates, in RPMI-1640 containing FCS (10% v/v), HEPES (1%), NEAA (1%), β-mercaptoethanol (0.0054%). Cells were grown for 48 h reaching confluency of approximately 80% before being treated with either samples or controls. Samples were either P-CoAc or P-TS-CoAc, both using a concentration of 170 µg particles per well, equivalent to NA concentrations of 1 µg mCherry and 5 µg pAmCyan per well. As negative controls, untreated cells and cells treated with either naked mRNA or pDNA were used. Cells treated with commercial transfection reagents were used as positive controls. For single transfection, JetMessenger was used for mRNA, whereas JetPrime was used for pDNA. As an additional positive control, a solid lipid nanoparticle (SLN) inspired by the current clinical standard of mRNA/siRNA delivery was prepared. Briefly, an aqueous solution of mRNA and pDNA in a mass ratio of 1:5 (pH = 4) was mixed with an ethanolic solution of the following lipids, DLin-MC3-DMA, DPPC, Cholesterol, and DSPE-PEG2000, in a molar ratio of 50:10:38.5:1.5 and at a final N/P ratio of 0.5 [42] . Both JetPrime and JetMessenger along with Lipofectin, polyethyleneimine, and solid lipid nanoparticles were used as controls using a combination of both pDNA and mRNA to assess their dualtransfection efficiency. JetPrime, JetMenssenger, and Lipofectin are commercial transfection reagents. JetMessenger is optimized for mRNA, and JetPrime is recommended for pDNA. Lipofectin can be used for both species. These commercial reagents were used per the manufacturer's protocols. High M W branched PEI is a golden standard among the polymers in terms of transfection efficacy, despite its cytotoxicity. Thus, PEI-NPs were prepared and included as a polymeric reference sample, using PEI: pDNA: mRNA mass ratio equivalent to the protamine sulfate: pDNA: mRNA mass ratio of 30:5:1 originally present in TS-CoAc. As an internal control, the same protamine sulfate concentration that was used for CoAc or TS-CoAc coating was used to formulate a protamine sulfate coacervate with pDNA, to which mRNA was added immediately before cell treatment, in the same pDNA: mRNA mass ratio used for either TS-CoAc or CoAc. Samples were incubated with cells for 6 h under shaking at 250 RPM, then removed and replaced with fresh medium and further incubated for 48 h. Cell harvesting was performed following washing twice with HBSS, where cells were detached using Trypsin-EDTA (100 µL), followed by the addition of 2% FCS in HBSS (900 µL). Samples were centrifuged at 4 • C and 300× g for 5 min, the pellet was rewashed in 1 mL HBSS then re-suspended and fixed in paraformaldehyde (4% w/v). Transfection efficiency was analyzed using flow cytometry (BD LSRFortessa TM Cell Analyzer Biosciences, Heidelberg, Germany), using the PE-Texas red channel for mCherry and AmCyan channel for pAmCyan. Flowjo version 10.6.1 was used for data processing.

Cytotoxicity of P-TS-CoAc was assessed in DC2.4, passages 10 to 12. Cells were seeded as previously described and then treated with P-TS-CoAc in a concentration of (340 µg·mL −1 , 170 µg·mL −1 , or 85 µg·mL −1 ). Cells were incubated with the samples Pharmaceutics 2021, 13, 1924 8 of 21 for 6 h under shaking at 250 RPM, following which the samples were removed, and the cells detached as previously described. As a positive control, cells killed by heating at 70 • C post detachment were used. Following the last washing step and before fixation with paraformaldehyde (4% w/v), cells were stained with live dead fixable stain 568/583 (PromoCell GmbH, Heidelberg, Germany) according to the manufacturer's protocol. The used kit stains only dead cells using cell membrane-impermeable amine-reactive peptides that can be detected on the PE-emission filter. The percentage of dead cells in the different samples was then measured by flow cytometry using the PE-A channel. Cell viability was calculated according to Equation (1) .

Cell viability (%) = total cell number − PE positive cell number total cell number × 100 (1)

DC2.4 cells were seeded in 8 well glass bottom µ-slide (Ibidi, Gräfelfing, Germany), precoated with 1 mg/mL bovine collagen type I solution, Purecol (CellSystems, Troisdorf, Germany), at a density of 25,000 cells/well. The transfection procedure was performed as previously described. Immediately before visualization, the cells were washed twice using HBSS and fixed for 5 min with 4% (v/v) paraformaldehyde (PFA; Electron Microscopy Sciences) in HBSS for 5 min at r.t. After rinsing with HBSS, cells were mounted and stored at 4 • C until CLSM analysis (Leica TCS SP8, Leica Microsystems, Mannheim, Germany). Image acquisition was conducted on a Leica TCS SP8 confocal imaging microscope with a 25× water immersion objective (Fluotar VISIR 25×/0.95) at 1024 × 1024 resolution. For AmCyan, fluorescence was detected between 495-550 nm (excited at 405 nm; 24% laser intensity), for mCherry, fluorescence was detected between 683-784 nm (excited at 561 nm; 10% laser intensity), both using a HyD detector. Images were then processed with the Leica Application Suite (LAS) X software.

Data were analyzed using Graph Pad Prism 8 for Windows (Version 8.01, GraphPad Software Inc., San Diego, CA, USA) and generally presented as the mean of individual values (generally 3-9 samples), with standard deviation indicated by the error bars. (N) refers to the number of experiments, (n) refers to the number of samples per experiment. A one-way ANOVA was performed for all test samples, followed by Tukey's post hoc test for assessment of inter-group individual differences. Data were considered statistically significant at a level of significance of p < 0.05 (* p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001).

Either non-stabilized coacervates (CoAc) or thermally stabilized coacervates (TS-CoAc) were prepared according to the scheme illustrated in (Figure 1 ), with the sample compositions indicated in (Table 1) . Thermal stabilization was performed as four subsequent heating-cooling cycles (55 ± 0.5 • C for 30 min followed by 0 ± 0.5 • C for 5 min per cycle). Subsequent protamine sulfate coating was performed via a microfluidic assembly to give protamine sulfate coated TS-CoAc (P-TS-CoAc).

We combined pAmCyan (pDNA) and gelatin via electrostatic interaction in Milli-Q purified water to form a coacervate-based core. We varied the gelatin to a pDNA (pAmCyan) ratio first ( Figure 2 ) and five gelatin to pDNA mass ratios were investigated (100:1, 70:1, 50:1, 30:1, 20:1 and 1:1 w/w). It was observed that a gelatin to pDNA mass ratio of 30:1 (CoAc 30 ) or 70:1 (CoAc 70 ) led to the formation of coacervates possessing significantly smaller diameters (170 nm and 151 nm, respectively) and smaller polydispersity indices (PDI) (0.17 and 0.21, respectively) compared to higher or lower ratios, which possessed particle diameters and PDIs of at least 257 nm and 0.32 in case of CoAc 100 (Figure 2c,d) . The intermediate-mass ratio of 50:1 (CoAc 50 ) showed the largest particle diameter of 1772 and a PDI of 0.52 (Figure 2c,d) . CoAc 30 and CoAc 70 showed slightly negative (−5.5 mV) and positive (1.8 mV) zeta potentials, respectively, while CoAc 50 showed a zeta potential of almost zero (Figure 2b ).

Pharmaceutics 2021, 13, 1924 10 of 23 a negative zeta potential. Hence, CoAc30 assembled at 37 °C was adopted for all subsequent experiments. Our data suggest that the temperature of the solution during the initial interaction between pDNA and gelatin crucially affects core size and PDI [45] . Gelatin and pDNA both possess a helical conformation in aqueous media below 40 °C , with persistence lengths of 10 nm and 50 nm, respectively [46, 47] . During the complexation of two polyions, the higher the chain flexibility of the two polyions, the better the interaction. The flexibility of a chain is an inverse function of its persistence length [47] . The persistence length of DNA was reported to be temperature-dependent, with a reduction in persistence length as the temperature increases [48] . Therefore, we assume that at 37 °C , pDNA has a smaller persistence length than it does at 23 °C , thus the proximity between the persistence lengths of gelatin and pDNA chains was higher at 37 °C than that at 23 °C , resulting in a better gelatin-pDNA interaction at 37 °C . The particles assembled at 37 °C had a smaller size and PDI compared to those formed at 23 °C . A mixing temperature of 55 °C produced Under the coacervation conditions (MQ water, pH = 6.2, 37 • C) used for gelatin type A (Bloom number 228) and pDNA (4.7 kbp), the gelatin to pDNA mass ratios of 30:1, 50:1, and 70:1 provided the three coacervates with the smallest zeta potentials. Above 70:1 and below 30:1, a surplus of the positively charged gelatin or negatively charged pDNA existed in the coacervates. The repulsive forces between molecules of the similarly charged predominant polyion reduced the packing density of coacervate and its storage modulus, observed as an increase in particle size and PDI, in accordance with what was previously described by Arfin et al. [43] . A slightly overcharged coacervate, as in the case of CoAc 30 and CoAc 70 , despite not having the highest packing density, or storage modulus is more kinetically stable than the high-density coacervates formed at the point of absolute charge neutralization (CoAc 50 ), which are more liable to aggregation over time due to their surface neutrality [44] . This explains why CoAc 30 and CoAc 70 showed acceptable particle sizes and PDIs as opposed to CoAc 50 . It is worth mentioning that upon changing the plasmid size between 2.6 to 7.2 Kbp, such CoAc could still be assembled at an acceptable size and PDI at a mass ratio of 30:1 gelatin to pDNA ( Figure S1 ).

Next, we optimized the core assembly temperature, where CoAc 100 showed a smaller diameter of 190 nm and PDI of 0.26 when assembled at 37 • C compared to coacervates assembled at 55 • C or 23 • C showing diameters of 541 nm and 246 nm, and PDI of 0.611 and 0.35, respectively (Figure 2c,d) . We found that using a gelatin to pDNA mass ratio of 30:1 and mixing temperature of 37 • C resulted in the smallest particle size and PDI, with a negative zeta potential. Hence, CoAc 30 assembled at 37 • C was adopted for all subsequent experiments.

Our data suggest that the temperature of the solution during the initial interaction between pDNA and gelatin crucially affects core size and PDI [45] . Gelatin and pDNA both possess a helical conformation in aqueous media below 40 • C, with persistence lengths of 10 nm and 50 nm, respectively [46, 47] . During the complexation of two polyions, the higher the chain flexibility of the two polyions, the better the interaction. The flexibility of a chain is an inverse function of its persistence length [47] . The persistence length of DNA was reported to be temperature-dependent, with a reduction in persistence length as the temperature increases [48] . Therefore, we assume that at 37 • C, pDNA has a smaller persistence length than it does at 23 • C, thus the proximity between the persistence lengths of gelatin and pDNA chains was higher at 37 • C than that at 23 • C, resulting in a better gelatin-pDNA interaction at 37 • C. The particles assembled at 37 • C had a smaller size and PDI compared to those formed at 23 • C. A mixing temperature of 55 • C produced CoAc 100 with a significantly higher PDI and much larger particle size than the two other mixing temperatures. When the mixture was warmed to 55 • C, above the helix-coil transition temperature of gelatin [49] , gelatin lost its helical structure, displaying a random coil conformation. We speculate that the loss of helical conformation compromised the structural synonymity of gelatin and pDNA, and despite pDNA possessing an even smaller persistence length and higher chain flexibility at this temperature, the coacervates formed between the pDNA's helix and gelatin's random coil were not as compact as the coacervates formed between pDNA's and gelatin's helices at 37 • C. Thus, the significance of maintaining the α-helical structure of gelatin type A by keeping the temperature of the mixture below the helix-coil transition threshold is critical for a successful initial interaction. This feature may also positively contribute to the stability of the system under physiological conditions.

Thermal treatment of CoAc 30 via four subsequent heating-cooling cycles to form TS-CoAc led to a significant enhancement in the system's colloidal stability. P-CoAc showed a progressive increase in particle diameter and PDI in MQ water starting as early as 3 days post coating, compared to the P-TS-CoAc, which resisted any core disruption by the protamine sulfate coat displaying no significant changes in particle diameter or PDI for 3 weeks (Figure 3a,b) . The measured lower stability of CoAc-NPs when getting in contact with a competing polycation, such as protamine, could be expected in a core that solely relies on electrostatic interactions. Without further stabilization, protamine can easily displace gelatin from the core, which would typically result in an increase in particle size and PDI due to the release of heterogenous gelatin components from the entire nano-system, and hence the observed particle disruption in the case of P-CoAc and the lack of it in the case of P-TS-CoAc. The successful deposition of protamine sulphate on the surface of TS-CoAc was accompanied by a reversal in zeta-potential from −3.9 mV in the case of TS-CoAc to 8.2 mV in the case of P-TS-CoAc (Figure 3c ). The enhanced colloidal stability of TS-CoAc compared to CoAc was further confirmed as a function of particle count (particle mL −1 ) using NTA (Figure 3d and Figure S2 ). Following a 4h incubation of TS-CoAc in RPMI-1640, the particle count dropped from 2.9 × 10 11 particles mL −1 to 1.1 × 10 11 particles mL −1 , on the other hand, CoAc decreased by one order of magnitude in particle count from 4.76 × 10 11 particles mL −1 to 3.93 × 10 10 particles mL −1 .

Gel migration assay following DNase I/RNase A digestion of both P-TS-CoAc and P-CoAc coacervates demonstrated a superior protective effect of the thermally stabilized system (P-TS-CoAc) as opposed to the non-stabilized system (P-CoAC) with regards to the core loaded pDNA Figure 3e ,f. Meanwhile, no detectable difference was observed between both the stabilized and unstabilized systems with regards to the surface-loaded mRNA's shielding. Yet both nanocarriers showed superior protection of both the NA cargos compared to the naked NA control. This experiment could also provide insight into the release inducing mechanisms for the two NA cargos from the proposed system. As a combination of proteolytic activity and polyanion exchange is required for the release of these cargos and hence the use of Trypsin and Heparin as exemplary release inducers in this assay.

A trend in enhanced protection of the pDNA against serum nucleases following thermal stabilization was observed. In 10% serum incubation for 3 h, TS-CoAc maintained 31.9% of intact pDNA, compared to 17.4% of CoAc particles ( Figure S5a,b) .

Circular dichroism data ( Figure 3g ) showed a reduction in the negative ellipticity of the peak at 204 nm, reported to coincide with the random coil structure of gelatin [50] . The 204 nm peak intensity followed the rank order gelatin > CoAc > TS-CoAc when all were measured at 37 • C, indicating that coacervate formation and thermal stabilization caused a slight decrease in sample randomicity and promoted a more ordered structure. All samples were prepared from the same gelatin stock to eliminate any variations that may arise due to differences in concentration rather than optical activity among the samples.

Gelatin type A is the acidic denaturation product of collagen. Being a denatured protein, gelatin possesses low antigenicity [51] , thus rendering it suitable for repeated administration. Gelatin possesses a lower charge density than most cationic polymers typically used in transfection, giving it a safety advantage [52] , yet this renders its coacervates with nucleic acids much less stable. To date, techniques utilizing gelatin nanocarriers for NA delivery rely heavily on chemical cross-linking, even of cationized gelatin, for particle preparation and stabilization. Some of the commonly reported cross-linkers include symmetrical bifunctional aldehydes, such as glutaraldehyde and glyoxal [53] , as well as EDC (1-ethyl-3-(3-dimethyl-aminopropyl) carbodiimide hydrochloride) [54] . In the specific scenario of nucleic acid delivery, such reactions may hinder the eventual release of the nucleic acid cargo from the gelatin matrix given the fact that both the nucleic acid and gelatin contain abundant amine groups and can thus become covalently bound. Covalently bound DNA-proteins are reported to hurdle the fidelity of gene expression in their host cell by interacting with the translation and transcription mechanisms of the cell via their DNA domain [55, 56] . Thus, when designing our system, we aimed to avoid chemical cross-linking and to rely on the intrinsic properties of both gelatin and DNA to find an alternative stabilization technique. Gelatin-DNA coacervates rely mainly on easily reversible electrostatic interactions, hydrogen bonds, and hydrophobic interactions between gelatin and DNA. Such non-covalent interactions are suitable for the physiological interactions between nucleic acids and proteins during various cellular processes [57] . Gelatin-DNA coacervates have demonstrated an ability to irreversibly transform from a coacervate to an anisotropic nanogel phase upon heating above gelatin's helix-coil transition temperature, followed by cooling below this temperature [58, 59] .

Thermal stabilization relies on the conversion of gelatin-pDNA coacervate in the case of CoAc to an anisotropic nanogel state (TS-CoAc), a phenomenon previously reported by Rawat et al. [58] , but remains unused as a tool for nucleic acid delivery to date. When heating the gelatin above its helix-coil transition temperature of 40 • C, it assumes a predominantly random coil chain morphology, which extends further across the coacervate's matrix, traversing more individual pDNA molecules along its length than a helix of equal length would. Upon sudden cooling of such coils in their extended state, the rapid loss of heat from the system is consumed in the formation of ionic and hydrogen bonds between the gelatin coils and pDNA, rather than allowing the coil a chance to resort back to its helix morphology. In this case, the pDNA acts as a scaffold supporting a matrix of extended gelatin coils even at a temperature below the helix-coil transition temperature. This physical method of cross-linking can be considered a safer option than most chemical cross-linkers commonly used for gelatin nanocarrier preparation, which compromise the intrinsic biodegradability and biocompatibility of the polymer [60, 61] . Thus, we selected thermal treatment of gelatin-pDNA coacervates as our physical stabilization technique of choice.

A microfluidic system was adopted for coating the cores under laminar flow conditions. This technique deposited a homogenous coat across the whole particle population without any visible compromise to core integrity, as observed in the TEM images ( Figure 4) .

We could not detect a distinct core-shell structure in the case of P-CoAc (Figure 4c ). Meanwhile, (Figure 4d and Figure S3) show the successful coating of the TS-CoAc with protamine sulfate, in protamine sulfate to gelatin mass ratio of 1:5, where an evident coreshell structure can be observed. This observation was further confirmed by the change in zeta potential from −3.9 mV to 8.2 mV (Figure 3c ). The fact that P-TS-CoAc shows a distinct core-shell structure as opposed to the P-CoAc, further demonstrates the value of thermal stabilization.

Following mRNA surface-loading on P-TS-CoAc, particle size and PDI showed no discernible difference from unloaded P-TS-CoAc for up to one week. This further supports the assumption that the system remains intact and is taken up as a unit by the cells, while no mRNA-protamine sulfate coacervate sub-populations are formed ( Figure S4 ).

Protamine is another peptide commonly used in the field of vaccination, that has been widely employed by CureVac AG in their RNactive ® technology, where it shows promise as an NA vaccine delivery tool [62] [63] [64] [65] [66] . Protamine is a naturally occurring membrane translocating peptide, with membrane translocation properties comparable to HIV-1 tat [67] . Protamine-DNA complexes bind to importins, which are transport proteins associated with the nuclear pore complex, thus facilitating the nuclear translocation of pDNA [38, 68] . All the aforementioned properties in addition to its established pharmaceutical application became our motivation to use it as a particle coating. potential from −3.9 mV to 8.2 mV (Figure 3c ). The fact that P-TS-CoAc shows a distinct core-shell structure as opposed to the P-CoAc, further demonstrates the value of thermal stabilization.

Following mRNA surface-loading on P-TS-CoAc, particle size and PDI showed no discernible difference from unloaded P-TS-CoAc for up to one week. This further supports the assumption that the system remains intact and is taken up as a unit by the cells, while no mRNA-protamine sulfate coacervate sub-populations are formed ( Figure S4 ). Protamine is another peptide commonly used in the field of vaccination, that has been widely employed by CureVac AG in their RNactive ® technology, where it shows promise as an NA vaccine delivery tool [62] [63] [64] [65] [66] . Protamine is a naturally occurring membrane translocating peptide, with membrane translocation properties comparable to HIV-1 tat [67] . Protamine-DNA complexes bind to importins, which are transport proteins associated with the nuclear pore complex, thus facilitating the nuclear translocation of pDNA [38, 68] . All the aforementioned properties in addition to its established pharmaceutical application became our motivation to use it as a particle coating. 

Entrapment efficiency (EE%) assessed using PicoGreen assay ( Table 2) showed that pDNA was fully incorporated into the system at a gelatin to pDNA mass ratio as low as 30:1. Both CoAc 30 and TS-CoAc 30 showed no considerable difference in EE % of pDNA (Table 2) . RiboGreen assay performed on P-TS-CoAc showed EE% of 97.81% of mRNA. The carrier packed more mRNA (1.884 × 10 12 ) than pDNA (1.076 × 10 12 ) molecules per 170 µg of particles. Moreover, on the particle level, based on the average particle count obtained using NTA (data not shown) each P-TS-CoAc particle packed approximately 5318 pDNA molecules and 9312 mRNA molecules. Calibration curves used in the establishment of these assays are provided in ( Figure S6 ). Table 2 . The entrapment efficiency (EE %) of the carriers for pDNA (pAmCyan) was assessed using PicoGreen assay, whereas RiboGreen assay was used for mRNA (mCherry) (N = 1, n = 3) . The average number of pDNA or mRNA molecules per particle dose (170 µg as the dose used per well in a 24 well-plate format) and the numbers of pDNA or mRNA molecules per particle (NP) were calculated based on the used amount of NAs and the particle count from NTA. 

Compared to untreated cells, murine dendritic cell line (DC2.4) cells treated with either the two-fold (340 µg·mL −1 ), the same (170 µg·mL −1 ) or half of (85 µg·mL −1 ) the particle concentration used in transfection efficiency studies showed cell viabilities of 91.9%, 97.1%, and 97.7%, respectively, following 6 h incubation ( Figure 5 ). Meanwhile, a 24 h incubation of 170 µg/mL particles showed 87.4% cell viability. A total of 170 µg/mL was selected for the 24 h extended viability assay because it was the dose to be used for the transfection trials. Protamine-mRNA-pDNA coacervates assembled using either equivalent protamine doses in P-TS-CoAc concentrations of 340, 170 or 85 µg/mL or 5-fold; these concentrations were also investigated and showed no significant difference in cytotoxicity compared to either P-TS-CoAc or untreated controls following 6 h incubation ( Figure S7 ). These data align with the established biocompatibility of the two major nanocarrier components, gelatin and protamine sulfate [60, 61] . 

Upon application of both protamine sulfate coated and mRNA surface-loaded P-CoAc and P-TS-CoAc to murine dendritic cell line DC2.4, both P-CoAc and P-TS-CoAc showed successful, simultaneous transfection of the cells with both mRNA (mCherry) and pDNA (pAmCyan). The transfection efficiency and level of protein expression of both pAmCyan and mCherry significantly surpassed all other test groups except for the JetM single transfection of mCherry transfection. Yet in the case of double transfection, despite the insignificant difference in transfection efficiency between JetM and both P-TS-CoAc and P-CoAc, JetM failed to cause any discernible transfection with pAmCyan ( Figure  6a,b) . Figure 6a ,b show comparable transfection efficiencies and AmCyan expression levels in the case of pDNA, with either P-TS-CoAc and P-CoAc. P-TS-CoAc showed a transfection efficiency of 37.6 ± 19.45% and an MFI of 686 ± 148 as opposed to 36.22 ± 19.21% and 670 ± 139 with P-CoAc, thus indicating that thermal stabilization did not reduce the transfection efficiency of pDNA. Both systems were more efficient than the commercial transfection reagent JetPrime (0.64 ± 0.57%, 152 ± 43), given the challenging nature of transfection in DC2.4. P-TS-CoAc showed a transfection efficiency and MFIs of 61.4 ± 21.6% and 909 ± 253 as opposed to 53.8 ± 22.3% and 794 ± 180 with P-CoAc for mRNA. We could resort the difference, though statistically insignificant, to the more efficient coating and more stable surface in the case of P-TS-CoAc, which allowed better binding and stabilization of the surface-loaded mRNA.

Protamine sulphate-NA coacervate serving as an internal control showed inconsistent transfection of both pDNA and mRNA (Figure 6) , which was negligible in most samples. JetMessenger (for mRNA) and JetPrime (for pDNA) were used as representa- Upon application of both protamine sulfate coated and mRNA surface-loaded P-CoAc and P-TS-CoAc to murine dendritic cell line DC2.4, both P-CoAc and P-TS-CoAc showed successful, simultaneous transfection of the cells with both mRNA (mCherry) and pDNA (pAmCyan). The transfection efficiency and level of protein expression of both pAmCyan and mCherry significantly surpassed all other test groups except for the JetM single transfection of mCherry transfection. Yet in the case of double transfection, despite the insignificant difference in transfection efficiency between JetM and both P-TS-CoAc and P-CoAc, JetM failed to cause any discernible transfection with pAmCyan (Figure 6a,b) . Figure 6a ,b show comparable transfection efficiencies and AmCyan expression levels in the case of pDNA, with either P-TS-CoAc and P-CoAc. P-TS-CoAc showed a transfection efficiency of 37.6 ± 19.45% and an MFI of 686 ± 148 as opposed to 36.22 ± 19.21% and 670 ± 139 with P-CoAc, thus indicating that thermal stabilization did not reduce the transfection efficiency of pDNA. Both systems were more efficient than the commercial transfection reagent JetPrime (0.64 ± 0.57%, 152 ± 43), given the challenging nature of transfection in DC2.4. P-TS-CoAc showed a transfection efficiency and MFIs of 61.4 ± 21.6% and 909 ± 253 as opposed to 53.8 ± 22.3% and 794 ± 180 with P-CoAc for mRNA. We could resort the difference, though statistically insignificant, to the more efficient coating and more stable surface in the case of P-TS-CoAc, which allowed better binding and stabilization of the surface-loaded mRNA.

Pharmaceutics 2021, 13, 1924 18 of 23 Protamine combines an ability to promote cytosolic delivery of mRNA, as well as nuclear translocation of pDNA via four specialized nuclear localization signal-like sequences in its structure [69, 70] , setting it apart from most of the aforementioned controls. Protamine sulphate-NA coacervate serving as an internal control showed inconsistent transfection of both pDNA and mRNA (Figure 6 ), which was negligible in most samples. JetMessenger (for mRNA) and JetPrime (for pDNA) were used as representatives of suc-cessful, widely applied commercial transfection reagents, which could serve as positive controls. Yet when used for co-transfection with both mRNA and pDNA, both systems displayed negligible transfection for the NA they were not designed to deliver, as well as a reduction in the transfection efficiency of their NA of specialization as opposed to our systems that demonstrated successful co-transfection. The same holds for Lipofectin, PEI, and DLin-MC3-DMA based SLN (Figure 6a) .

Protamine combines an ability to promote cytosolic delivery of mRNA, as well as nuclear translocation of pDNA via four specialized nuclear localization signal-like sequences in its structure [69, 70] , setting it apart from most of the aforementioned controls. However, when a protamine coacervate was assembled at the same protamine:mRNA:pDNA mass ratios as P-TS-CoAc and used as an internal control using the same NA doses, the transfection performance of the protamine coacervate was inferior to P-TS-CoAC. This could be resorted to a trojan horse effect exerted by gelatin-pDNA coacervate core, because, in such an arrangement, a considerable fraction of the anionic charges of pDNA could be occluded within the gel core, while only a fraction of the surface-exposed pDNA interacts with the protamine coat, sparing more of protamine's cationic groups for endosomal disruption. Such an arrangement could be allowing the protamine to function at an apparently higher NP ratio despite the lower actual protamine dose. An additional possible explanation for the superior dual transfection performance of P-CoAc and P-TS-CoAc compared to the controls may also be due to a unique time-resolved release and translation of NA cargo from these two systems; this could hence be the subject of a more detailed future study.

Moreover, the confocal microscopy images of the highest performing treatments reveal that cells treated with P-TS-CoAc displayed visibly more consistent transfection patterns in the case of mRNA ( Figure 7b ) and pDNA (Figure 7e ) than cells treated with the P-CoAc (Figure 7c,g) . This might indicate that the enhancement in colloidal stability of the system via thermal stabilization, in addition to providing better shielding of the core-loaded pDNA and a more stable surface for a stable loading of mRNA, allowed the system to remain intact for longer during transfection and for the co-loaded NAs to be taken up as a unit, a feature that can prove valuable in a vaccination context of antigen-adjuvant co-delivery. This observation suggests that thermal stabilization may have enhanced both the transfectional and co-transfectional capabilities of the system. Cells treated with commercial transfection reagent displayed a strong expression of mCherry in the case of JetMessenger ( Figure 7d ) and a weaker yet more homogenous expression of AmCyan in the case of JetPrime (Figure 7h) . A detailed gating strategy is provided in (Figure S8 ).

Pharmaceutics 2021, 13, 1924 19 of 23

However, when a protamine coacervate was assembled at the same protamine:mRNA:pDNA mass ratios as P-TS-CoAc and used as an internal control using the same NA doses, the transfection performance of the protamine coacervate was inferior to P-TS-CoAC. This could be resorted to a trojan horse effect exerted by gelatin-pDNA coacervate core, because, in such an arrangement, a considerable fraction of the anionic charges of pDNA could be occluded within the gel core, while only a fraction of the surface-exposed pDNA interacts with the protamine coat, sparing more of protamine's cationic groups for endosomal disruption. Such an arrangement could be allowing the protamine to function at an apparently higher NP ratio despite the lower actual protamine dose. An additional possible explanation for the superior dual transfection performance of P-CoAc and P-TS-CoAc compared to the controls may also be due to a unique time-resolved release and translation of NA cargo from these two systems; this could hence be the subject of a more detailed future study. Moreover, the confocal microscopy images of the highest performing treatments reveal that cells treated with P-TS-CoAc displayed visibly more consistent transfection patterns in the case of mRNA ( Figure 7b ) and pDNA (Figure 7e ) than cells treated with the P-CoAc (Figure 7c,g) . This might indicate that the enhancement in colloidal stability of the system via thermal stabilization, in addition to providing better shielding of the coreloaded pDNA and a more stable surface for a stable loading of mRNA, allowed the system to remain intact for longer during transfection and for the co-loaded NAs to be taken up as a unit, a feature that can prove valuable in a vaccination context of antigen-adjuvant co-delivery. This observation suggests that thermal stabilization may have enhanced both the transfectional and co-transfectional capabilities of the system. Cells treated with commercial transfection reagent displayed a strong expression of mCherry in the case of Jet-Messenger ( Figure 7d ) and a weaker yet more homogenous expression of AmCyan in the case of JetPrime (Figure 7h) . A detailed gating strategy is provided in (Figure S8 ). 

This study shows an approach to produce nanocarriers based on gelatin, a pharmaceutically established biopolymer, using a mild and straightforward preparation technique to load polynucleic acid cargos suitable for gene delivery. The shown improvement of colloidal stability by thermal stabilization could be essential for the further development of a product with sufficient storage stability. As expected for a system made from a biocompatible, biodegradable protein material, no cytotoxicity was observed in the concentration range successful for transfection.

A core-shell system was prepared by taking advantage of an intrinsic property of the two core components, gelatin and pDNA, to form an irreversible gel when heated together. The stability of this gel-core allowed for the deposition of a protamine sulfate shell. We loaded mRNA on the particle's shell while maintaining pDNA in the core. Transfection of both nucleic acids was observed with comparable transfection efficiencies from both pDNA and mRNA when used in a mass ratio of 5:1, as opposed to clinical, experimental, and commercial transfection reagents, where such co-transfection was not feasible.

We here present a proto-type NA carrier with unique co-transfectional capabilities. A vast pool of applications can be based on or expanded off the concept, both in the areas of vaccine delivery, as well as protein replacement therapies. In this study, we employed commercial fluorescent reporter molecules of rapid onset of expression and long-expression product half-life. Using selected combinations of nucleotides, we think the interesting possibility of gene expression at varying time scales could be achievable and should be further studied. In future studies, the potential of this system to optimize the expression kinetics and location of NA cargos could be explored. Besides clinical applications, this system could also serve as a research tool to study differences between expression kinetics of more than one NA cargo in parallel.

The following are available online at https://www.mdpi.com/article/ 10.3390/pharmaceutics13111924/s1, Figure S1 : DLS assessment of CoAc and TS-CoAc assembled using pDNA of varying sizes., Figure S2 : Screenshots of NTA analysis of CoAc and TS-CoAc immediately after addition to RPMI-1640 at 37 • C in a ratio of 1:10 v/v, and after 4 h incubation in such medium., Figure S3 : Transmission electron microscopy of unstained P-TS-CoAc, Figure S4 : Particle sizes (nm) and PDI of P-TS-CoAc before loading with mCherry, 15 min after loading with mCherry and 7 days after loading with mCherry. Figure S5 : Gel electrophoresis for assessment of serum stability of pAmCyan cargo in coacervate and TS-CoAc following 3 h incubation in 10% serum, Figure S6 : PicoGreen and RiboGreen assays' calibration curves, Figure S7 : Cytotoxicity assay of protamine coacervate, Figure S8 : Gating information of DC2.4 during the in-vitro transfection assay. 

Development and clinical trials of nucleic acid medicines for pancreatic cancer treatment

Therapeutic nucleic acids: Current clinical status

The promise of mRNA vaccines: A biotech and industrial perspective

DNA vaccines in veterinary use

Takes Key Action in Fight Against COVID-19 By Issuing Emergency Use Authorization for First COVID-19 Vaccine

COVID-19 vaccine frontrunners and their nanotechnology design

Nucleic acid delivery for therapeutic applications

Lipoplexes versus nanoparticles: PDNA/SiRNA delivery

Lyophilized chitosan nanoparticles for PDNA 766 and SiRNA delivery: Physico-chemical properties, transfection efficiency, and cytotoxicity

Strategies for simultaneous and successive delivery of RNA

Lipid nanoparticle formulations for enhanced co-delivery of SiRNA and mRNA

Latest development on RNA-based drugs and vaccines

Incorporation of pseudouridine into mRNA yields superior nonimmunogenic vector with increased translational capacity and biological stability

A facile method for the removal of dsRNA Contaminant from in vitro-transcribed mRNA

Type I interferons interfere with the capacity of mRNA lipoplex vaccines to elicit cytolytic T cell responses

Evading innate immunity in nonviral mRNA delivery: Don't shoot the messenger

Suppression of RNA recognition by toll-like receptors: The impact of nucleoside modification and the evolutionary origin of RNA

Expression of therapeutic proteins after delivery of chemically modified mRNA in mice

Co-delivery of nucleoside-modified mRNA and TLR agonists for cancer immunotherapy: Restoring the immunogenicity of immunosilent mRNA

Tailoring mRNA vaccine to balance innate/adaptive immune response

Strategies for controlling the innate immune activity of conventional and self-amplifying mRNA therapeutics: Getting the message across

Local delivery of mRNA-encoding cytokines promotes antitumor immunity and tumor eradication across multiple preclinical tumor models

Nanoparticle design to induce tumor immunity and challenge the suppressive tumor microenvironment

Antibody-antigen-adjuvant conjugates enable co-delivery of antigen and adjuvant to dendritic cells in cis but only have partial targeting specificity

Toll-like receptors on tumor cells facilitate evasion of immune surveillance

Triggering CD40 on endothelial cells contributes to tumor growth

Intratumoral delivery of TriMix mRNA results in T-cell activation by cross-presenting dendritic cells

TriMix and tumor antigen mRNA electroporated dendritic cell vaccination plus ipilimumab: Link between T-cell activation and clinical responses in advanced melanoma

The nuclear pore complex: The gateway to successful nonviral gene delivery

Transcriptional targeting of dendritic cells for gene therapy using the promoter of the cytoskeletal protein fascin

Targeting dendritic cells with antigen via dendritic cell-associated promoters

Promoter architecture dictates cell-to-cell variability in gene expression

Pros and cons of PDNA and mRNA transfection to study mRNA translation in mammalian cells

The Flt-1 promoter for transcriptional targeting of teratocarcinoma

Efficient encapsulation and release of RNA molecules from gelatin-based nanoparticles

Collagen-vs. gelatine-based biomaterials and their biocompatibility: Review and perspectives

Cationic cholesterol promotes gene transfection using the nuclear localization signal in protamine

Dendritic cell targeted liposomes-protamine-DNA complexes mediated by synthetic mannosylated cholestrol as a potential carrier for DNA vaccine

Multifunctional pharmaceutical nanocarriers

Quantifying the average number of nucleic acid therapeutics per nanocarrier by single particle tracking microscopy

Crommelin, D. mRNA-lipid nanoparticle COVID-19 vaccines: Structure and stability

Overcharging, thermal, viscoelastic and hydration properties of DNA-gelatin complex coacervates: Pharmaceutical and food industries

Methods for characterization of nanoparticles

Manufacture of gelatin/gelatin coacervate microcapsules

Temperature dependence of DNA persistence length

Effect of persistence length on binding of DNA to polyions and overcharging of their intermolecular complexes in aqueous and in 1-methyl-3-octyl imidazolium chloride ionic liquid solutions

Temperature dependence of DNA condensation at high ionic concentration

Helix-coil transition of gelatin: Helical morphology and stability

Applications of circular dichroism for structural analysis of gelatin and antimicrobial peptides

Nanofibrous smart bandages for wound care. Wound Health Biomater

Systematic adjustment of charge densities and size of polyglycerol amines reduces cytotoxic effects and enhances cellular uptake

Poly(ethylene glycol)-modified thiolated gelatin nanoparticles for glutathione-responsive intracellular DNA delivery

Delivery by cationic gelatin nanoparticles strongly increases the immunostimulatory effects of CpG oligonucleotides

DNA-protein crosslink proteolysis repair

Formation and repair of DNA-protein crosslink damage

DNA-protein cross-links: Formation, structural identities, and biological outcomes

DNA-gelatin complex coacervation, UCST and first-order phase transition of coacervate to anisotropic ion gel in 1-methyl-3-octylimidazolium chloride ionic liquid solutions

Coacervation in biopolymers

Assessment of the characteristics and biocompatibility of gelatin sponge scaffolds prepared by various crosslinking methods

Cytotoxicity evaluation of gelatin sponges prepared with different cross-linking agents

A phase I/IIa study of the mRNA-based cancer immunotherapy CV9201 in patients with stage IIIB/IV non-small cell lung cancer

Self-adjuvanted mRNA vaccination in advanced prostate cancer patients: A first-in-man phase I/IIa study

Fotin-Mleczek, M. A novel, disruptive vaccination technology

Safety and immunogenicity of a mRNA rabies vaccine in healthy adults: An open-label, non-randomised, prospective, first-inhuman phase 1 clinical trial

Phase Ib evaluation of a self-adjuvanted protamine formulated mRNA-based active cancer immunotherapy, BI1361849 (CV9202), combined with local radiation treatment in patients with stage IV non-small cell lung cancer

Protamine as an efficient membrane-translocating peptide

Protamine increases transfection efficiency and cell viability after transfection with calcium phosphate nanoparticles

Evaluation of nuclear transfer and transcription of plasmid DNA condensed with protamine by microinjection: The use of a nuclear transfer score

Protamine sulfate enhances lipid-mediated gene transfer

Agarose research grade was purchased from Serva ® , Heidelberg, Germany. Disodium dihydrate ethylenediamine tetra-acetic acid (EDTA-Na 2 ) was purchased from Roth GmbH+Co. KG, Karlsruhe, Germany. DNA loading dye (6×) was purchased from Thermo Fisher Scientific, Waltham, MA, USA. DNA Ladder 250-10,000 bp was purchased from PEQLAB Biotech GmbH, Erlangen, Germany. Ethidium bromide 10 mg·mL −1 was purchased from Sigma-Aldrich, Darmstadt, Germany. Live/dead fixable stain (568/583) was purchased from PromoCell GmbH, Heidelberg, Germany. Quant-iT™ PicoGreen™ dsDNA Assay Kit and RiboGreen™ RNA Assay-Kit, DNase I, DNase I buffer, 50 mM EDTA, RNase A, and Ribolock were purchased from Thermo Fisher Scientific, Darm-Acknowledgments: The authors deeply acknowledge Jörg Schmauch at Saarland University, Experimentalphysik INM-Leibniz-Institute for New Materials for the TEM experiments. The authors acknowledge Hanzey Yasar and Jana Westhues for assistance with in-vitro assessment protocols, Robin Gierse for assistance with Circular Dichroism assay, as well as Marius Hittinger for logistic support. The graphical abstract and (Figures 1 and 2a) were prepared using Biorender.

The authors declare no conflict of interest.