TYPE Original Research PUBLISHED 09 May 2023 DOI 10.3389/fimmu.2023.1128582 Exploiting endocytosis for OPEN ACCESS transfection of mRNA for EDITED BY Kirill Afonin, University of North Carolina at Charlotte, cytoplasmatic delivery using United States REVIEWED BY cationic gold nanoparticles Rong Hai, University of California, Riverside, United States Muriel F. Gustà 1,2,3, Michael J. Edel 4,5,6, Vivian A. Salazar 1, Massimo Masserini, University of Milano-Bicocca, Italy Belén Alvarez-Palomo 7, Manel Juan 4, Massimo Broggini 8, *CORRESPONDENCE Giovanna Damia 8, Paolo Bigini 8, Alessandro Corbelli 8, Neus G. Bastús Fabio Fiordaliso 8, Alexander Barbul 9, Rafi Korenstein 9, neus.bastus@icn2.cat Vı´ctor Puntes Neus G. Bastús 1,3* and Vı́ctor Puntes 1,2,3,10* victor.puntes@icn2.cat 1 Institut Català de Nanociència i Nanotecnologia (ICN2), Consejo Superior de Investigaciones RECEIVED 20 December 2022 Científicas (CSIC), The Barcelona Institute of Science and Technology (BIST), Barcelona, Spain, ACCEPTED 21 April 2023 2 Vall d’Hebron Institut de Recerca (VHIR), Barcelona, Spain, 3 Biomedical Research Networking Center PUBLISHED 09 May 2023 in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Madrid, Spain, 4 Hospital Clı´nic de Barcelona, Servei Immunologia-IDIBAPS, Barcelona, Spain, 5 Unit of Anatomy and Embryology, CITATION Universitat Autònoma de Barcelona, Faculty of Medicine, Barcelona, Spain, 6 University of Western Gustà MF, Edel MJ, Salazar VA, Alvarez- Australia, Faculty of Medicine, Discipline of Medical Sciences and Genetics, School of Biomedical Palomo B, Juan M, Broggini M, Damia G, Sciences, Perth, WA, Australia, 7 Banc de Sang i Teixits, Cell Therapy Service, Barcelona, Spain, Bigini P, Corbelli A, Fiordaliso F, Barbul A, 8 IRCCS‐Istituto di Ricerche Farmacologiche Mario Negri, Milano, Italy, 9 Tel Aviv University, Sackler Korenstein R, Bastús NG and Puntes V School of Medicine, Tel Aviv-Yafo, Israel, 10Institució Catalana de Recerca i Estudis Avançats (ICREA), (2023) Exploiting endocytosis for Barcelona, Spain transfection of mRNA for cytoplasmatic delivery using cationic gold nanoparticles. Front. Immunol. 14:1128582. doi: 10.3389/fimmu.2023.1128582 COPYRIGHT Introduction: Gene therapy holds promise to cure various diseases at the © 2023 Gustà, Edel, Salazar, Alvarez-Palomo, fundamental level. For that, efficient carriers are needed for successful gene Juan, Broggini, Damia, Bigini, Corbelli, delivery. Synthetic ‘non-viral’ vectors, as cationic polymers, are quickly gaining Fiordaliso, Barbul, Korenstein, Bastús and Puntes. This is an open-access article popularity as efficient vectors for transmitting genes. However, they suffer from distributed under the terms of the Creative high toxicity associated with the permeation and poration of the cell membrane. Commons Attribution License (CC BY). The use, distribution or reproduction in other This toxic aspect can be eliminated by nanoconjugation. Still, results suggest that forums is permitted, provided the original optimising the oligonucleotide complexation, ultimately determined by the size author(s) and the copyright owner(s) are and charge of the nanovector, is not the only barrier to efficient gene delivery. credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, Methods: We herein develop a comprehensive nanovector catalogue distribution or reproduction is permitted comprising different sizes of Au NPs functionalized with two different cationic which does not comply with these terms. molecules and further loaded with mRNA for its delivery inside the cell. Results and Discussion: Tested nanovectors showed safe and sustained transfection efficiencies over 7 days, where 50 nm Au NPs displayed the highest transfection rates. Remarkably, protein expression was increased when nanovector transfection was performed combined with chloroquine. Cytotoxicity and risk assessment demonstrated that nanovectors are safe, ascribed to lesser cellular damage due to their internalization and delivery via endocytosis. Obtained results may pave the way to design advanced and efficient gene therapies for safely transferring oligonucleotides. KEYWORDS gold nanoparticles, cationic, transfection, gene therapeutics, safety Frontiers in Immunology 01 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 Introduction smaller complexes with mRNA, intended to prevent degradation and facilitate cytosolic mobility. Besides, approaches that work for Developing efficient gene therapies depends on the means for one cell line might not perform well for others, suggesting that the transferring oligonucleotides (DNA or RNA) into the cell. The most mechanisms that NPs use for cell entry and cell trafficking are common vectors used are replication-defective vector systems based essential factors to consider when designing efficient transfection on two types of viruses: retroviruses and adenoviruses. These nanovectors (18). vectors have been engineered to drastically reduce the The second aspect is the cellular uptake of the nanovector. transcriptional activity of the virus, virtually eliminating the Endocytosis is the primary mechanism for the uptake of possibility of viral reactivation. However, uncontrolled integration nanovector- oligonucleotide complexes (19). During this process, into the genome can potentially lead to insertional mutagenesis if the loaded nanovector is engulfed by the cell membrane and the integration of vector DNA into host cells is placed near an delivered into the cell within a vesicle. The internalization process oncogene, posing serious concerns in their clinical application (1). depends on nanovector size and charge. The highest cellular Additionally, their production is complex, and their production internalization is observed for NPs in the size ranges of 25–50 under Good Manufacturing Practices is burdened with strict nm (20). Alternatives for cell entry other than endocytosis are regulations. Last but not least, DNA-insertion therapies in the membrane permeation and membrane fusion, which often result in context of CAR-T cell therapy, where ex-vivo transfection high toxicity, especially permeation (21, 22). Similarly, surface reprograms T cells to destroy B cells in the case of leukaemia, charge determines cellular uptake. The negatively charged cell deplete patients of B cells for life with the corresponding health and membrane enhances the uptake of positively charged NPs. societal burden. As an alternative, the idea of mRNA transfection However, the uptake of positively charged NPs may disrupt the will lead to transient cell therapy, dramatically reducing genetic integrity of the cell membrane and lead to an increase in toxicity therapy side effects. In addition, the transfection with mRNA has to (23) unless size and structure are provided. be performed in the cytoplasm, opening the venue for non-viral The third aspect is the endosomal escape of the nanovector. transfection vectors. Once inside the cell, the nanovector complex is trapped within a Unlike viral vectors, nonviral ones rely on forming noncovalent vesicle. Several mechanisms have been identified to escape the assemblies between mRNA (a polyanion), cationic polymers, and endosome into the cytosol, the most popularly known as the lipidic moieties. Although a leading class of synthetic gene-delivery “proton sponge mechanism” (24). The proton sponge mechanism vehicles, cationic polymers suffer from high toxicity, and their relies on the fact that present amines are not protonated under efficiency does not compare to viral systems (2). To overcome physiological conditions. These basic moieties can buffer the these limitations, the delivery of oligonucleotides using decreasing pH within the endosome. As more protons are nanovectors, a nanoparticle-based transfection vector, is attracting pumped, more counter ions (mainly chloride) influx into the increasing attention (3–6). This includes lipidic and polymeric NPs endosome for electroneutrality. This fact increases the osmotic composed of biocompatible units that self-assemble encapsulating pressure and, ultimately, the passive diffusion of water into the nucleic acids. Also, inorganic NPs which can be functionalized with endosome. Consequently, the endosome continues to swell until the DNA or mRNA and can be used for in vitro transfection increasing membrane stress leads to membrane rupture and a applications. Among the candidates, Au nanoparticles (NPs) are release of the contents. remarkably interesting due to their small size and monodispersity, In this context, the study’s main objective is to design and low cytotoxicity, low immunogenicity, biocompatibility, develop transfection nanovectors for releasing nucleic acids as straightforward synthesis and easy functionalization (7). Indeed, messenger RNA (mRNA) inside the cell ready for transfection. the chemical functionalization of Au NPs with cationic molecules The introduction of mRNA by nonviral transfection vectors allows (5, 8–12), especially polyethylenimine (PEI) molecules (13, 14), the gene’s transient expression, which presents several medical favours efficient oligonucleotide adsorption by high-affinity advantages. For instance, note that people treated with DNA electrostatic interactions, ultimately enabling transfection. In any CAR-T cell immunotherapy get deprived of B cells for life (25, case, the interactions between the innate immune system and 26). The nanovector consists of a cationic Au NP where mRNA is nanoparticles and derived objects can be especially immunogenic. adsorbed for further delivery into the cytosol. There are two critical Thus, purity, solubility, and surface state has to be precisely points in the development of the nanovector: i) the tight absorption controlled (15). of enough mRNA and ii) the efficient cytoplasmic release of the The efficient loading of the nanovectors is ultimately mRNA following the “proton sponge” effect. The number of mRNA determined by its size and charge. By adjusting its size, its loading molecules adsorbed onto the nanovector is expected to depend on capacity is modified (16). Thus, the load vs carrier ratio increases as the size of the nanovector and the density of amino groups present the vector becomes smaller. Similarly, the loading depends on the at their surface, and this density will also determine the efficiency of charge of the nanovector. Cationic polymers possess many positive the cytoplasmic release. In addition, the absorption of the mRNA charges providing strong interactions with oligonucleotides, which onto the NP, as occur with other molecules as proteins, protects are negatively charged. As simple as that, results suggest that them from degradation. optimising the oligonucleotide complexation is not the only Thus, a nanovector catalogue has been developed comprising barrier to gene delivery (8, 9, 13, 14, 17). A higher grafting different sizes of Au NP cores, later functionalized with two density of PEI does not always result in more compact and different cationic molecules. These nanovectors were further Frontiers in Immunology 02 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 loaded with mRNA for their application as nonviral transfection acid (AUT) or polyethyleneimine 2kDa (PEI). Both cationic vectors. Cationic molecules have already been used as transfection molecules contain amine terminal groups providing a positive agents, but with significant toxicity concerns (27–30). However, its charge to the Au NPs at physiological pH while differing in their absorption to the NP’s surface promotes the loss of flexibility and molecular structure and density of amino groups. While AUT the membrane pore formation ability. Instead, PEI-derived NPs consists of an 11-carbon chain with a terminal amine and a thiol (nanovectors) attach to the cell membrane, not crossing it but group in the opposite site that pseudo-covalently binds to gold. The inducing endocytosis. Consequently, its toxic aspect is eliminated employed PEI is a branched polymer that binds electrostatically to thanks to nanoconjugation. Indeed, detoxification by the NP. Linear PEI was also tested but increased toxicity without nanoconjugation has been observed before in different systems increasing efficacy (data not shown). (31). Thus, the monolayer coverage of Au NPs allows for tuning Citrate-stabilized Au NPs were produced using a well- the charge and structure to maximize transfection efficiency while established seeded growth approach based on the citrate reducing associated toxicity. reduction of HAuCl4 (38, 39). Citrate is a good capping agent because it only binds loosely at the Au NP surface, being easily replaced by other ligands such as thiol- or amine-containing Results and discussion ligands, which have higher binding affinities for Au surfaces (~ 45 Kcal/mol (40) and ~ 6 Kcal/mol (41), respectively). Representative Synthesis, cationic functionalization and transmission electron microscopy (TEM) images of 5 nm, 20 nm mRNA loading of citrate-stabilized Au NPs and 50 nm Au NPs are shown in Figure 1, revealing their high monodispersity and quasi-spherical morphology. Citrate-stabilized The first critical point in the design of the nanovector is the Au NPs are stable in aqueous media and display a well-defined efficient adsorption of mRNA for their further delivery into the surface plasmon resonance (SPR) peak that red-shifts and increases cytosol. The number of molecules adsorbed onto the nanovector in intensity as NPs increases in size. depends on the size and density of amino groups at their surface. As synthesized citrate-stabilized Au NPs were functionalized Therefore, a nanovector catalogue comprising different sizes of Au with AUT and PEI by incubating them with an excess of the NPs functionalized with two different cationic molecules and cationic molecule at pH 2.5 for 24 hours. Remarkably, under further loaded with mRNA was developed. By adjusting the size these extremely acidic conditions, the stability of citrate-capped of the Au NPs and the nature of the cationic molecule, their loading Au NPs is compromised due to loss of electrostatic repulsion, and capacity was modified. Characterization by UV-Vis spectroscopy, aggregation is only prevented by the effective conjugation of the DLS and Z-Potential was performed at each step and summarized selected molecules. Citrate has three pKa points which are 6.3, 4.7, in Table 1. The sizes of choice were 5, 20 and 50 nm as this size and 3.1. Therefore, below pH~ 3, all the hydroxyl residues become regime favours NP’s endocytosis (32–35) and presents proven protonated, losing their negative charge, and, in turn, NPs lose the stability in physiological media (36, 37). As synthesized Au NPs electrostatic repulsion provided by the citrate capping. The presence were further functionalized with either 11-amino-1-undecanethiol of a self-assembled AUT monolayer provides colloidal stability to TABLE 1 Characterization of the nanovector catalogue. NPs NPs-AUT NPs-AUT+RNA NPs-PEI NPs-PEI+RNA Au 5 nm SPR (nm) 510 521 533 516 519 DLS (nm) 6.1 ± 1.2 11.6 ± 3.6 62.7 ± 25.5 15.6 ± 4.9 15.1 ± 3.6 Z-Pot (mV) -59.7 +5.8 -10.3 +26.2 +9.3 Au 20 nm SPR (nm) 522 526 531 523 524 DLS (nm) 30.4 ± 10.6 37.0 ± 13.3 54.6 ± 22.9 38.9 ± 15.3 44.9 ± 19.6 Z-Pot (mV) -44.4 +18.6 -28.6 +34.4 +21.7 Au 50 nm SPR (nm) 531 533 540 531 534 DLS (nm) 49.6 ± 14.5 58.8 ± 25.9 75.0 ± 35.7 67.5 ± 30.2 72.8 ± 26.4 Z-Pot (mV) -41.2 +31.2 -25.7 +44.3 +34.9 Summary of sizes, optical properties and surface charge of citrate-capped Au NPs of 5, 20 and 50 nm, cationic functionalized Au NPs with AUT of PEI, and loaded with mRNA. Note that the pH at which z-Potential of the citrate-capped NPs was measured (pH~8.6) is different from the other measurements, where NPs were dispersed in MES buffer (pH~5). This fact fundamentally impacts the surface charge of the NP. Frontiers in Immunology 03 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 FIGURE 1 Nanovector Catalogue. Representative transmission electron microscopy images of highly monodisperse citrate-stabilized Au NPs of ∼5 nm, ∼20 nm, and ∼50 nm. UV-Vis spectra of citrate-capped Au NPs of 5, 20 and 50 nm, cationic functionalized Au NPs with AUT of PEI, and loaded with mRNA. Absorbance is normalized to l400 nm for a better comparison. NPs through the positively charged amine-terminal residue that measurements, revealed that cationic functionalization leads to avoids aggregation by electrostatic repulsion. Therefore, highly positively charged Au NPs. Independently of their size, an appropriate conjugation conditions are critical to retaining NP increased surface charge is observed for PEI than AUT conjugates, stability during the coating process (see Figure S1, Figure S2). explained by the higher amine density of the PEI coating due to its Characterization by UV-Vis spectroscopy indicates the effective branched polymeric nature (44). conjugation of Au NPs to cationic molecules. The absorption Functionalized NPs were purified and redispersed in MES spectrum is sensitive to the Au NP environment, and an buffer (pH ~ 5.5) for their later loading with oligonucleotides. observable red-shift of the SPR band of about 6-10 nm can be Based on our previous expertise (36, 37, 45), we added the seen once the Au NPs are functionalized with the cationic molecules cationic NPs to the oligonucleotide solution to maximize NPs (42, 43). The shift occurs within a few minutes and then remains surface coverage while avoiding uncontrolled aggregation during unaltered, suggesting that the conjugation process is quick and that the mixture of both solutions (46). The ratio of oligonucleotide NPs are stable for long periods. The extent of the redshift depends molecules to NP was optimized for each NP’s size tested to achieve on the Au NPs size, the structure of the cationic molecule and the the maximum loading while avoiding NP aggregation (see Figure nature of the anchoring group (42). Thus, smaller Au NPs S3). The incubation was performed at 4 °C for 24 hours to minimize functionalized with AUT molecules bond via thiol groups exhibit the risk of nucleic acid degradation while ensuring surface the largest SPR shift. Although PEI is relatively big, it interacts saturation (Figure S4 shows complete kinetics of the loading electrostatically with NP’s surface by the positively charged amine process). Samples were purified before characterization analysis. residues, not forming covalent bonds, ultimately leading to smaller As seen in Figure 1, at any given NP size and cationic coating, a SPR shifts. red shift in the SPR peak position of about 1-2 nm can be observed, The functionalization process was further assessed by dynamic confirming the effective oligonucleotide loading. Accordingly, their light scattering (DLS). An apparent increase in the hydrodynamic hydrodynamic diameter increased, and their z-potential dropped size of the NPs is observed after conjugation. Remarkably, due to the presence of the negatively charged mRNA molecules at monomodal distribution profiles were obtained in all cases, the acidic pH of work (47). Yet, the increase in the hydrodynamic indicating that cationic Au NPs are colloidally stable. diameter is larger for PEI than for AUT-functionalized Au NPs. Measurement of NP surface charge, performed by z-potential These results, coupled with the fact that the surface charge doesn’t Frontiers in Immunology 04 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 invert to negative values in the case of PEI-functionalized Au NPs, standard transfection methods, such as TransIT ® or point out that significant differences attributed to the disposition of Lipofectamine®, show good transfection rates but have very low the cationic coating at the NP’s surface could exist. As previously cell viability. It is claimed that their internalization pathway is mentioned, AUT forms a regular self-assembled monolayer, through membrane disruption, which ultimately causes their whereas PEI, with a branched polymeric nature and high toxicity (48). In contrast, nanovectors internalize via molecular weight, is most likely in a mushroom configuration. endocytosis, significantly reducing the associated toxicity. To The fact that PEI-functionalized Au NPs remain positive after evaluate the safety of nanovectors in HEK293 cells, we first oligonucleotide loading has a significant impact on the interaction measured the cell viability by Resazurin reduction using of the nanovectors with the cell surface and has a critical role in Prestoblue and then Annexin V/PI staining was performed to their transfection efficiency. Finally, aiming to calculate the loading analyze cell viability. (Figure 2). of oligonucleotides to cationic Au NPs, quantification was Some previous studies reported that Au NPs cytotoxicity is performed by Nanodrop. For this, after the loading process, Au mediated by their size, which inhibited the proliferation and NPs were purified, and the supernatant was analyzed. triggered cell cycle arrest (49–51). However, the present results Spectrophotometric results reveal a loading of 48% (9.3 ng/µl) for suggest no toxic effect related to NP size, at last in the studied NPs-AUT and 45% (8.7 ng/µl) for NPs-PEI for 50 nm Au NPs. All range. Yet, cationic functionalization of the Au NPs did in all, the presented results, along with the nanovector stability compromise cell viability and membrane integrity in some (Figure S5) and release studies (Figure S6), provided a deep cases. A slight decrease in cell viability in HEK293 cells was understanding of the nanovectors properties and proved the high determined with an exposition of 5 nm Au NPs coated with PEI stability of the nanovector-oligonucleotide complex. and PEI-RNA at the maximum concentration of 3.3x10 (13) NP/ mL. On the contrary, AUT-coated 5nm Au NP do not show any cytotoxic effect. Regarding 20 nm and 50 nm Au NPs, non- Nanovector cytotoxicity studies significant variation in the cell viability between Au-treated and non-treated cells was quantified even when exposing the cells to A key point in the success of nanovectors as nonviral the highest concentration, corresponding to 2.7x1012 NP/mL transfection vehicles must be their low cytotoxicity. Other and 3x1011 NP/mL, respectively. A B FIGURE 2 Nanovector Cytotoxicity Studies. The action of Au NP coated with PEI and AUT, and loaded with oligonucleotides was tested at different concentrations. (A) The percentage of resazurin reduction is shown at the maximum Au NP’s concentration. The dashed-black line corresponds to control values for reference, and the grey area overlaid to its relative standard deviation. (B) After 24h, 48h, or 7 days of HEK293 cells exposition to cationic NPs, alone or loaded with mRNA, Annexin V/PI staining was performed to analyze cell viability. Cell state was classified as live, in early or late apoptosis, or dead according to the relative intensity of each marker. Frontiers in Immunology 05 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 To gain further insight into transfection-associated toxicity, for their use as delivery vectors. Cationic AuNPs non-loaded with HEK293 cells were exposed to nanovectors and sorted based on DNA rapidly absorb negatively charged proteins from the medium their viability state. Annexin V/PI staining was used to categorize losing their cationic charge and may cause uncontrolled aggregation cell populations as live, in early or late apoptosis, and dead. Viability (Figure S5, Table S1). controls were also included, cells without any treatment for the negative control and lethal H2O2 doses for positive cell death control. In this experiment, TransIT® was also included to Endosomal escape compare its toxicity with the NP-based nanovectors carrying similar amounts of oligonucleotides. Results show an evident Reaching the cytoplasmic space is critical for the delivery of difference in the viability profile of TransIT® compared to mRNA into the cytosol, where the ribosomes that will translate it nanovectors at short times. Cationic NPs, either loaded with into the coded protein are located. NPs with sizes ranging 5-100 nm mRNA or not, present a cell population distribution similar to enter the intracellular space via endocytosis, but the subcellular fate the negative control. However, for TransIT® exposed cells, the of the NPs will depend on their properties (32, 52). Upon vesicle proportion of cells in a pro-apoptotic state is much higher, formation, they enter into the endocytic pathway, where pH comparable with those treated with H2O2. The percentage of early progressively decreases with the vesicle maturation process for and late apoptosis was maintained at 48 h. Accordingly, the dead digestion. However, NPs with pH buffering capacity can inhibit cell proportion increased at each incubation time. In contrast, cells this process, disrupt the endosomal membrane, and escape from the transfected with the nanovectors preserved their viability, and the endosome by the so-called proton sponge effect (53–55). Two apoptosis entrance was delayed. At 7 days post-exposure, viability parallel studies were performed to investigate the ability of the values stabilized. Only a non-significant slight increase in the pro- nanovectors to escape from the endosome (Figure 3). apoptotic populations can be observed for TransIT® and 5 nm Au First, the proton buffering capacity was tested in vitro by NPs-PEI. These results confirm that NP-based formulations are safe monitoring a solution’s acidification in cationic Au NPs. The A B FIGURE 3 Endosomal Escape. (A) The proton buffer capacity of the cationic-coated Au NPs was analyzed by measuring the decrease in pH as a function of HCl volume added to the solution. (B) Au NP-PEI Intracellular Trafficking on HEK293 cells. Representative transmission electron microscopy images from HEK293 cells exposed to PEI-coated 20 nm Au NPs. The proposed internalization pathway is schematically represented at the different stages: (1) membrane attachment, (2) internalization in endosomes, maturation to (3) amphisomes and later (4) lysosomes. (5) Shows NPs dispersed in the cytoplasmic space after escaping from the endosomes. Red arrows indicate time evolution. Frontiers in Immunology 06 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 proton sponge efficiency of the NPs was evaluated as a function of protocol with nanovectors (Figure S7). It was observed that the the HCl added; thus, H+ needed to reach the same pH value. As transfected population increased with the amount of mRNA observed, PEI and AUT-coated NP solutions present a delay in the indicating that despite the high amounts of mRNA, the ribosomes pH drop compared to the aqueous solution used as a reference. In were far from saturation, probably due to the slow and sustained detail, the Au NPs-PEI show a much greater ability to capture release of mRNA in the cytosol. protons than NPs-AUT. This correlates with the amine Cellular viability was evaluated by flow cytometry using concentration of the PEI-coated NPs being higher. These results Annexin/PI combined staining to study apoptosis phenomena in were correlated with the findings from the NP intracellular those expressing GFP cells to gain further insights into the trafficking study. The observation of cells exposed to cationic NP transfection events. Thus, GFP-positive cells were sorted into live, by TEM reveals that NPs follow an internalization pathway via early or late apoptosis, or dead depending on the relative intensity of endocytosis. After NPs-PEI stuck to the cell membrane, they were both viability markers. Figure 4B shows the expression in live cells internalized in endocytic vesicles. In a typical endocytosis process, at 24h, 48h and 7 days. HEK293 cells transfected with TransIT® the maturation of these vesicles has several steps, from early and late show the highest GFP fluorescence signal but with relatively high- endosomes to amphisomes, to finally reach the active lysosome intensity variability within the same population. Regarding the PEI- stage, where the proteases digest the cargo. In this case, it could be nanovectors, 50 nm NP show the highest efficiency on all time observed that close to 85% of the internalized NPs-PEI were found points, and GFP expressions are maintained for 7 days, while that either in amphisomes or in lysosomes. However, the 15% remaining from TransIT® has started decreasing. were located dispersed in the cytoplasmic space. This fact entails Interestingly, when GFP expression is reviewed as the that NPs-PEI escaped from the endocytic pathway at some point. proportion of live to total transfected cells (Figure 4C), differences As discussed before, the coating of Au NPs with PEI confers between TransIT® and nanovectors dramatically change. These proton sponge capacities allowing for an endosomal escape, as results reveal a high and prolonged transfection efficiency of results suggest. However, the loading of cationic NP with HEK293 cells by PEI-nanovectors, coupled with low cytotoxicity, oligonucleotides and their interfacing with biological media may since the apoptosis cell ratio is not enhanced. It can be seen that the impact on their interation with cells. In fact, the interaction with amount of healthy transfected cells is similar in both cases, components from biological fluids -as the potential formation of a TransIT® and AuNPs, but while there is a large population of protein corona- could influence nanovectors cell internalization, transfected apoptotic cells with the former, they are almost none in distribution and fate (56). In our case, cells are transfected in serum the latter, especially at short times. Additionally, the observation of free medium avoiding interaction with proteins in the cell culture. the samples under the CLSM enabled the image of the Au NPs simultaneously by reflectance mode (57–59) (Figure S8). These findings, coupled with the TEM observations of the NPs Transfection: GFP expression intracellular trafficking, may indicate that despite most of the cells having internalized nanovectors, they are not yet dispersed in the The effectiveness of Au transfection in HEK293 cells was cytoplasmic space. Therefore, endosomal escape may happen later assayed using 5, 20 and 50 nm Au NP coated with PEI and AUT, for these nanovectors, delaying the mRNA delivery and expression, loaded with green fluorescence protein, GFP mRNA. The GFP which also agrees with the sustained GFP expression observed by expression was analyzed at short (24 and 48 h) and long (7 days) flow cytometry. Additionally, variability in GFP expression could be times post-transfection, by flow cytometry and fluorescence ascribed to differences in the cell cycle stage between the HEK293 microscopy. The observation of the transfected cells by population (60). fluorescence microscopy allowed for a first visual inspection of Aiming to test the versatility of the developed nanovectors, the samples to evaluate cell morphology and GFP expression different cells lines were also trasnfected with GFP mRNA loaded on qualitatively. In addition, the transfection efficiency from the total 5, 20 and 50 nm Au NP-PEI. The transfection efficiency of the cell population was measured by flow cytometry. nanovectors on Jurkat cells, an immortalized line of human T Figures 4A, B shows the transient expression of GFP. While all lymphocyte cells that are used to study acute T cell leukaemia and the Au NP-PEI could transfect the HEK293 cells, the transfection RAW264.7 as a model of macrophage has been tested. Immune performed with the AUT-coated nanovectors was unsuccessful cells, and particularly T cells morphology is specially challenging for since no GFP expression was observed because the NPs did not cytoplasmatic delivery since the majority of the cell volume is reach the cytosol, either because they did not release the mRNA. occupied by the cell nucleous. Obtained results demonstrate the Interestingly, the expression of the GFP protein was preserved for 7 capacity of the nanovectors to deliver mRNA to different cell lines days which could be adscribed to a slow mRNA from the (Figure S9). nanovector (Figure S6). Some morphological and adherence changes were observed, especially after exposition to 5nm NPs- PEI and the commercial reagent TransIT®, consistent with necrosis Transfection: Cloroqiuine Effect as reported in the cytotoxicity studies. Additionally, to evaluate ribosomal occupancy and saturation different concentrations of Here, the combined use of nanovectors with chloroquine is mRNA were tested, which allowed to standardize the transfection explored to enhance the transfection efficiency (Figure 5). Frontiers in Immunology 07 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 A B C FIGURE 4 GFP Expression. (A) Wide-Field fluorescence overlaid onto Bright-Field images of transfected HEK293 cells at 24h, 48h, and 7 days. Scale bar = 150 µm. (B) Quantifying the nanovectors transfection based on the mean GFP expression at 24h, 48h, and 7 days. Cell viability was analyzed by Annexin V/PI iodine staining, and the cell state was classified as live, in early or late apoptosis, or dead. (C) Summary of the GFP-expressing live cells relative to the total GFP-expressing cells. Chloroquine, as largely reported in the literature, is a common Conclusions compound used to halt endosomal maturation, which in turn acts to boost NP endosomal escape by enhancing the proton sponge effect In this work role of cationic functionalized Au NPs as (61, 62). So, HEK293 cells were treated with chloroquine for 4 hours transfection vectors has been explored. Au NPs of 5 nm, 20 prior to transfection with PEI-nanovectors and GFP expression at nm and 50 nm were synthesized with high monodispersity 24h was visualized by wide-field fluorescence microscopy. following a previously reported seeded-growth method. NPs The statistical analysis from the GFP expression quantification were successfully coated with either AUT or PEI to provide shows a significant fluorescence signal increase for 5nm and 50nm cationic surface charge and then loaded with mRNA, which PEI-nanovectors, and higher but not so significant for 20nm. constitutes the nanovector construct. We studied stability in Conversely, no enhancement effect was observed in cells biological media and the ability to release the previously loaded transfected with AUT-coated nanovectors (data not shown). mRNA over time. Nanovectors also displayed proton sponge Accordingly, these results support the hypothesis that Au NPs- capacity, and further microscopy studies suggested their delivery PEI reaches the cytoplasmic space via endosomal escape by the into the cytosol via endosomal escape. The use of PEI-coated Au proton sponge mechanism. Further, this study provides new NPs as mRNA delivery vectors showed a relatively low insights into the characterization of nanovectors that ensure transfection efficiency but sustained GFP protein expression suitable mRNA delivery into the cells without a cytotoxic effect. over 7 days, where 50 nm Au NPs displayed the highest Frontiers in Immunology 08 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 A B FIGURE 5 Effect of Chloroquine in transfection. (A) Transfection efficiency based on the mean GFP expression at 24h. (* p>0.05) (B) Wide-field Fluorescence images of HEK293 cells transfected with PEI-coated nanovectors, alone or in combination with chloroquine, at 24h. Scale bar = 150 µm. transfection rates. Cytotoxicity and risk assessment performed Materials and methods by Prestoblue and Annexin V/PI assays demonstrated are safe and didn’t induce significant cell damage, except for 5 nm Au- Chemicals PEI NPs that showed higher toxicity, similar to TransIT transfection control. Protein expression was increased when Gold(III) chloride trihydrate (HAuCl4·3H2O), trisodium citrate nanovector-mRNA transfection was performed combined with (Na3C6H5O7), tannic acid (C76H52O46), potassium carbonate chloroquine. The mRNA transfection with NPs induced less (K 2 CO 3 ), amino-undecanethiol (AUT), poly-ethyleneimine cellular damage and mortality due to their internalization and branched Mn2000 (PEI), 2-(N-morpholino)ethanesulfonic acid delivery via endocytosis. All in all, cationic Au NPs were proven buffer solution (MES), sodium hydroxide (NaOH), hydrogen to be safe, nonviral vectors for mRNA delivery into cells, with a chloride (HCl), oligonucleotide model 600-800 bases (D1626), wide margin for improvement. Sodium Phosphate Dibasic (Na 2 HPO 4 ), Sodium phosphate In conclusion, the introduction of mRNA into the cells for monobasic (NaH2PO4), poly-L-lysine, Paraformaldehyde (PFA), protein overexpression is an alternative to viral vectors that Triton-X, Bovine Serum Albumin (BSA), Sodium Chloride includes several advantages: no threat of mutagenic insertion, (NaCl), and Calcium Chloride (CaCl2) were purchased from no threat of viral particle reactivation, accessible to dose Sigma-Aldrich. Dulbecco’s Modified Eagle Medium (DMEM), control, and synthetic animal product-free production, which Foetal Bovine Serum (FBS), Hoechst 3342 (H1399), Prolong altogether makes it an attractive approach for clinical use since antifade mounting medium (11559306), Optimem Medium, they allow for a transient expression of the desired gene. Not Pacific Blue-Annexin V, Propidium iodide (PI), accutase, 4-(2- only that, but non-viral vectors have gained importance in hydroxyethyl)-1-piperazineethanesulfonic acid buffer (HEPES), recent years because of their safety in handling and ease of and Prestoblue were purchased from Thermo Fisher. Phalloidin application compared with viral vectors. In contrast, they are Alexa Fluor 647 (ab176759) was purchased from Abcam. Clean customizable, non-pathogenic, relatively safe, and easily CAP eGFP mRNA (5 moU) was purchased from Tebu-Bio. produced and scaled-up. TransIT ® -LT1 Transfection Reagent was purchased from Frontiers in Immunology 09 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 MirusBio. All chemicals were used as received without further positive charges of the amine residues of AUT interact with the purification. Distilled water passed through a Millipore system negatively charged hydroxyl residues of citrate and crosslink (r = 18.2 MW) was used in all experiments. All glassware was triggering NPs aggregation. The conjugation time was analyzed first rinsed with acetone and then with Millipore water before use. by monitoring the NPs by UV-Vis from 5 min to 1 month. Finally, the purification process of the AUT-coated NPs was studied. The conjugated NPs were precipitated by centrifugation twice, and Gold nanoparticle synthesis resuspended to the initial volume, first with HCl 2 mM and then with MES 10 mM. 5 nm citrate-stabilized Au NPs were producted following Piella Functionalization of Gold Nanoparticles with PEI. The optimal et al. In detail, a 150 mL of freshly prepared reducing solution of PEI concentration and pH were studied for Au NPs PEI-coating. sodium citrate (SC, 2.2 mM) containing 0.1 mL of tannic acid (TA, On the first case, 50nm Au NPs were concentrated 10-fold relative 2.5 mM) and 1 mL of potassium carbonate (K2CO3, 150 mM) was to the synthesis concentration by centrifugation. Next, PEI heated with a heating mantle in a 250 mL three-necked round- solutions with concentrations ranging between 50-200 µM were bottom flask under vigorous stirring. When the temperature prepared in HCl 34mM (pH~7). NPs (10% to final volume) were reached 70°C, 1 mL of tetrachloroauric acid (HAuCl4, 25 mM) rapidly added into the PEI solution under vigorous stirring. After was injected. The colour of the solution changed rapidly to black- 1h, samples were characterized by UV-Vis. 10-fold concentrated grey (less than 10 s) and then to orange-red in the following 1−2 50nm Au NPs were conjugated to PEI (200 µM) at different pH min. The solution was kept at 70°C for 5 min more to ensure conditions ranging from 2 to 7. NPs (10% to final volume) were complete reaction of the gold precursor. Immediately after the rapidly added into the PEI solution under vigorous stirring. After synthesis and in the reaction same vessel, the sample was diluted 24h, samples were characterized by UV-Vis. The conjugated NPs by extracting 55 mL and adding 55 mL of SC (2.2 mM). When the were precipitated by centrifugation, resuspended to the initial temperature reached again 70°C, two injections of 0.5 mL of volume with water and characterized again by UV-Vis. HAuCl4 (25 mM) on a time interval of 10 min were done. This growing step comprising sample dilution plus 2 injections of HAuCl4 was repeated until the particles reached the desired size. Loading of cationic gold nanoparticles 20nm and 50nm citrate-stabilized Au NPs were produced with oligonucleotides following Bastus et al. In detail, a solution of 2.2 mM sodium citrate (SC) in Milli-Q water (150 mL) was heated with a heating Optimization of the NP : RNA ratio mantle in a 250 mL three-necked round-bottomed flask for 15 min 50 nm (at 3x1011) NP/mL) Au NPs coated with AUT were used, under vigorous stirring. A condenser was utilized to prevent the previously purified and dispersed in MES 10 mM. Single-stranded evaporation of the solvent. After boiling had commenced, 1 mL of DNA (ssDNA) with a molecular weight similar to an average HAuCl4 (25 mM) was injected. The colour of the solution changed mRNA construct was used as an oligonucleotide model. For from yellow to bluish grey and then to soft pink in 10 min. ssDNA loading, first 900 µl of 2-fold serial dilutions in MES Immediately after the synthesis of the Au seeds and in the same 10mM were prepared, ranging from 53-0.41 µg/mL. Next, 100 µl reaction vessel, the reaction was cooled until the temperature of the of NPs were rapidly added onto the ssDNA and the mixture was solution reached 90°C. Then, 1 mL of a HAuCl4 solution (25 mM) gently homogenized. Thus, the final relative ssDNA : NP ratios was injected. After 30 min, the reaction was finished. This process was ranged from [39-5000]. Samples were incubated for 24 h at 4°C repeated twice. After that, the sample was diluted by extracting 55 mL under stirring. Next day, samples were characterized by UV-Vis of sample and adding 53 mL of Mili-Q water and 2 mL of 60 mM spectroscopy, DLS and Z-Pot before and after purification. For sodium citrate. This solution was then used as a seed solution, and the purification, NPs were precipitated by centrifugation, supernatant process was repeated again until the particles reached the desired size. was discarded and pellets were resuspended in MES 10 mM to the initial volume. Functionalization of nanoparticles Loading kinetics 50 nm Au NPs coated with AUT were loaded with ssDNA. Functionalization of Gold Nanoparticles with AUT. First Briefly, 900µl of NPs dispersed in MES 10 mM were added onto 100 parameter explored for a stable functionalization of Au NPs was µl of ssDNA to a final ratio DNA : NP=300. Samples were kept at 4 ° the concentration of AUT. For this, 20nm Au NPs were C under stirring. At each time point, 1 mL of sample was taken for concentrated 10-fold relative to the synthesis concentration by characterization. For purification, NPs were precipitated by centrifugation (conditions were set according the Stokes law for centrifugation, supernatant was stored for ssDNA quantification each particle size). Next, AUT solutions with concentrations and pellets were resuspended in MES 10 mM to the initial volume. ranging between 50-400 µM were prepared in HCl 10 mM Conjugates were analyzed by UV-Vis spectroscopy, DLS and (pH<3). NPs (10% to final volume) were rapidly added into the additionally Z-Potential was measured after purification. The AUT solution under vigorous stirring. After 1h, samples were quantification of the ssDNA loaded on the NPs was extrapolated characterized by UV-Vis. Note that at pH values above 3, NPs from the measurement of the supernatants at 24h by Nanodrop aggregate and precipitate upon dispersion in the AUT solution. The (Nanodrop 2000 Spectrophotometer, ThermoFisher). Frontiers in Immunology 10 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 Stability of nanovectors and 7 days cell viability was measured. For this, 10 µL of PrestoBlue To study the stability of nanovectors, 50 nm Au NPs coated was added to each well, plates were incubated for 2h and with AUT and PEI, alone or loaded with ssDNA, were used. For fluorescence was measured (lex531nm, lem572nm) by Varioskan this, NP solution was diluted 1:10 in the media of study and LUX (Thermo Fisher Scientific). All experiments were carried out in incubated for 24h at 4 °C. Different biologically relevant media triplicate, and data was treated and calculated with were tested: Optimem (pH 7.4) and Phosphate Buffer (PB) 10mM OriginLab software. (pH 7.4). NPs dispersed in MES 10 mM (pH 5) were used as a control. NP stability was studied by UV-Vis and DLS. After 24 h Annexin V/propidium iodide samples were characterized. Au NPs were precipitated by To determinate the cell viability, HEK293 cells were stained centrifugation, the pellets were redispersed in the media of study with Pacific Blue- Annexin V/propidium iodide (PI) in accord with and Z-Potential was measured. the manufacturer’s recommendations to determine cell viability. Briefly, HEK293 cells were collected by cell detachment using accutase and washed with PBS. After centrifugation cells were In vitro experiments resuspended in 100 µL of Annexin binding buffer (10 mM HEPES, 140 mM NaCl and 2.5 mM CaCl2). 5 µL of Annexin V Cells culture and PI (1 mg/mL) were added and incubated at room temperature HEK293 cell culture was maintained in culture in 75 cm2 tissue for 15 minutes. After the incubation period, additional 400 µL of the culture flask using DMEM with heat- inactivated foetal bovine binding buffer was added. Acquisition was configured to stop after serum (FBS) at 10% at 37 °C and humidified 5% CO2. recording 10,000 events within the HEK293 cell population. TEM imaging of cultured cells exposed to NPs HEK293 cells were seeded on a 10 cm petri dish at 100.000 cell/ Transfection efficiency 2 cm and incubated overnight. 20 nm Au NPs coated with PEI were added dropwise onto cell cultures and gently homogenized. At 24 h Transfection of mRNA with nanovectors cells were fixed with 2.5% glutaraldehyde in 0.1 M PB. Next, To evaluate the transfection capacity of gold nanoparticles samples were embedded in paraffin following a standard protocol. coated with PEI and AUT, HEK- 293 cells were cultured in For observation, paraffin-embedded samples were sectioned using a DMEM with FBS 10% in 24-well plate at 50.000 cells/mL. The ultra-microtome and transferred carbon-coated copper TEM grid. transfection was performed with 60-70% confluence and final mRNA concentration of 1000 ng. After 2h the incubation at 37°C Proton sponge efficiency of cationic the DMEM medium was removed and replaced for 900 µL of gold nanoparticles Optimem medium. Next, specific colloidal ratios [mRNA : NP] First, the pH of a cationic Au NPs solution was adjusted to 7.3 were added (100 µL) for each nanoparticle size for 5 nm Au NPs with NaOH. Then, pH was monitored continuously as a known [5:1], 20 nm [50:1] and 50 nm [300:1]. The next day 100µL of FBS volume of HCl (10 mM) was added dropwise on the Au NPs were added to each well and left for 48h and 7 days after solution under stirring, until pH 4 was reached. A solution of Mili Q transfection process. The transfection and cell viability water was used as a control. The proton sponge efficiency of the percentages were evaluated by confocal microscopy and cationic NPs was calculated based on the HCl volume added, flow cytometry. normalized to Au surface (nm2). Chloroquine effect To further inside to the proton sponge mechanism the Cytotoxicity transfected HEK-cells were treated with chloroquine at 20 mM for 4 hours before nanovector transfection, performed as described Cytotoxicity assessment of nanovectors above. At 24h cells were visualized by Wide-Field fluorescence The action of Au NPs in the viability of the HEK-293 cells was microscopy and GFP signal intensity quantified. Statistical analysis evaluated by PrestoBlue and Annexin V/Propidium Iodide assay, was performed by the 2-way ANOVA test, using the GraphPad according to the manufacturer’s recommendations. Prism software. For significance, p>0.005 was considered. Prestoblue Flow cytometry 5 HEK293 cells were seeded to 1x10 cell/mL in 96-well plate The percentage of Green Fluorescence Protein (GFP) during 24 hours before to Au NPs exposition. Serial dilutions of expression after transfection was analyzed with BD LSRFortessa™ nanoparticles were added at final concentration ranging from Cell Analyzer. Forward and side-scatter areas (FSC-A, SSC-A) in a 3.3x1013 - 1.2x1011 for Au 5nm NPs/mL, 2.7x1012) - 1.0x1010 for linear scale were used to gate HEK293 population, and GFP Au 20 nm NPs/mL and 3x1011 - 1.1x109 for Au 50nm NPs/mL. To expression was detected by excitation through 480-500nm. To assay the ratio oligonucleotide:NPs cytotoxicity the PEI/AUT determinate the cell viability, HEK293 cells were stained with nanoparticles were loaded with oligonucleotide. After 24 h, 48 h Pacific Blue- Annexin V/propidium iodide (PI) in accord with the Frontiers in Immunology 11 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 manufacturer’s recommendations. Briefly, HEK293 cells were Funding collected by cell detachment using accutase and washed with PBS. After centrifugation cells were resuspended in 100 µL of Annexin We acknowledge financial support from the Spanish Ministerio binding buffer (10mM HEPES, 140mM NaCl and 2.5mM CaCl2). 5 de Ciencia, Innovació n y Universidades (MCIU) (RTI2018-099965- µL of Annexin V and PI (1mg/mL) were added and incubated at B-I00, AEI/FEDER,UE) proyectos de I+D+i de programació n room temperature for 15 minutes. After the incubation period, conjunta internacional MCIN/AEI (CONCORD, PCI2019- additional 400 µL of the binding buffer was added. Acquisition was 103436) cofunded by the European Union and Generalitat de configured to stop after recording 10,000 events within the HEK293 Catalunya (2017-SGR-1431). ICN2 is supported by the Severo cell population. Ochoa program from Spanish MINECO (SEV-2017-0706) and is funded by the CERCA Programme/Generalitat de Catalunya. Wide-field fluorescence microscopy Au nanoparticles transfection efficiency in HEK293 cells was calculated by GFP expression analyzed by Wide-Field Conflict of interest microscopy. To this end, HEK293 cells were visualized in Thunder Wide-Field Fluorescence Microscope (Leica). For GFP The authors declare that the research was conducted in the imaging, a 475 nm LED was used for excitation while the absence of any commercial or financial relationships that could be emission channel was set to 506-532 nm. For deconvolution of construed as a potential conflict of interest. each image, we utilized the algorithm Small Volume Computational Clearing (SVCC). Publisher’s note Data availability statement All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, The raw data supporting the conclusions of this article will be or those of the publisher, the editors and the reviewers. Any product made available by the authors, without undue reservation. that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Author contributions Supplementary material MB, GD, PB, RK, ME, MJ, NB, VP conceived of the presented idea MG, ME, VS, AA-P, AB, AC, and FF designed and performed The Supplementary Material for this article can be found online at: the experiments. All authors contributed to the article and approved https://www.frontiersin.org/articles/10.3389/fimmu.2023.1128582/ the submitted version. full#supplementary-material References 1. Marshall E. Gene therapy death prompts review of adenovirus vector. Science colloidal gold nanoparticles. Biochim Biophys Acta (2007) 1770(5):747–52. doi: (1999) 286(5448):2244–5. doi: 10.1126/science.286.5448.2244 10.1016/j.bbagen.2007.01.012 2. Phillips AJ. The challenge of gene therapy and DNA delivery. J Pharm Pharmacol 9. Niidome T, Nakashima K, Takahashi H, Niidome Y. Preparation of primary (2001) 53(9):1169–74. doi: 10.1211/0022357011776603 amine-modified gold nanoparticles and their transfection ability into cultivated cells. 3. Gemeinhart RA, Luo D, Saltzman WM. Cellular fate of a modular DNA delivery Chem Commun (Camb) (2004) 17):1978–9. doi: 10.1039/b406189f system mediated by silica nanoparticles. Biotechnol Prog (2005) 21(2):532–7. doi: 10. Ghosh PS, Kim CK, Han G, Forbes NS, Rotello VM. Efficient gene delivery 10.1021/bp049648w vectors by tuning the surface charge density of amino acid-functionalized gold 4. Kneuer C, Sameti M, Bakowsky U, Schiestel T, Schirra H, Schmidt H, et al. A nanoparticles. ACS Nano (2008) 2(11):2213–8. doi: 10.1021/nn800507t nonviral DNA delivery system based on surface modified silica-nanoparticles can 11. Li P, Li D, Zhang L, Li G, Wang E. Cationic lipid bilayer coated gold efficiently transfect cells. vitro. Bioconjugate Chem (2000) 11(6):926–32. doi: 10.1021/ nanoparticles-mediated transfection of mammalian cells. Biomaterials (2008) 29 bc0000637 (26):3617–24. doi: 10.1016/j.biomaterials.2008.05.020 5. Xiang JJ, Tang JQ, Zhu SG, Nie XM, Lu HB, Shen SR, et al. IONP-PLL: a novel 12. Pujals S, Bastú s NG, Pereiro E, Ló pez-Iglesias C, Puntes VF, Kogan MJ, et al. non-viral vector for efficient gene delivery. J Gene Med (2003) 5(9):803–17. doi: Shuttling gold nanoparticles into tumoral cells with an amphipathic proline-rich 10.1002/jgm.419 peptide. ChemBioChem (2009) 10(6):1025–31. doi: 10.1002/cbic.200800843 6. Ojea-Jimenez I, Comenge J, Garcia-Fernandez L, Megson ZA, Casals E, Puntes 13. Thomas M, Klibanov AM. Conjugation to gold nanoparticles enhances VF. Engineered inorganic nanoparticles for drug delivery applications. Curr Drug polyethylenimine’s transfer of plasmid DNA into mammalian cells. Proc Natl Acad Metab (2013) 14(5):518–30. doi: 10.2174/13892002113149990008 Sci (2003) 100(16):9138–43. doi: 10.1073/pnas.1233634100 7. Shukla R, Bansal V, Chaudhary M, Basu A, Bhonde RR, Sastry M. 14. Cebrian V, Martin-Saavedra F, Yague C, Arruebo M, Santamaria J, Vilaboa N. Biocompatibility of gold nanoparticles and their endocytotic fate inside the cellular Size-dependent transfection efficiency of PEI-coated gold nanoparticles. Acta Biomater compartment: a microscopic overview. Langmuir (2005) 21(23):10644–54. doi: (2011) 7(10):3645–55. doi: 10.1016/j.actbio.2011.06.018 10.1021/la0513712 15. Ernst LM, Casals E, Italiani P, Boraschi D, Puntes V. The interactions between 8. Noh SM, Kim WK, Kim SJ, Kim JM, Baek KH, Oh YK. Enhanced cellular delivery nanoparticles and the innate immune system from a nanotechnologist perspective. and transfection efficiency of plasmid DNA using positively charged biocompatible Nanomaterials (2021) 11(11):20. doi: 10.3390/nano11112991 Frontiers in Immunology 12 frontiersin.org Gustà et al. 10.3389/fimmu.2023.1128582 16. Hill HD, Millstone JE, Banholzer MJ, Mirkin CA. The role radius of curvature 40. Sellers H, Ulman A, Shnidman Y, Eilers JE. Structure and binding of plays in thiolated oligonucleotide loading on gold nanoparticles. ACS Nano (2009) 3 alkanethiolates on gold and silver surfaces: implications for self-assembled (2):418–24. doi: 10.1021/nn800726e monolayers. J Am Chem Soc (2002) 115(21):9389–401. doi: 10.1021/ja00074a004 17. Sandhu KK, McIntosh CM, Simard JM, Smith SW, Rotello VM. Gold 41. Di Felice R, Selloni A. Adsorption modes of cysteine on Au(111): thiolate, nanoparticle-mediated transfection of mammalian cells. Bioconjug Chem (2002) 13 amino-thiolate, disulfide. J Chem Phys (2004) 120(10):4906–14. doi: 10.1063/1.1645789 (1):3–6. doi: 10.1021/bc015545c 42. Bastus NG, Piella J, Puntes V. Quantifying the sensitivity of multipolar (Dipolar, 18. Gigante A, Li M, Junghänel S, Hirschhäuser C, Knauer S, Schmuck C. Non-viral quadrupolar, and octapolar) surface plasmon resonances in silver nanoparticles: the transfection vectors: are hybrid materials the way forward? Medchemcomm (2019) 10 effect of size, composition, and surface coating. Langmuir (2016) 32(1):290–300. doi: (10):1692–718. doi: 10.1039/C9MD00275H 10.1021/acs.langmuir.5b03859 19. Conner SD, Schmid SL. Regulated portals of entry into the cell. Nature (2003) 43. Schulz F, Vossmeyer T, Bastus NG, Weller H. Effect of the spacer structure on 422(6927):37–44. doi: 10.1038/nature01451 the stability of gold nanoparticles functionalized with monodentate thiolated poly 20. Zhang S, Li J, Lykotrafitis G, Bao G, Suresh S. Size-dependent endocytosis of (ethylene glycol) ligands. Langmuir (2013) 29(31):9897–908. doi: 10.1021/la401956c nanoparticles. Adv Mater (2009) 21(4):419–24. doi: 10.1002/adma.200801393 44. Hill RJ. Hydrodynamics and electrokinetics of spherical liposomes with coatings 21. Meng Z, O’Keeffe-Ahern J, Lyu J, Pierucci L, Zhou D, Wang W. A new of terminally anchored poly(ethylene glycol): numerically exact electrokinetics with developing class of gene delivery: messenger RNA-based therapeutics. Biomaterials self-consistent mean-field polymer. Phys review. E Statistical nonlinear soft matter Phys Sci (2017) 5(12):2381–92. doi: 10.1039/C7BM00712D (2004) 70(5 Pt 1):051406. doi: 10.1103/PhysRevE.70.051406 22. Grabbe S, Haas H, Diken M, Kranz LM, Langguth P, Sahin U. Translating 45. Vitali M, Casals E, Canals F, Colome N, Puntes V. Simple spectroscopic nanoparticulate-personalized cancer vaccines into clinical applications: case study with determination of the hard protein corona composition in AuNPs: albumin at 75. RNA-lipoplexes for the treatment of melanoma. Nanomedicine (Lond) (2016) 11 Nanoscale (2020) 12(29):15832–44. doi: 10.1039/D0NR02379E (20):2723–34. doi: 10.2217/nnm-2016-0275 46. Piella J, Bastus NG, Puntes V. Size-dependent protein-nanoparticle interactions 23. Foroozandeh P, Aziz AA. Insight into cellular uptake and intracellular in citrate-stabilized gold nanoparticles: the emergence of the protein corona. Bioconjug trafficking of nanoparticles. Nanoscale Res Lett (2018) 13(1):339. doi: 10.1186/ Chem (2017) 28(1):88–97. doi: 10.1021/acs.bioconjchem.6b00575 s11671-018-2728-6 47. Sherbet GV, Lakshmi MS, Cajone F. Isoelectric characteristics and the secondary 24. Boussif O, Lezoualc’h F, Zanta MA, Mergny MD, Scherman D, Demeneix B, et al. A structure of some nucleic acids. Biophysics structure Mech (1983) 10(3):121–8. doi: versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: 10.1007/BF00537554 polyethylenimine. Proc Natl Acad Sci (1995) 92(16):7297–301. doi: 10.1073/pnas.92.16.7297 48. Meng Z, O’Keeffe-Ahern J, Lyu J, Pierucci L, Zhou D, Wang W. A new 25. Namuduri M, Brentjens RJ. Medical management of side effects related to CAR developing class of gene delivery: messenger RNA-based therapeutics. Biomater Sci T cell therapy in hematologic malignancies. Expert Rev Hematol (2016) 9(6):511–3. doi: (2017) 5(12):2381–92. doi: 10.1039/C7BM00712D 10.1080/17474086.2016.1183479 49. Vales G, Suhonen S, Siivola KM, Savolainen KM, Catalan J, Norppa H. 26. Jackson HJ, Rafiq S, Brentjens RJ. Driving CAR T-cells forward. Nat Rev Clin Genotoxicity and Cytotoxicity of Gold Nanoparticles In Vitro: Role of Surface Oncol (2016) 13(6):370–83. doi: 10.1038/nrclinonc.2016.36 Functionalization and Particle Size. Nanomaterials (2020) 10(2):271. doi: 10.3390/ nano10020271 27. Wang T, Larcher L, Ma L, Veedu R. Systematic screening of commonly used commercial transfection reagents towards efficient transfection of single-stranded 50. Sani A, Cao C, Cui D. Toxicity of gold nanoparticles (AuNPs): a review. Biochem oligonucleotides. Molecules (2018) 23(10):2564. doi: 10.3390/molecules23102564 biophysics Rep (2021) 26:100991. doi: 10.1016/j.bbrep.2021.100991 28. Kulkarni JA, Myhre JL, Chen S, Tam YYC, Danescu A, Richman JM, et al. 51. Jawaid P, Rehman MU, Zhao QL, Misawa M, Ishikawa K, Hori M, et al. Small Design of lipid nanoparticles for in vitro and in vivo delivery of plasmid DNA. size gold nanoparticles enhance apoptosis-induced by cold atmospheric plasma via Nanomedicine-Nanotechnology Biol Med (2017) 13(4):1377–87. doi: 10.1016/ depletion of intracellular GSH and modification of oxidative stress. . Cell Death j.nano.2016.12.014 Discovery (2020) 6(1):83. doi: 10.1038/s41420-020-00314-x 29. Khare P, Dave KM, Kamte YS, Manoharan MA, O’Donnell LA, Manickam DS. 52. Saha K, Rahimi M, Yazdani M, Kim ST, Moyano DF, Hou S, et al. Regulation of Development of lipidoid nanoparticles for siRNA delivery to neural cells. AAPS J macrophage recognition through the interplay of nanoparticle surface functionality (2022) 24(1). doi: 10.1208/s12248-021-00653-2 and protein corona. ACS Nano (2016) 10(4):4421–30. doi: 10.1021/acsnano.6b00053 30. Kumar Y, Kuche K, Swami R, Katiyar SS, Chaudhari D, Katare PB, et al. 53. Wanling L, Jenny KWL. Endosomal escape pathways for non-viral nucleic acid Exploring the potential of novel pH sensitive lipoplexes for tumor targeted gene delivery systems. In: Brian C, editor. Molecular regulation of endocytosis. IntechOpen: delivery with reduced toxicity. Int J Pharm (2020) 573:12. doi: 10.1016/ Rijeka (2012). p. p Ch. 17. j.ijpharm.2019.118889 54. Behr J-P. The proton sponge: a trick to enter cells the viruses did not exploit. 31. Comenge J, Sotelo C, Romero F, Gallego O, Barnadas A, Parada TG, et al. Chimia (1997) 51(1-2):34. doi: 10.2533/chimia.1997.34 Detoxifying antitumoral drugs via nanoconjugation: the case of gold nanoparticles and 55. Smith SA, Selby LI, Johnston APR, Such GK. The endosomal escape of cisplatin. PloS One (2012) 7(10):e47562. doi: 10.1371/journal.pone.0047562 nanoparticles: toward more efficient cellular delivery. Bioconjug Chem (2019) 30 32. Shan Y, Ma S, Nie L, Shang X, Hao X, Tang Z, et al. Size-dependent endocytosis (2):263–72. doi: 10.1021/acs.bioconjchem.8b00732 of single gold nanoparticles. Chem Commun (Camb) (2011) 47(28):8091–3. doi: 56. Cox A, Andreozzi P, Dal Magro R, Fiordaliso F, Corbelli A, Talamini L, et al. 10.1039/c1cc11453k Evolution of nanoparticle protein corona across the blood–brain barrier. ACS Nano 33. Lesniak A, Salvati A, Santos-Martinez MJ, Radomski MW, Dawson KA, Aberg (2018) 12(7):7292–300. doi: 10.1021/acsnano.8b03500 C. Nanoparticle adhesion to the cell membrane and its effect on nanoparticle uptake 57. Gibbs-Flournoy EA, Bromberg PA, Hofer TP, Samet JM, Zucker RM. Darkfield- efficiency. J Am Chem Soc (2013) 135(4):1438–44. doi: 10.1021/ja309812z confocal microscopy detection of nanoscale particle internalization by human lung 34. Cheng X, Tian X, Wu A, Li J, Tian J, Chong Y, et al. Protein corona influences cells. Part Fibre Toxicol (2011) 8(1):2. doi: 10.1186/1743-8977-8-2 cellular uptake of gold nanoparticles by phagocytic and nonphagocytic cells in a size- 58. Wang F, Chen B, Yan B, Yin Y, Hu L, Liang Y, et al. Scattered light imaging dependent manner. ACS Appl Mater Interfaces (2015) 7(37):20568–75. doi: 10.1021/ enables real-time monitoring of label-free nanoparticles and fluorescent biomolecules acsami.5b04290 in live cells. J Am Chem Soc (2019) 141(36):14043–7. doi: 10.1021/jacs.9b05894 35. Liu X, Huang N, Li H, Jin Q, Ji J. Surface and size effects on cell interaction of 59. Garcia-Fernandez L, Garcia-Pardo J, Tort O, Prior I, Brust M, Casals E, et al. gold nanoparticles with both phagocytic and nonphagocytic cells. Langmuir (2013) 29 Conserved effects and altered trafficking of cetuximab antibodies conjugated to gold (29):9138–48. doi: 10.1021/la401556k nanoparticles with precise control of their number and orientation. Nanoscale (2017) 9 36. Casals E, Pfaller T, Duschl A, Oostingh GJ, Puntes V. Time evolution of the (18):6111–21. doi: 10.1039/C7NR00947J nanoparticle protein corona. ACS Nano (2010) 4(7):3623–32. doi: 10.1021/nn901372t 60. Yu JN, Ma SF, Miao DQ, Tan XW, Liu XY, Lu JH, et al. Effects of cell cycle status 37. Casals E, Pfaller T, Duschl A, Oostingh GJ, Puntes VF. Hardening of the on the efficiency of liposome-mediated gene transfection in mouse fetal fibroblasts. J nanoparticle-protein corona in metal (Au, Ag) and oxide (Fe3O4, CoO, and CeO2) Reprod Dev (2006) 52(3):373–82. doi: 10.1262/jrd.17097 nanoparticles. Small (2011) 7(24):3479–86. doi: 10.1002/smll.201101511 61. Fredericksen BL, Wei BL, Yao J, Luo T, Garcia JV. Inhibition of endosomal/ 38. Bastus NG, Comenge J, Puntes V. Kinetically controlled seeded growth synthesis lysosomal degradation increases the infectivity of human immunodeficiency virus. J of citrate-stabilized gold nanoparticles of up to 200 nm: size focusing versus ostwald Virol (2002) 76(22):11440–6. doi: 10.1128/JVI.76.22.11440-11446.2002 ripening. Langmuir (2011) 27(17):11098–105. doi: 10.1021/la201938u 62. Heath N, Osteikoetxea X, de Oliveria TM, Lazaro-Ibanez E, Shatnyeva O, 39. Piella J, Bastú s NG, Puntes V. Size-controlled synthesis of Sub-10-nanometer Schindler C, et al. Endosomal escape enhancing compounds facilitate functional citrate-stabilized gold nanoparticles and related optical properties. Chem Mater (2016) delivery of extracellular vesicle cargo. Nanomedicine (Lond) (2019) 14(21):2799–814. 28(4):1066–75. doi: 10.1021/acs.chemmater.5b04406 doi: 10.2217/nnm-2019-0061 Frontiers in Immunology 13 frontiersin.org