Gelatin methacryloyl granular scaffolds for localized mRNA delivery

Messenger RNA (mRNA) therapy is the intracellular delivery of mRNA to produce desired therapeutic proteins. Developing strategies for local mRNA delivery is still required where direct intra‐articular injections are inappropriate for targeting a specific tissue. The mRNA delivery efficiency depends on protecting nucleic acids against nuclease‐mediated degradation and safe site‐specific intracellular delivery. Herein, novel mRNA‐releasing matrices based on RGD‐moiety‐rich gelatin methacryloyl (GelMA) microporous annealed particle (MAP) scaffolds are reported. GelMA concentration in aerogel‐based microgels (µgels) produced through a microfluidic process, MAP stiffnesses, and microporosity are crucial parameters for cell adhesion, spreading, and proliferation. After being loaded with mRNA complexes, MAP scaffolds composed of 10% GelMA µgels display excellent cell viability with increasing cell infiltration, adhesion, proliferation, and gene transfer. The intracellular delivery is achieved by the sustained release of mRNA complexes from MAP scaffolds and cell adhesion on mRNA‐releasing scaffolds. These findings highlight that hybrid systems can achieve efficient protein expression by delivering mRNA complexes, making them promising mRNA‐releasing biomaterials for tissue engineering.


INTRODUCTION
Gene therapy is performed by intracellularly introducing genetic materials to supplement the protein deficiency caused by mutations or to enhance a biological activity by overexpression of a gene. [1,2]Genes can be transferred via vectors directly to the patient (known as in vivo gene therapy), or patient cells can be genetically modified in vitro and then returned to the patient (known as ex vivo gene therapy). [3]n patients, gene delivery may be applied to treat the causes of a broad range of diseases instead of only relieving the symptoms. [4]Furthermore, gene therapy has been used for preventing or treating diseases and their complications. [5,6]ene delivery is a promising strategy for illness in which small molecules or protein-based drugs cannot achieve the required outcomes. [7]ne of the most significant challenges of gene delivery relies on developing new strategies capable of improving low levels of gene transfer in the nucleus or cytoplasm and its expression.Typically, viral and non-viral vectors facilitate the intracellular trafficking of exogenous nucleic acids through complex tissues and cellular barriers. [2]Among the successful viral vectors, adenovirus, adeno-associated virus, lentivirus, and retrovirus are the most commonly used for vector construction.However, there are still several concerns due to undesirable effects, including response from the host's immune system against the vector, dose-related toxicity, short-lived or insufficient transgene expression, and mutagenicity and tumorigenicity. [8,9]n this sense, the development of specific lipid-based non-viral vectors has gained attention due to their availability, better physical stability, less immunogenicity, more straightforward genetic engineering and production, and consequently better cost effectiveness than viral vectors. [10]The delivery of non-viral vectors loaded with messenger RNA (mRNA) is an alternative method to produce therapeutic proteins with advantages over plasmid delivery, resulting in rapid protein expression.Unlike conventional DNA delivery, synthetic mRNA is converted into protein after escaping from the endosome and entering the cytoplasm without the need to enter the cell nucleus or interact with DNA. [11]As a result, there is no incorporation into the host genome, significantly reducing the probability of insertional mutations.
However, one of the significant challenges of mRNA therapy associated with non-viral vectors is the development of strategies that can improve low levels of gene transfer, which occurs due to inefficient transport into the cytoplasm and rapid degradation of vectors and genetic material. [12]ecently, mRNA lipid-based nanocarriers have been successfully developed by Moderna and Pfizer/BioNTech to induce systemic immunity against SARS-CoV-2. [10]Both mRNA vaccines work by introducing an mRNA that corresponds to a small piece of a specific protein, spike protein, found on the SARS-CoV-2's outer membrane.Another major challenge of mRNA therapy lies in developing approaches for local gene delivery to avoid undesired systemic side effects.In this perspective, the development of mRNA-releasing matrices has been explored to achieve controlled and localized gene expression in vivo in tissue engineering, [13] vaccines, [14] and cancer immunotherapy. [15,16]arious hydrogel-based biomaterials have been investigated for the controlled and long-term release of different payloads while supporting tissue growth.The high water content of the hydrogels gives them excellent biocompatibility, providing physical resemblance to tissues. [17]20][21] In tissue engineering, traditional "bulk" hydrogels, produced via conventional techniques, have some limitations over specific applications, particularly when an injection or smaller sizes are required.As an alternative to traditional bulk hydrogels, small-scale gels have stood out with several unique properties that make them attractive for biomedical applications. [22]In this sense, dropletbased microfluidic has been used to generate microgels (µgels) for tissue engineering, [23,24] loading and delivery of bioactive molecules, nanoparticles, [25][26][27][28] viral vectors, [29,30] and non-viral vectors. [31,32]Droplet microfluidics has also been applied to produce µgels to build granular hydrogels, also known as microporous annealed particle (MAP) hydrogels. [24,33]AP scaffolds have emerged as a revolutionary tool in tissue engineering, offering a versatile and promising platform for regenerative medicine.These scaffolds, composed of a 3D network of hydrophilic polymers, possess a high water content and biocompatibility, resembling the natural extracellular matrix of tissues.Their unique porous structure provides an ideal environment for cells to adhere, proliferate, and differentiate, mimicking the physiological conditions necessary for tissue growth. [34]MAP scaffolds are highly tunable, allowing researchers to adjust their mechanical properties, porosity, and degradation rates to match specific tissue requirements. [34,35]or instance, Griffin et al. described accelerated wound healing by injectable MAP scaffolds. [24]During the formation of the MAP scaffold in laboratory conditions, cells introduced into the process proliferated and created intricate 3D networks within 48 h.Additionally, when tested in vivo, the injectable MAP scaffold promoted cell migration, leading to fast skin tissue regeneration and tissue structure formation within 5 days.Liu et al. explored immune cell characteristics within restricted and unrestricted biomaterials, utilizing spherical µgels ranging from 40 to 130 µm. [36]The findings revealed that MAP scaffolds facilitated wound healing with an IgG1-biased Th2 response.Particularly, MAP scaffolds created with large µgels stimulated a well-balanced pro-regenerative macrophage reaction, leading to improved wound healing marked by the regeneration of mature collagen and decreased inflammation levels.
In contrast, Hsu et al. reported that using MAP scaffolds incorporating a nerve growth factor gradient integrated into the interconnected pores of the nerve conduit enhances cell migration. [37]This process leads to substantial bridging effects on peripheral nerve defects, resulting in axon outgrowth of up to 4.7 mm and a twofold increase in axon fiber density within just 4 days in vivo.Li et al. created tissue-like structures using cell-laden MAP scaffolds to achieve long-term maintenance and chondrogenesis of human bone marrow-derived mesenchymal stem cells. [38]A 4-arm poly(ethylene glycol)-N-hydroxysuccinimide (NHS) crosslinker-induced covalent bonding between the MAP scaffold and the surrounding tissue mimic.The resulting MAP scaffold promoted the upregulation of chondrogenic markers in gene expression and glycosaminoglycan levels.The regenerated tissue within the µgels exhibited characteristic hyaline-like cartilage features.It revealed a favorable distribution and higher content of type II collagen in the MAP scaffolds compared to bulk hydrogel and pellet cultures.
In previous studies, MAP scaffolds have shown promising results for the delivery of pDNA-loaded non-viral vectors, [39,40] proteins, [41,42] and drug-loaded nanoparticles [43] while keeping cell elongation and migration, which are typically inhibited in traditional hydrogels due to the absence of pore interconnectivity. [34]One strategy to sustainably release mRNA and produce continuously therapeutic proteins in situ is the development of matrices loaded with mRNA.Even though mRNA-releasing matrices are emerging technologies, their application is still scarce in the literature, requiring further study to be applied clinically. [18,21]erein, we reported, for the first time, a microporous gelatin methacryloyl (GelMA) scaffold for the transgene expression of mRNA.GelMA MAP scaffolds have already been shown to be suitable for cell adhesion, migration, and proliferation. [33,44]Yet, there have not been any biological assays conducted on MAP scaffolds produced from GelMA aerogel-based µgels. [45]Thus, taking advantage of GelMA's excellent biocompatibility and easy handling, we developed a biodegradable mRNA-releasing GelMA MAP scaffold to understand how cells behave in a 3D mRNA-releasing matrix coupled with well-designed mRNA lipid-based nanocarriers.mRNAs were complexed into commercial lipid-based nanocarriers; resultant mRNA lipoplexes were loaded onto GelMA aerogel-based µgels during rehydration.In addition, fibroblast cells were topically seeded on GelMA MAP scaffolds produced through the drying process with varying GelMA concentrations to study the effects of cell proliferation and network formation.After the MAP annealing steps, efficient mRNA loading was achieved.In vitro transfection efficiency was seen by either mRNA lipid-based carrier diffusion or after cell adhesion on MAP scaffolds.The microporous structure of scaffolds provided a local microenvironment for cell infiltration and proliferation while achieving the effective production of desired proteins.

GelMA µgels and MAP characterization
Gene-loaded matrices, which combine 3D matrices or scaffolds with nucleic acids, may allow a more prolonged delivery of transgenes and a more localized effect of its expression, avoiding side effects caused by off-targets and increasing the in situ concentration of the transgenes for an extended period. [20,21,46]However, the lack of pore interconnectivity may inhibit cell elongation and migration in traditional bulk hydrogels.Microporous scaffolds have been proposed to overcome drawbacks in bulk hydrogels, promoting cell infiltration, spreading, and proliferation. [34]Herein, we used semi-synthetic GelMA granular scaffolds for sustained release of mRNA loaded in lipid nanoaggregates.
GelMA is a biocompatible and biodegradable polymer with cell-adhesive peptides (RGD) ligands that provide extracellular matrix-cell interactions without adding cell-adhesion sites. [47]Mechanical properties of matrices/scaffolds play a crucial role in cellular behaviors, including cell adhesion, spreading, migration, and differentiation.Also, cell-matrix interactions can significantly affect cell gene expression. [48]or instance, cells cultured on higher-stiffness scaffolds may have limited mobility and space, likely inhibiting cell spreading and growth. [23]Gene-loaded porous scaffolds have often been compared to non-porous scaffolds; in the first case, typically, genes can be released gradually with scaffold degradation and cellular infiltration. [39]In addition, porous scaffolds can also eventually increase and prolong transgene expression. [49]ue to the high dependence between mechanical properties and morphological aspects of scaffolds and cell behaviors, we first tested producing GelMA µgel with different GelMA concentrations, which were used to produce GelMA MAP scaffolds with well-defined characteristics.After the microfluidic production of µgels, a lyophilization process was used as a post-treatment technique to generate aerogel-based µgels; this is a simple method to produce MAP scaffolds which only depends on the rehydration of aerogels, followed by crosslinking by UV-light exposure. [45]The high production rate of µgels was achieved using a highthroughput microfluidic system based on a step-emulsion process.One critical characteristic of granular scaffolds relies on their packing condition, which is related to the pore size distribution, individual µgel size, and interconnectivity among the µgel building blocks.Together, these features may govern the cell penetration, adhesion, and proliferation within the granular scaffold.Figure 1 describes the morphological properties (µgel size, coefficient of variation [CV], and porosity), mechanical, and rheological properties of GelMA MAP scaffolds.
One advantage of using GelMA as the main composition of µgels and MAP scaffolds is that GelMA can be readily crosslinked by using a photoinitiator and light irradiation (UV or visible light).It is possible because GelMA has many methacryloyl groups; this can also be modulated easily by changing the GelMA synthesis parameters. [47]t is worth mentioning that the photo-crosslinking process allows hydrogel production with shape fidelity and stability when incubated under physiological conditions.GelMA µgels were produced through a high-throughput microfluidic device followed by an oil removal process and MAP formation (Figure 1A).The size distributions of pre-gel solution droplets and GelMA µgels after the MAP formation were characterized using the ImageJ/FIJI particle analyzer to assess the µgels' production.µgels remain polydisperse with CV values higher than 5% with a slight decrease in µgels produced using 20% of GelMA (Figure 1Bi).Before the crosslinking steps and oil removal process, µgels presented average diameters of 47 ± 4 (5%), 49 ± 6 (10%), and 43 ± 5 µm (20%) (Figure 1Bii).The average size of GelMA µgels collected after lyophilization, resuspension, and MAP formation showed diameters equal to 69 ± 1 µm (5%), 54 ± 8 (10%), and 65 ± 1 (20%) (Figure 1Biii).
Figure 1C shows the pore size and porosity analysis of different concentrations of GelMA in MAP scaffolds (5%, 10%, and 20%).The void space was investigated using a high-molecular-weight molecule solution.Fluorescent fluorescein isothiocyanate-dextran was added to fill the space within the GelMA MAP scaffolds.Z-stack images (20 zslices spanning a total distance of 100 µm) were taken to analyze the void fraction (Figure 1Cii).Significant differences in void fraction between MAP scaffolds were observed, with values ranging from 0.15 (5% and 10%) to 0.18 (20%), which can be described as low-packing systems. [49]According to reported studies, µgels ranging from 50 µm up to 100 µm can generate void fractions around 0.2 and 0.1, which can be considered low packing and high or very high packing conditions, respectively. [50]In addition, all beaded scaffolds (5%, 10%, and 20%) had median pore diameters of ∼10 µm (Figure 1Ciii).The pore size distribution and porosity are essential since cells can easily infiltrate, depending on the scaffold properties.Cells can penetrate easier in loosely packed than tightly packed microporous scaffolds.Previous studies that corroborate our data have demonstrated that packing density is also a function of the stiffness of µgels and CV, wherein the void fraction decreases, increasing CV values. [51]nvestigating the stiffness of the MAP scaffold is also essential, as it may influence cell mobility (adhesion and migration).We conducted compressive stress versus strain tests to verify the stiffness of GelMA MAP scaffolds.Compressive moduli as a function of GelMA concentration, which range from 20 ± 3 kPa (5%) to 39 ± 9 kPa (20%) (Figure 1Di,ii).These limit values, 5% and 20%, were significantly different.On the other hand, there was no significant difference between the compressive moduli of 10% and 20%.This behavior could be seen in conventional bulk GelMA scaffolds (5% and 20%). [52]omprehensive rheological characterization is also critical in understanding the cellular mechano-sensing of the granular microenvironment.Thus, the rheological properties were obtained for all annealed MAP scaffolds after crosslinking.The storage moduli (Gʹ) of MAP scaffolds were not significantly different (Figure 1Ei).Next, the storage (Gʹ) and loss (G″) moduli were measured versus oscillatory shear strain and angular frequency.The strain sweep experiments determined the linear viscoelastic region at a constant frequency (Figure 1Eii).The linearity strain limit on the linear viscoelastic region ranged from 6% (5% and 10% GelMA) to 8% (20% GelMA).In addition, gel-like characteristics were identified in all samples since Gʹ > G″.A decrease in Gʹ at strain >15% can be easily observed for all granular scaffolds, suggesting similar brittle fracture under oscillatory shear.By applying an oscillatory shear strain of 0.1%, the Gʹ and G″ versus angular frequency curves demonstrate that the granular scaffolds behaved as elastic gels with independent Gʹ up to 10 rad/s and almost one magnitude order larger than G″ (Figure 1Eiii).
One of the main parameters affecting nucleic acid release and cell infiltration is the biomaterial degradation rate.GelMA granular scaffolds were incubated in a collagenase solution to mimic in vivo degradability.The collagenase concentration used for this assay is described in the literature as the concentration found during a wound-healing process. [53]n Day 4, it is possible to observe a significant difference between all samples (Figure 1F).The complete degradation of 5% and 10% GelMA could be seen in less than 7 days and upward of 14 days for 20% GelMA.Increased concentrations of GelMA are more resistant to collagenase due to the increased gel stiffness and reduced swelling capacity. [54]welling can affect the degradation of GelMA because it can influence the diffusion of water and degradation products into and out of the hydrogel matrix.
On the other hand, the crosslinking density of GelMA can also affect swelling and degradation.When GelMA is highly crosslinked, it becomes more resistant to swelling and degradation because the crosslinks create physical barriers that limit the diffusion of water and enzymes.These findings suggest that different GelMA concentrations on MAP scaffolds had minimal effects on morphological aspects (e.g., porosity, pore size, and CV) and rheological properties.In contrast, the degradation rate of GelMA MAP scaffolds is a function of gel stiffness, swellability, and crosslinking density.As an essential design parameter for tissue engineering and gene delivery, [55] the mechanical properties were further investigated according to cell behavior.

Effect of GelMA MAP scaffolds on in vitro cell proliferation
Cell infiltration and migration within the GelMA MAP scaffolds were determined by topically seeding the cells on crosslinked MAP scaffolds (not adding them directly into the µgel slurry).Previous studies have shown that by mixing NIH/3T3 murine fibroblasts with 20% GelMA µgels slurry, cells could proliferate over 7 days after MAP formation. [33,44]n another recent study, authors demonstrated that after topical seeding, NIH/3T3 murine fibroblast could easily migrate within 3D printed granular bioinks composed of 10% GelMA µgel building blocks, keeping cell viability over 7 days post seeding. [56]In these previous studies, other GelMA concentrations were not discussed.Herein, in vitro analyses were performed to assess the topical cell seeding, infiltration, viability, and metabolic activity (proliferation).
Furthermore, the drying process of GelMA µgels to generate the MAP scaffolds has not been tested before regarding biological assays. [45]Thus, to study the effects of cell proliferation and network formation, C3H/10T1/2 fibroblast cells were topically seeded on GelMA MAP scaffolds produced through the drying process with varying GelMA concentrations (5%, 10%, and 20%).Figure 2 shows data from infiltration, adhesion, and proliferation of fibroblast cells cultured on GelMA MAP scaffolds for 3 days.
After 24 h post seeding, almost 71% of fibroblast cells infiltrated and adhered to 10% GelMA MAP scaffolds (Figure 2Ai).In contrast, only 47% and 57% adhered to 5% and 20% scaffolds.The granular scaffolds could support cell viability for at least 1 day.However, after 3 days of culture, there was no significant increase in metabolic activity for cells seeded on a 5% GelMA MAP scaffold (Figure 2Aii).It was noted that after 2 days of culture, higher cell densities were found within the stiffer MAP scaffold conditions, 29 kPa (10%) and 39 kPa (20%).The increased cell density on MAP with higher GelMA concentrations could also be justified due to the increased crosslinking density and RGD cell-binding motif. [51]AP scaffolds with 10% GelMA present more homogeneous cell distribution than 5% and 20% GelMA (Figure 2Aiii).As previously described, MAP scaffolds with 5% GelMA could not support cell adhesion and proliferation after 2 days of culture.The MAP scaffolds with 10% and 20% GelMA, which present well-connected µgels, allowed the rapid formation of 3D cellular networks, as shown by live/dead and actin/4′,6-diamidino-2-phenylindole (DAPI) staining (Figure 2Bi).The homogeneous proliferation was also confirmed through the nucleus density distribution analysis.Cell-free regions were mainly observed in 5% of MAP scaffolds.Our results corroborate with recent studies in which authors stated that lower annealing degrees allow cells to proliferate faster than MAP scaffold with high annealing degrees, likely due to the ability of cells to prolif- erate in a less stiff scaffold, less crosslinked gels. [23]Based on these results, 10% GelMA MAP scaffolds were adopted as the optimal concentration for the following studies.To show that cells can easily infiltrate within the MAP scaffold with 10% of GelMA, a representative z-stack maximum intensity projection image was taken after 24 h of cell seeding.Fibroblast cells were stained with actin/DAPI along the z-direction (300 µm depth) after penetration among the microscale void spaces (Figure 2Bii).These reported results also align with the infiltration and metabolic activity assay results.

mRNA-Lipofectamine ™ MessengerMAX ™ (LP) optimization
As a proof-of-concept, the gold standard Lipofectamine MessengerMax was used as the transfection reagent to load mRNA before loading it into GelMA MAP scaffolds.Cationic lipid-based non-viral carriers typically present a synthetic cationic lipid in its composition, facilitating complexation with phosphate groups in nucleic acids.In this study, mRNA was complexed with the lipid-based delivery carrier to achieve better intracellular trafficking and endosomal escape.Different mRNA:LP ratios (1:0.75, 1:1.5, and 1:3; mg/µL) were tested to choose the best mRNA-LP composition.The best composition was selected by analyzing the loading efficiency, cell metabolic activity when exposed to mRNA-LP complexes, and 2D transfection efficiency (Figure 3).
The loading efficiencies were verified using the mRNA accessibility assay after mRNA-LP complexes formation.Figure 3Ai depicts that almost 95% of mRNA was complexed to mRNA-LP (1:3 w/v).In contrast, 73% and 25% values were found for 1:5 and 1:0.75, respectively.By increasing fourfold the LP volume and consequently the presence of positively charged lipids, the loading efficiency increased more than fourfold-accessibility changes due to the strong interaction between mRNA and LP.The cytotoxicity of mRNA-LPs was also investigated by measuring the metabolic activities after exposing fibroblasts to the complexes for 24 and 48 h.In Figure 3Aii, a non-significant difference in cell metabolic activity was found between the different complexes regardless of the mRNA:LP ratios tested.The highest transfection efficiency of up to 49% was found when exposing fibroblasts to mRNA-LP (1:3 w/v) (Figure 3Bi).The enhanced transfection could result from this condition's high mRNA loading efficiency.In contrast, for 1:1.5 and 1:0.75 (w/v), the transfection efficiency values were 38% and 33%, respectively, Figure 3Bii.The same condition was recently tested for the delivery of therapeutic mRNA-based biomaterials. [57]Thus, mRNA-LP at the ratio of 1:3 w/v was selected as the best condition for mRNA delivery for all the subsequent studies (Figure 3Biii).RiboGreen, an RNA staining agent, was used to quantify the amount of mRNA released from LP loaded on GelMA MAP scaffolds.We used heparin, an anionic macromolecule, to dissociate the mRNA from LP.This step is necessary since RiboGreen dye only becomes intensely fluorescent when RNAs are not complexed (free mRNA).Complexed mRNA becomes unreachable by nucleic acid-binding dyes.We investigated a series of heparin dilutions to dissociate mRNA from mRNA-LP (1:3 w/v), aiming to find the best amount to release the mRNA from LP. Accessibility was expressed as the percentage of control (free mRNA without LP) as a function of the mRNA:HEP ratio.Only the mRNA:HEP ratio over 1:10000 could disrupt the complexes and completely free the mRNA (Figure 3Aiii).Thus, to quantify mRNA release from the GelMA MAP scaffolds, the 1:10000 mRNA:HEP mass ratio was applied to mRNA-LP complexes after release from GelMA MAP scaffolds.

2.4
Development of GelMA MAP scaffolds incorporating mRNA-LP

2.4.1
Loading efficiency and in vitro kinetics release profile One alternative to promote tissue regeneration is to target gene expression directly through the delivery of genes encoding desired protein, which can stimulate pro-regenerative behaviors.Gene-loaded scaffolds can generally transfect transplanted or infiltrating host cells. [55]In this sense, some MAP characteristics, such as microporosity, leverage the replacement of conventional non-porous scaffolds, in which cell elongation and migration are inhibited.Recent studies have already shown promising results for cell spreading, proliferation, and pDNA transgene expression on MAP scaffolds. [39,51,58]A recent study from the same research group investigated how a non-viral gene delivery system, composed of linear polyethylenimine condensed with pDNA and coated with hyaluronic acid, could be loaded into injectable granular hydrogel scaffolds without any aggregation phenomena. [39]As an alternative to therapeutic protein production, the delivery of mRNA stands out as a promising approach to replace pDNA delivery.Unlike conventional DNA delivery, synthetic mRNA is rapidly converted into protein after escaping from the endosome and entering the cytoplasm without the need to enter the cell nucleus or interact with DNA. [59]n this perspective, GelMA MAP scaffolds were engineered and loaded with mRNA lipid-based nanocarriers.Ideally, loading mRNA nanocarriers into matrices can help protect the cargo from RNase degradation and prolong transgene expression. [13,15,16]In this sense, forming a hybrid system composed of mRNA complexes and µgel building blocks could be an extra barrier to protect cargo from nucleases, ensuring that it is translated into protein by the ribosome in the cytoplasm.Another advantage is that the hydrogel network can hide temporally any toxicity presented in payloads. [60]Particularly, at the same time that mRNA-LP can be released from the granular scaffolds, its microporosity can allow better cell adhesion, spreading, and proliferation.Thus, after the production of GelMA µgels through a highthroughput microfluidic device, the emulsion µgels were purified to generate aerogel-based µgels (powdered µgels).Later, two methods were investigated for mRNA-LP loading into GelMA MAP scaffolds: (i) passive loading and (ii) direct loading processes (Figure 4A).
The mRNA-LP loading efficiency was assessed using the RiboGreen probe after the passive and direct loading processes (Figure 4Ai,ii).mRNA was quantified by collecting the supernatant (non-loaded mRNA-LP complexes) by cen-trifugation and using heparin to free mRNA from LP.The efficiency values for the passive and direct loading processes were around 35% and 98% (Figure 4Bi).Our results showed that loading mRNA-LP while rehydrating the aerogel-based gels was more efficient than in the preformed MAP scaffold.The high efficiency in the direct loading process results from the interaction between attractive forces on GelMA µgels, which are lightly negatively charged, and the positively charged mRNA-LP complexes.Figure 4Bii shows the homogeneous distribution of RiboGreen-labeled mRNA-LP in an z-stack image after the direct loading process, followed by three wash steps.
The homogenous distribution of mRNA-LP within the GelMA MAP scaffold is only possible due to the electrostatic interaction between the mRNA complexes and scaffolds (Figure 4Bii,iii).In addition, Figure 4D confirms that mRNA-LP was not only present on the surface of scaffolds, but they also diffused into the scaffold matrix.
Nucleic acid-loaded cationic non-viral carriers are widely explored due to their spontaneous interaction with carboxylic (COO − ) and hydroxyl (OH − ) groups presented in negatively charged hydrogel-based materials.The zeta potential values typically found for gelatin and GelMA are −4.5 and −7.85 mV, respectively. [42]The difference between the zeta potential values is due to the methacrylamide grafting in the amino groups of gelatins.Due to the presence of a cationic lipid, LP can be easily complexed with negatively charged nucleic acids, which could overcome the electrostatic repulsion of the negatively charged cell membrane.This interaction allows higher uptake of cationic nanoparticles than neutral or negatively charged particles.In this sense, incorporating mRNA-LP within the scaffolds may protect the cargo and control the release of the mRNA-LP into the surrounding media.
The in vivo degradation of GelMA hydrogels may differ due to the presence of proteases, which depend on the applied tissue.For instance, collagenase type II is the predominant proteolytic enzyme present during the healing of normal and non-healing wounds. [53]Therefore, to simulate the in vivo environment, the release kinetic profile of mRNA-LP from the GelMA MAP scaffolds was investi-gated in DPBS-collagenase type II and only DPBS buffer (Figure 4C).The GelMA MAP scaffolds loaded with mRNA-LP using the direct loading method were immersed in the presence of collagenase type II and pure-DPBS buffer to test the release kinetic of complexes.In the presence of collagenase type II at 37 • C, mRNA-LP-loaded GelMA MAP scaffolds released about 60% of mRNA in 5 days.In contrast, in the absence of collagenase, DPBS at 37 • C, around 50% of mRNA was released in 5 days.It indicates that the mRNA-LP release mechanism over time is a combination of diffusion, degradation, and swelling of GelMA µgels.The burst release observed during the first hours can be related to the mRNA-LP complexes on the surface of the MAP scaffolds.

2.4.2
Gene transfection ability of mRNA-LP-loaded GelMA MAP scaffolds in 2D and 3D in vitro culture The transfection efficacy from mRNA-LP-loaded GelMA MAP scaffolds was assessed in 2D and 3D in vitro cultures (Figure 5).Fibroblast cells were seeded on well plates with mRNA-LP-loaded GelMA MAP scaffolds placed on cell culture inserts (2D in vitro culture) (Figure 5A).Fibroblast cells were seeded directly on mRNA-LP-loaded GelMA MAP scaffolds (3D in vitro culture) (Figure 5B).After the seeding and incubation for 2 days, eGFP transfection rates were assessed through fluorescence imaging and flow cytometry.Blank MAP scaffolds and MAP scaffolds loaded with free mRNA were used as part of the control groups.For imaging analysis, cells were also stained with DAPI.
As a comparison, 2D fibroblast cell culture was exposed for 48 h to mRNA-LP-loaded GelMA MAP scaffolds (Figure 5Ai) using a transwell system to assess the transgene expression of fibroblast due to the release of mRNA-LP from the GelMA MAP scaffolds (Figure 5Aii).Two days post-cell seeding, eGFP transfection rates were analyzed by flow cytometry (Figure 5Aiii).The results showed that 13% of the cells could express eGFP, as observed in the mRNA-LP MAP group but not in the control group (free mRNA).The difference between transfection outcomes in Figure 5iv, direct mRNA-LP and mRNA:LP-loaded GelMA MAP assisted transfection, is likely due to the absence of mRNA-LP release, which may also depend on GelMA MAP scaffolds degradation rate; it may take around a week to be fully completed (Figure 1F).Moreover, as shown in Figure 4C only 45% of mRNA-LP may be released after 48 h.Further studies may be carried out to explore longterm transfection and release of mRNA from GelMA MAP scaffolds.This result matches the described release kinetic study presented in Figure 4C and Figure 5Aii, which indicates that the GelMA MAP-based gene delivery system could release mRNA-LP complexes and deliver mRNA to the cells.
Following the gene transfection in the 2D culture of fibroblasts, gene transfection studies in 3D culture were also performed to simulate the in vivo and localized transfection.In this assay, cells were seeded on the top of mRNA-LP-loaded GelMA MAP scaffolds (Figure 5Bi).Forty-eight hours post-cell seeding, cells stained by DAPI and eGFP transfection were observed by confocal microscopy.Maximum intensity projection images showed that mRNA-LPloaded GelMA MAP scaffolds effectively transfected cells.Due to the presence of RGD peptide motifs in GelMA MAP scaffolds, fibroblasts could easily adhere and proliferate while being transfected by mRNA-LP, which indicates that gene transfection may also occur in a 3D environment.

CONCLUSION
This research demonstrates the feasibility and potential of combining GelMA and step-emulsion droplet-based microfluidics to produce µgels and granular scaffolds.mRNA has not been fully explored in tissue regeneration, mainly due to the limitations of the developed delivery systems.Herein, we developed a novel granular scaffold platform for mRNA delivery.We revealed that the hybrid system composed of lipid-based gene carriers loaded on a 3D GelMA MAP scaffold showed high mRNA-LP loading efficiency.The enhanced cell infiltration, adhesion, and proliferation in GelMA MAP scaffolds suggest it can provide a promising new platform for local gene therapies, especially in regenerative tissue approaches.The use of mRNA as a therapy is still in its early stages.Still, our results offer new insights into its potential applications, such as non-healing wounds and other medical conditions.Our findings demonstrate that GelMA MAP scaffolds can effectively deliver mRNA and gene expression can be achieved while supporting cell adhesion and proliferation, making it a promising avenue for future application in tissue regeneration.
In light of the potential applications in tissue engineering, we envision several promising future directions for our mRNA-loaded MAP material in therapeutic contexts.One of the key areas of exploration involves personalized medicine, wherein patient-specific mRNA sequences can be encapsulated within the MAP scaffold.This customization would allow for tailored therapies, addressing individual patient needs and variations in healing responses.Additionally, the adaptability of our MAP scaffold opens avenues for combination therapies.By incorporating mRNAs encoding diverse growth factors, cytokines, or even CRISPR-Cas9 components, the scaffold could be designed to address multifaceted aspects of tissue regeneration.For instance, in diabetic wound healing, our material could be further optimized to deliver mRNAs targeting specific pathways involved in angiogenesis, inflammation resolution, and collagen synthesis.Moreover, investigating the potential of this material for organ-specific regeneration, such as cardiac tissue repair or neural regeneration, presents exciting opportunities.Further studies will focus on refining the specificity and efficiency of mRNA delivery, ensuring controlled release kinetics, and enhancing the overall therapeutic outcomes in diverse clinical applications.

Microfluidic device fabrication
The parallelized step emulsification devices used to generate GelMA µgels were fabricated as previously reported by de Rutte et al. [23] Briefly, standard soft lithography was employed to fabricate polydimethylsiloxane (PDMS) devices with a 10:1 w/w ratio for PDMS/curing agent (Sylgard 184 Silicon Elastomer kit, Dow Corning, USA).PDMS devices were sealed to coverslips by oxygen (O 2 ) plasma surface activation and then treated with 3% (v/v) of trichloro (1H, 1H, 2H, 2H-perfluorooctyl) silane in Novec 7500 oil, followed by washing with Novec oil (3 M) to keep the glass surface hydrophobic.Later, the devices were maintained at 60 • C for 1 h to remove the excess solvent in the microchannels.The device was formed by wedge = 150 µm and N = 500.

Synthesis of GelMA
GelMA was synthesized according to previously described methods. [47]Briefly, 10 g of Type A porcine skin gelatin (Sigma-Aldrich G2500, USA) was dissolved in 100 mL of Dulbecco's phosphate-buffered saline (DPBS) at 50 • C, followed by adding 8 mL of methacrylic anhydride (MA) dropwise.The methacrylation reaction was kept for 1 h under magnetic stirring at 50 • C.Then, the mixture was centrifuged at 1000 × g for 2 min to remove the MA excess.DPBS buffer was added to the mixture at a 1:1 ratio and then dialyzed for 1 week against deionized water at 40 • C (12-14 kDa Mw).After dialysis, the solution was lyophilized and kept at room temperature.

GelMA µgels production
GelMA µgels were produced as previously described. [61]riefly, GelMA (5%, 10%, or 20% wt/vol) was dissolved in DPBS with 0.5% (w/v) of photoinitiator (Irgacure 2959) at 80 • C until complete dissolution.In the microfluidic process, GelMA aqueous solution was kept as the dispersed phase, and the fluorocarbon oil HFE Novec ™ 7500 oil (3 M, USA) with surfactant 0.5 wt% PicoSurf™ (Sphere fluidics, UK) as the continuous phase.Two syringe pumps fed both streams into the microdevice (Harvard Apparatus, USA).The oil and GelMA flow rates were kept at 150 and 10 µL/min.The whole microfluidic setup (syringe, tubing, and microfluidic device) was kept at 37 • C during the entire process using a heater to prevent sol-gel transition (GelMA gelation) and channel clogging.The µgels in oil were collected in a tube at 4 • C to allow the physical crosslinking of µgels.

Morphological and size analysis
Droplets and GelMA µgels were imaged using a Zeiss Inverted Microscope with Axiocam 503 mono and quantified through the ImageJ software (US National Institute of Health, USA).CV was determined using n = 100 droplets or µgels.

GelMA MAP scaffold production
Granular GelMA scaffolds were produced as described previously. [45]Two steps are required: (1) GelMA µgels drying: GelMA µgels in oil were kept at −80 • C for 24 h and then lyophilized to remove the oil.This process generates dry GelMA aerogel-based µgels, which is only possible because the Novec ™ 7500 oil has high heat conductivity, volatility, and a low freezing point; (2) Crosslinking to produce granular GelMA scaffolds: dry GelMA µgels were resuspended in DPBS (pH 7) with 0.5% (w/v) of photoinitiator at 4 • C. The µgels were concentrated and packed using a pulse centrifuge (GmCLab mini centrifuge, Gilson, France).The µgels were then transferred to a PDMS mold (diameter = 6 mm; height = 1 mm) and exposed to UV light (360-480 nm) at 100 mW cm −2 for 1 min using an Omnicure system (Excelitas Technologies, USA) to form the covalent crosslinked granular scaffold.Granular scaffold formation was also possible using 10 mW cm −2 for 2 min or 25 mW cm −2 for 1 min.Before crosslinking, the µgels were kept at 4 • C during the entire process using a cold-ice bath.

Porosity
The MAP scaffold void space was analyzed by immersing the scaffold in a fluorescein isothiocyanate-dextran solution (500 kDa-15 mM).Confocal images were acquired using a Zeiss inverted LSM 710 confocal microscope, 20 slices in a distance of 100 µm (2 per condition; n = 3).After thresholding, the overall porosity was calculated using ImageJ software and the Voxel Counter plugin.Representative maximum intensity projections of granular hydrogels using dextran can demonstrate the effect of µgel size on the packing degree and porosity.

Compression modulus
The Instron mechanical tester (Instron 5542, USA) was used for the compression analysis at a rate of 1 mm min −1 .Granular GelMA scaffolds were molded using a PDMS mold (diameter = 8 mm; height = 1 mm) and then crosslinked by UV exposure with 100 mW cm −2 for 1 min.Before the mechanical test, the granular GelMA scaffold was kept in DPBS for 1 h at room temperature.The compressive modulus (stress/strain) was evaluated by calculating the slope of the stress-strain curve in the linear region (0%-10% strain).

Rheology
Oscillatory shear rheological tests were conducted in an MCR 302 Rheometer (Anton Paar, Austria) with a parallel plate geometry (diameter-8 mm, sandblasted measuring plate, PP08/S).The oscillatory frequency sweep analysis was run at 0.1-100 rad/s under an oscillatory strain of ∼0.1% (LVE region).Viscoelastic moduli versus oscillatory strain (0.01%-100%) were measured at 1 rad/s in 20 min.The scaffolds were kept in an enclosed chamber at room temperature with natural oil to prevent dehydration.

Degradability assay
Granular scaffolds produced with 5%, 10%, and 20% (wt/vol) GelMA (diameter = 6 mm; height = 1 mm) were placed in 1.5 mL centrifuge tubes filled with 500 µL of 2 U mL −1 of collagenase type II in DPBS and incubated at 37 • C for 14 days.The enzymatic solution was changed every 2 days.At predetermined time points, the samples were centrifuged to remove the collagenase solution, then the samples were freeze-dried and weighed.The granular scaffold weight loss (%) was determined using Equation ( 1), wherein M 0 and M 1 are related to the initial and final mass.

PDMS cell culturing devices
A poly(methyl methacrylate) (PMMA) custom negative mold was used to produce PDMS (10:1 w/w ratio for PDMS/curing agent) culturing devices.The negative mold was placed in degassed PDMS and then cured at 60 • C for 4 h.All devices were sealed to coverslips by O 2 plasma surface activation.
The culture wells comprised a cylindrical culture section (diameter = 6 mm; height = 3.2 mm), enabling 50 µL of media.In addition, a cylindrical media reservoir (diameter = 7 mm; height = 6.2 mm) was added above the first culturing section, allowing a maximum of 200 µL media.Before use, 30 µL of 1% agarose was added to the bottom to prevent cell adhesion on the glass surface.

In vitro culture onto GelMA MAP scaffolds
The murine C3H/10T1/2 fibroblast cell line was obtained from the American Type Culture Collection (ATCC).The cells were cultured in Dulbecco's Modified Eagle's Medium/High Glucose (DMEM) supplemented with 10% fetal bovine serum (FBS) (Life Technologies, USA) and 1% penicillin/streptomycin (Invitrogen, USA) at 37 • C and 5% CO 2 .For the 3D topical cell seeding, 1 × 10 5 cells in 20 µL were seeded on the 5%, 10%, and 20% of GelMA MAP scaffolds (diameter = 6 mm; height = 1 mm) placed in cell culturing devices.For these in vitro assays, GelMA MAP scaffolds were placed on a cell non-adhesive surface to avoid cell adhesion.
The seeded scaffolds were kept in the incubator at 37 • C and 5% CO 2 for 30 min to enable cell adhesion.After the first incubation, a complete cell culture media was added to the seeded scaffolds and then incubated again.The media was changed every other day.Metabolic activity was analyzed on Days 1 and 3 using the PrestoBlue ™ (Invitrogen, USA) assay.
C3H/10T1/2 fibroblasts-seeded MAP scaffolds from Day 1 were also imaged using live/dead and F-actin/DAPI using a confocal microscope.Cells seeded on the scaffold were stained with the live/dead assay (Thermo Fisher Scientific, USA) and imaged.Alexa-Fluor 488-phalloidin (dilution 1:100) (Life Technologies, USA) and DAPI (dilution 1:1000) (Sigma, USA) were used for F-actin and cell nuclei staining to check cell adhesion after 24 h.All the steps were performed at room temperature while protecting the sample from the light.Cells were imaged using a Zeiss inverted LSM 710 confocal microscope.

Preparation of mRNA-LP complexes
Before loading the mRNA-LP complexes on MAP scaffolds, some intermediary mRNA-LP optimizations were performed to find the best condition for transfecting fibroblast cells on 2D culture assay.Later, the same condition was used to test the efficacy of MAP scaffolds to load and release mRNAs in 2D and 3D cultures.mRNA-LP complexation was performed according to the manufacturer's protocol.Briefly, mRNA CleanCAp ® eGFP mRNA (Trilink Biotechnologies, USA) and Lipofectamine ™ MessengerMAX ™ (LP) were diluted in diethyl pyrocarbonate-treated water (DEPC) (Thermo Fisher, USA) and then complexed in different mRNA:LP ratios (1:0.75, 1:1.5, and 1:3; w/v).After complexation, the complexes were kept for 5 min to allow the mRNA incorporation.

mRNA accessibility assay
mRNA accessibility assay to the fluorescence probe Ribo-Green was performed at different mRNA:LP ratios (1:0.75, 1:1.5, and 1:3; w/v).The assay was carried out in a 96well microplate by adding 100 µL of the working solution to 100 µL of assembled complexes.All samples were diluted to a final mRNA mass of 0.1 µg with the working solution.After 4-5 min of incubation, the absolute fluorescence was measured using a multimodal microplate reader (Thermo Scientific, USA) with excitation/emission wavelengths (480/520 nm).Fluorescence intensity profiles (accessibility) were described as the mRNA percentage (control) and plotted as a function of mRNA:LP.

Cell transfection assay
C3H/10T1/2 fibroblasts were seeded in 96-well plates at a density of 1 × 10 4 cells per well.After 24 h, cells were semiconfluent (70%−90%), and they were washed with DPBS, followed by the addition of mRNA-LP in free serum and antibiotic medium.Next, cells were incubated for 4 h at 37 • C; after incubation, the media was exchanged and replaced with a fresh one.After 48 h, the eGFP fluorescence intensity was assessed using flow cytometry.Additional samples were also stained with DAPI to acquire fluorescence images.

Metabolic activity assay
The metabolic activity was analyzed on Days 1 and 2 using the PrestoBlue ™ (Invitrogen, USA) assay.

Heparin decomplexation assay
mRNAs were dissociated from LP following the previously described method. [62]Briefly, the decomplexation assay was performed to find the best dose of heparin to release all mRNA from LP wherein mRNA:LP equal to 1 µg:3 µL.Heparin, which is an anionic polysaccharide can easily be used to dissociate mRNA from the LP (cationic charged particle).Different dilutions of heparin sodium salt from porcine intestinal mucosa (250 KU) (Sigma, USA) were used to dissociate mRNA from the complex.Heparin solution was added to the mRNA-LP complexes in predetermined weight ratios between heparin and mRNA and incubated for 10 min.Then, the dissociated mRNA was assessed using RiboGreen and compared with the control group (free mRNA).

Preparation of GelMA MAP scaffold loaded with mRNA-LP
To fabricate the mRNA-LP loaded GelMA MAP scaffolds, mRNA and LP in a 1:3 w/v ratio were complexed in 0.5% photoinitiator (Irgacure 2959) in DEPC water (Thermo Fisher, USA) and kept for 5 min to allow the mRNA incorporation.Two methods were tested for loading mRNA-LP complexes to GelMA MAP scaffolds: (i) passive loading and (ii) direct loading process.

Passive loading
GelMA aerogel-based µgels were resuspended in 0.5% photoinitiator (Irgacure 2959) in DEPC water and concentrated using a pulse centrifuge to remove the supernatant.The packed µgels were transferred to a PDMS mold (diameter = 6 mm; height = 1 mm) and exposed to UV light (100 mW cm −2 for 60 s) to form the covalent crosslinked granular scaffold.Finally, the passive loading was performed by keeping the MAP scaffold in the preformed mRNA-LP dispersion for 15 min (3 mg of µgels and 0.5 µg of mRNA per sample).
4.17.2Direct loading mRNA-LP dispersion at 4 • C was used to resuspend the GelMA aerogel-based µgels (5 min incubation to allow mRNA-LP loading into GelMA µgels).The µgels loaded with mRNA-LP were packed and crosslinked as described above (3 mg of µgels and 0.5 µg of mRNA per sample).

Loading efficiency of mRNA-LP on MAP scaffold
An indirect fluorescence analysis method measured the mRNA loading efficiency (LE) in GelMA (µgels and GelMA MAP scaffolds).Briefly, physically crosslinked GelMA µgels and GelMA MAP scaffolds loaded with mRNA-LP through both processes (direct and passive loading) were washed three times.In this step, 10 µL heparin (100 µg/µL) was added into the collected supernatants to decomplex mRNA from LP.After 10 min, the non-loaded mRNA-LP was assessed using Quant-iT RiboGreen RNA Assay Kit (Invitrogen, USA).Then, the LE was calculated according to Equation (2), in which the mass of mRNA found in GelMA µgels or GelMA MAP scaffolds is related to the mass of mRNA found on supernatants, and the initial fluorescence is associated with the mass of mRNA loaded in GelMA µgels or GelMA MAP scaffolds.The sample was assayed in triplicate (n = 3).Fluorescence was measured using a fluorescence microplate reader, Varioskan LUX multimodal system (Thermo Scientific, USA), at standard fluorescein wavelengths (excitation 480 nm, emission 520 nm) and then converted to mRNA mass amount.

=
Mass of mRNA in GelMA MAP (experimental) Theoretical initial mass of mRNA in GelMA MAP × 100 (2)

mRNA-LP distribution on MAP scaffolds
The distribution of mRNA-LP in the z-direction of GelMA MAP scaffolds was also assessed.mRNA was mixed and incubated with a fluorescent probe, RiboGreen, at a ratio of 1:4 (v/v) for 30 min at room temperature.RiboGreenlabeled mRNA was then used to prepare the lipoplexes and loaded to MAP scaffolds through the direct loading process as previously described.GelMA MAP scaffolds containing labeled mRNA-LP were washed three times and imaged using confocal microscopy to obtain z-stacks.

4.20
In vitro mRNA release study from mRNA-LP complexes At the optimal ratio of heparin/mRNA (1:10000), the amount of mRNA in solution after dissociation should be approxi-mately the same before being complexed with LP.Heparin sulfate solutions with predetermined weight ratios between heparin and LP were added to the mRNA-LP complexes and incubated for 15 min.Then, the concentration of mRNA in the solution was assessed using RiboGreen Kit according to the manufacturer's protocol at 1, 3, 6, 24, 48, and 72 h after incubation at 37 • C.

4.21
In vitro mRNA release study from GelMA MAP scaffolds mRNA-LP-loaded MAP scaffold was soaked in DPBS with or without type II collagenase (2 U/mL).At predetermined time points, a 50 µL suspension of each sample was incubated with heparin for 15 min to dissociate the mRNA.RiboGreen was added to quantify the concentration of mRNA released from the GelMA MAP scaffold.The percentage was calculated based on the amount released at a given time relative to the amount loaded.

2D cell transfection model
C3H/10T1/2 fibroblasts were seeded in 48-well plates at a density of 5 × 10 4 cells per well.After incubation for 24 h, the media was replaced with a media supplemented with 2 U mL −1 collagenase.Transwell inserts (3 µm pore size) were placed in the plate, and an mRNA-LP-loaded GelMA MAP scaffold was left on the upper membrane.After 2 days of incubation, the eGFP fluorescence intensity was checked using flow cytometry and fluorescence microscopy.

3D cell transfection model
For the 3D cultured model, 5 × 10 4 fibroblasts were topically seeded onto previously prepared mRNA-LP-loaded GelMA MAP scaffold (diameter = 6 mm; height = 1 mm) loaded with 0.5 µg mRNA-LP.After 2 days of incubation, MAP scaffolds were stained with DAPI and imaged using a Zeiss inverted LSM 710 confocal microscope to assess the cell transfection in the z-direction.

Statistical analysis
All data were presented as the mean ± standard deviation of at least n = 3. Statistical analyses were performed using oneway ANOVA and Tukey's post hoc test (a = 0.05).The results were calculated through GraphPad Prism 8 software with statistical tests, each figure and method are described with significance, based on samples replicated in three technical and independent experiments.

A C K N O W L E D G E M E N T S
The authors acknowledge funding from the National Institutes of Health (HL140951, HL137193, CA257558, and DK130566), the Coordination for the Improvement of Higher

F I G U R E 2
Effect of gelatin methacryloyl (GelMA) microporous annealed particle (MAP) scaffold matrix concentration and stiffness on in vitro cell infiltration, adhesion, and proliferation.(A) Cell viability on MAP scaffolds.(i) Cell infiltration percentage 24 h post seeding.(ii) Metabolic activity versus incubation time measured by PrestoBlue ® assay.(iii) Representative live (green) and dead (red) fluorescence images of C3H cells after 2 days of culture on GelMA MAP scaffolds (*p < 0.03, ***p < 0.0004, and ****p < 0.0001, n = 3; NS indicated non-statistically significant differences).Scale bar: 100 µm.(B) Cell adhesion and distribution on GelMA MAP scaffolds.(i) Fluorescent maximum intensity projection (300 µm z-stacks) images of C3H/10T1/2 fibroblast on GelMA MAP scaffolds 2 days post seeding.(ii) Corresponding heat map (darker red indicates a region with higher nuclei).(iii) Distribution of nucleus density in the fields of view.Cells were stained with Alexa488-phalloidin and 4′,6-diamidino-2-phenylindole to assess actin and nuclei morphologies.Scale bar: 100 µm.(iv) Fluorescent 3D projection of the 300 µm z-stacks of 10% GelMA MAP scaffold illustrating the infiltration process within the internal porous structure.An orthogonal image is shown on the right panel.All data are represented as mean ± SD.

F I G U R E 3
Evaluation of mRNA:LP ratios on messenger RNA (mRNA) loading efficiency, cell viability, and in vitro transfection efficiency.(A) (i) Relative fluorescent intensity (% of control, free mRNA) of mRNA quantified with RiboGreen.(ii) Metabolic activity versus incubation time (24 and 48 h) measured by PrestoBlue ® assay, no statistically significant differences were found between the samples.(iii) Relative fluorescent intensity (% of control, free mRNA) of mRNA quantified with RiboGreen after incubating the mRNA-LP (1:3 w/v) complex with heparin solutions.Heparin can completely free the mRNA when the mass ratio of mRNA/HEP is over 1:10000.Values indicated with **** p < 0.0001, n = 3, were significantly different compared to mRNA/HEP = 1:10000 and free mRNA.(B) 2D in vitro gene transfection.(i) Flow cytometry analysis of fibroblast cells treated with different mRNA/LP ratios (1:0.75, 1:1.5, and 1:3, w/v).(ii) 2D gene transfection rates based on the results of flow cytometry (**p < 0.004 and ****p < 0.0001, n = 3).(iii) Fluorescence microscopy images of 2D fibroblast cells treated with mRNA/LP -1:3 (w/v).Fluorescence microscopy images, eGFP expression in green, and cell nuclei were labeled with 4′,6-diamidino-2-phenylindole.Scale bar: 50 µm.All data are represented as mean ± SD.

F I G U R E 4
Development of gelatin methacryloyl (GelMA) microporous annealed particle (MAP) scaffolds loaded with mRNA-LP, loading efficiency, and delivery kinetics.(A) Schematic of direct or passive loading process performed by loading MAP GelMA scaffolds with mRNA-LP.(i) Passive and (ii) direct mRNA-LP loading efficiency on 10% GelMA MAP scaffolds.(B) (i) Loading efficiency calculated on the supernatant of loaded µgels.(ii) 3D rendering of 10% GelMA MAP scaffold loaded with RiboGreen-labeled mRNA-LP after direct loading (Depth: 160 µm).(iii) Interaction between GelMA scaffold and positively charged mRNA-LP.(C) Messenger RNA (mRNA) release profiles in PBS with or without collagenase over 72 h determined using Quant-iT RiboGreen RNA Assay Kit after heparin addition.All data are represented as mean ± SD, n = 3. (D) 3D renderization of RiboGreen-labeled mRNA-LP labeled in GelMA µgel after direct loading.Scale bar: 20 µm.

F
U N D I N G I N F O R M AT I O N National Institutes of Health (HL140951, HL137193, CA257558, and DK130566); São Paulo Research Foundation (FAPESP) (Grants # 2018/18523-3, 2021/11564-9, and 2021/07057-4); National Council for Scientific and Technological Development (CNPq) (productivity grant 304815/2022-5; Ministry of Education (RS-2023-00240729); MSIT (Ministry of Science and ICT), Korea, under the ITRC (Information Technology Research Center) support program (IITP-2023-RS-2023-00258971) supervised by the IITP (Institute for Information & Communications Technology Planning & Evaluation); Korea University Grant (K2326671)C O N F L I C T O F I N T E R E S T S TAT E M E N T DD and The Regents of the University of California-Los Angeles (UCLA) have financial interests in TempoTherapeutics, which is commercializing MAP technology.D ATA AVA I L A B I L I T Y S TAT E M E N TAll data that support the findings of this study are included in the article.O R C I DBruna Gregatti Carvalho https://orcid.org/0000-0002-4314-4368Lucimara Gaziola de la Torre https://orcid.org/0000-0002-8179-1160Ali Khademhosseini https://orcid.org/0000-0002-2692-1524Natan Roberto de Barros https://orcid.org/0000-0001-8689-4110RE F E R E N C E S Torre thanks the National Council for Scientific and Technological Development (CNPq) (productivity grant 304815/2022-5).The cell line (C3H10T1/2, Clone 8) utilized in this research was provided by Dr. Jonathan Kelber from the Developmental Oncogene Laboratory at California State University, Northridge (CSUN).Han-Jun Kim, would like to acknowledge the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education (RS-2023-00240729).This research was supported by the MSIT (Ministry of Science and ICT), Korea, under the ITRC (Information Technology Research Center) support program (IITP-2023-RS-2023-00258971) supervised by the IITP (Institute for Information & Communications Technology Planning & Evaluation).This work was also supported by a Korea University Grant (K2326671).