pmc Adv Mater Adv Mater 379 blackwellopen 9885358 ADMA Advanced Materials (Deerfield Beach, Fla.) 0935-9648 1521-4095 pmc-is-collection-domain yes pmc-collection-title Wiley Open Access Collection PMC13261376 PMC13261376.1 13261376 13261376 41814945 10.1002/adma.202510532 ADMA72768 1 Research Article Research Article A Cryoprotectant‐Compatible Nanoporous Platform for Stable and Scalable Delivery of Biopharmaceuticals Lee Sian 1 2 Kim Seongchan 1 3 Chong Jooyeun 4 Kim Min‐gyeong 5 6 Lim Kayeong 5 6 Lee Nan‐Ee 1 Koo Jahyun 2 Kang Jiheong 4 jiheongkang@snu.ac.kr Lee Hyojin https://orcid.org/0000-0001-6367-9485 1 6 7 hyojinlee@kist.re.kr 1 Biomaterial Research Center Biomedical Research Division Korea Institute of Science and Technology (KIST) Seoul Republic of Korea 2 School of Biomedical Engineering Korea University Seoul Republic of Korea 3 College of Pharmacy and Research Institute of Pharmaceutical Sciences Gyeongsang National University Jinju Gyeongsangnam‐do Republic of Korea 4 Department of Chemistry Seoul National University Seoul Republic of Korea 5 Biomedical Research Division Korea Institute of Science and Technology (KIST) Seoul Republic of Korea 6 Division of Bio‐Medical Science &Technology KIST School University of Science and Technology Seoul Republic of Korea 7 SKKU‐KIST Department of Integrative Biotechnology College of Biotechnology and Bioengineering Sungkyunkwan University Suwon Gyeonggi Republic of Korea 12 3 2026 12 6 2026 38 33 515466 10.1002/adma.v38.33 e10532 17 12 2025 04 6 2025 03 3 2026 12 06 2026 13 06 2026 17 06 2026 © 2026 The Author(s). Advanced Materials published by Wiley‐VCH GmbH https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the terms of the http://creativecommons.org/licenses/by-nc-nd/4.0/ License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made. ABSTRACT CRISPR‐Cas9 ribonucleoproteins (RNPs) represent a promising class of biopharmaceuticals for treating genetic and complex diseases. However, their clinical translation is limited by instability during storage and delivery. Lyophilization offers a potential solution, though conventional approaches often compromise structural integrity and bioactivity under non‐cryogenic conditions. Here, we have developed a nanostructured delivery platform, designated Nano Banker & Blowball (NB 2 ), which features a blowball‐like architecture and tunable nanoscale pores. These pores are designed to protect RNPs and enable controlled release. The freeze‐dried formulation (FNB 2 ) integrates optimized cryoprotectants and a surface‐engineered nanoparticle design, preserving morphology and function without excessive excipients. FNB 2 exhibits rapid rehydration and retains ∼70% of gene editing activity post‐lyophilization, enabling robust functional gene editing in vitro and in vivo. It also maintains long‐term stability and supports efficient cellular uptake, enabling administration via multiple routes. FNB 2 represents a scalable and robust platform for genetic therapeutics, vaccines, and biologics, particularly well‐suited for resource‐limited and emergency medical applications. A cryoprotectant‐compatible nanoporous platform enables ambient‐stable and scalable delivery of gene editing therapeutics. By combining hierarchical pore architecture with optimized lyophilization chemistry, the system preserves Cas9‐RNP activity post‐freeze‐drying. This work introduces a versatile strategy for cold chain‐independent storage and minimally invasive administration of biopharmaceuticals, advancing clinical and field applications of next‐generation genetic medicines. ambient‐stable biopharmaceuticals freezed‐dried formulation gene editing nano banker & blowball scalable therapeutic platform National Research Foundation of Korea 10.13039/501100003725 RS‐2022‐NR072331 RS‐2024‐00403376 Korea Health Industry Development Institute, Ministry of Health & Welfare, Republic of Korea RS‐2023‐KH135060 RS‐2024‐00438476 KIST Institutional Program pmc-status-qastatus 0 pmc-status-live yes pmc-status-embargo no pmc-status-released yes pmc-prop-open-access yes pmc-prop-olf no pmc-prop-manuscript no pmc-prop-legally-suppressed no pmc-prop-has-pdf yes pmc-prop-has-supplement yes pmc-prop-pdf-only no pmc-prop-suppress-copyright no pmc-prop-is-real-version no pmc-prop-is-scanned-article no pmc-prop-preprint no pmc-prop-in-epmc yes pmc-license-ref CC BY-NC-ND source-schema-version-number 2.0 cover-date 12 June 2026 details-of-publishers-convertor Converter:WILEY_ML3GV2_TO_JATSPMC version:6.8.0 mode:remove_FC converted:12.06.2026 1 Introduction The emergence of advanced biopharmaceuticals, particularly gene editing technologies such as CRISPR‐Cas9 ribonucleoproteins (RNPs), has redefined therapeutic strategies for genetic diseases, cancers, and other complex conditions [ 1 , 2 , 3 , 4 ]. However, the clinical translation of these fragile macromolecules remains hindered by their intrinsic instability during storage and delivery [ 5 ]. The global pandemic of SARS‐CoV2 has further highlighted the fragility of existing biotherapeutic infrastructure, revealing the global dependence on cold chain logistics and the limitations of current delivery modalities [ 6 , 7 , 8 ]. The temperature sensitivity of mRNA vaccines, for instance, has underscored the urgent need for scalable, thermostable platforms capable of preserving functional bioactivity in ambient or field operational settings, an especially critical demand in low‐resource and emergency response environments [ 7 , 9 , 10 ]. As the therapeutic pipeline expands to include diverse DNA, RNA, and protein‐based agents, technologies that enable long‐term stabilization without compromising efficacy are increasingly essential [ 6 , 11 , 12 , 13 , 14 , 15 , 16 ]. Lyophilization, freeze‐drying, is a widely used technique for enhancing the biological stability of aqueous formulations by converting them into solid, dry formats. Standard protocols employ cryoprotectants and buffers to mitigate freezing‐induced damage and pH shifts. However, these approaches often offer incomplete protection, leading to loss of structural integrity, reduced rehydration efficiency, and diminished therapeutic potency [ 8 , 11 , 12 , 17 , 18 ]. Moreover, conventional lyophilized systems typically exhibit low drug loading capacity and poor scalability, thus limiting their applicability for complex or high‐dose biologics. Nanoparticle‐based carriers have emerged as a promising alternative solution to address these limitations [ 12 , 19 ]. However, lipid and polymer‐based systems frequently suffer from formulation complexity, batch‐to‐batch variability, and structural degradation during freeze‐drying and reconstitution [ 11 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ]. These limitations have been shown to result in significant therapeutic losses, exceeding 60%, reduced cellular uptake, and inconsistent performance, thereby underscoring the need for more robust delivery and stabilization platforms [ 11 ]. Here, we introduce Nano Banker & Blowball (NB 2 ). This nanostructured delivery system has been engineered to stabilize and deliver CRISPR‐Cas9 RNP with high efficiency and precision. The Cas9‐RNP complex comprises a protein for gene cleavage and a sgRNA for sequence recognition, thus serving as a suitable model to demonstrate the biopharmaceutical protection of nanoparticles through gene editing efficiency analysis [ 27 , 28 ]. The NB 2 features a blowball‐like structure with precisely controlled nano‐sized pores designed to encapsulate and protect large biomolecular complexes. This architecture offers sufficient internal volume for efficient RNP loading while maintaining nanoparticle integrity during the stresses of lyophilization. NB 2 is specifically engineered to provide a robust and protective microenvironment essential for preserving the bioactivity of gene editing complexes (Figure 1a ). The freeze‐dried formulation (FNB 2 ) incorporates an optimized combination of cryoprotectants and buffering agents to stabilize RNP during dehydration and storage. This formulation supports long‐term storage at ambient temperatures, thereby eliminating the need for cold chain infrastructure. Following rehydration, FNB 2 rapidly restores structural and functional integrity, achieving over 40% gene knockout efficiency in vitro and in vivo. A comprehensive suite of in vitro and in vivo studies has been conducted to evaluate the biocompatibility, stability, and functional efficacy of FNB 2 , thereby substantiating its effectiveness. This work establishes FNB 2 as a versatile and scalable platform for next‐generation therapeutics, offering a promising solution for expanding the accessibility and impact of gene editing therapies, vaccines, and biologics—particularly in low‐resource and emergency medical environments. FIGURE 1 Preparation and characterization of NB 2 and FNB 2 . (a) Schematic illustration of protein protection and particle stabilization during the freeze‐drying of Cas9‐RNP loaded NB 2 . (b) Representative transmission electron microscopy (TEM) and (c) Scanning electron microscopy (SEM) image of NB 2 and FNB 2 . Scale bar is 500 nm. (d) Nitrogen adsorption/desorption–derived BET surface area and mean pore size of NB 2 and FNB 2 . Data are presented as mean ± SD ( n = 3 independent nanoparticle batches). Statistical significance was evaluated using an unpaired two‐tailed Student's t‐test. (e) Loading contents of Cas9‐RNP into NB 2 and FNB 2 . Data are presented as mean ± SD ( n = 3 independent experiments). Statistical significance was determined using an unpaired two‐tailed Student's t‐test. (f) Zeta‐potential changes of Cas9‐RNP, NB 2 , Cas9‐RNP loaded NB 2 (Cas9‐RNP@NB 2 ), Cas9‐RNP loaded FNB 2 (Cas9‐RNP@FNB 2 ) and rehydration of Cas9‐RNP loaded FNB 2 (Cas9‐RNP@FNB 2 Hydration). Data are presented as mean ± SD ( n = 3 independent experiments). Statistical significance was evaluated using one‐way ANOVA followed by Tukey's post hoc test. (g) Z‐average values and polydispersity index measured using dynamic light scattering (DLS) analysis. Data are presented as mean ± SD ( n = 3 independent experiments). Statistical significance was assessed using one‐way ANOVA followed by Tukey's post hoc test. 2 Results and Discussion 2.1 Characterization of NB 2 and FNB 2 The NB 2 platform demonstrated a high loading efficiency for Cas9 ribonucleoprotein (Cas9‐RNP) complexes, attributable to its substantial surface area, considerable pore volume, and robust adsorptive properties. The pore size of NB 2 exceeded 10 nm, which is sufficient to encapsulate large biomacromolecules, including RNA, plasmid DNA, and Cas9‐RNPs. This facilitates effective encapsulation and protection from enzymatic degradation in biological environments, which is essential for in vivo stability [ 29 , 30 , 31 , 32 , 33 ]. The physicochemical stability of the FNB 2 was evaluated using parameters such as particle size, zeta potential, and loading capacity, in order to assess potential degradation. Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) revealed the characteristic spherical, blowball‐like morphology of NB 2 and FNB 2 , with an average diameter of ∼350 nm (Figure 1b,c ; Figure S1a ). Nitrogen adsorption–desorption analysis indicated a pore size of 16.86 nm and Brunauer–Emmett–Teller (BET) surface area of 525.9 m 2 /g for NB 2 , whereas FNB 2 exhibited a reduced pore size of 12.54 nm and surface area of 165.9 m 2 /g (Figure 1d ). This decrease is likely to be a consequence of partial pore occlusion by freeze‐drying excipients. Upon rehydration, these excipients dissolved, restoring the pore structure without morphological deformation (Figure S1b,c ). The Cas9‐RNP‐loaded NB 2 (Cas9‐RNP@NB 2 ) was successfully prepared by combining Cas9‐RNP and NB 2 in phosphate‐buffered saline (PBS, pH 7.4) at room temperature for 60 min. The loading capacities for Cas9‐RNP on NB 2 and FNB 2 were 26.8% and 23.9% (w/w), respectively, as determined by bicinchoninic acid (BCA) assay (Figure 1e ). These values exceed those previously reported for lipid and polymeric nanoparticles, thereby underscoring the advantage of the inorganic NB 2 framework in maintaining high payload retention post‐lyophilization [ 33 , 34 ]. In order to assess structural changes during loading and freeze‐drying, zeta potential and hydrodynamic size were monitored. The NB 2 displayed a positive surface charge (+25.4 ± 1.7 mV), enabling electrostatic interaction with negatively charged Cas9‐RNPs (−21.3 ± 2.5 mV). Upon complex formation, a decrease in zeta potential to +18.5 ± 0.9 mV was observed, thereby confirming successful binding. Following lyophilization, a further reduction to +11.6 ± 0.9 mV was observed, attributable to the presence of excipients (Figure 1f ). Furthermore, the dynamic size of NB 2 increased from 366.9 ± 1.6 nm to 452.1 ± 6.2 nm in the presence of freeze‐drying preservatives (Figure 1g ; Figure S1b ). Upon rehydration, the particle size reverted to 370.9 ± 2.4 nm, indicating that the structural integrity of the particles was maintained and that the freeze‐drying agents were effectively removed (Figure 1g ; Figure S1c ). The collective results demonstrate that NB 2 offers a structurally robust, porous architecture suitable for high‐efficiency loading, protection, and release of Cas9‐RNPs. Moreover, the freeze‐drying process successfully preserved the functionality and structural integrity of the loaded nanoparticles, ensuring their viability for therapeutic gene editing applications upon rehydration. This highlights the FNB 2 ‐based platform's potential for in vivo delivery of gene editing tools with minimal activity loss. 2.2 Optimization of Stabilization Techniques for Freeze‐Dried Formulations: FNB 2 and Enzymatic Bioactivity Freeze‐drying (lyophilization) is a foundational technique in biopharmaceuticals, frequently employed to convert liquid formulations into stable solid forms, thereby improving shelf‐life and enabling ease of transportation [ 8 , 12 ]. However, freeze‐drying poses significant challenges, as biomolecules are subjected to various stressors during the process, potentially compromising their structural integrity. Enzymes, in particular, are susceptible to denaturation due to disruption of hydrogen bonding when water molecules freeze, and polar groups are exposed, leading to loss of activity and structural stability [ 35 , 36 ]. To mitigate these risks, excipients are integral to maintaining protein stability throughout the freeze‐drying cycle. Stabilizers that are frequently employed include non‐reducing disaccharides (e.g., sucrose, trehalose), polysaccharides (e.g., dextran), and bulking agents (e.g., mannitol). The function of these excipients is to replace water‐protein hydrogen bonds and form a protective matrix around biomolecules [ 37 ]. In parallel, pH fluctuations during the freezing and drying processes pose additional risks, especially for pH‐sensitive proteins such as Cas9. Buffer systems must therefore be carefully optimized to maintain pH stability and enzymatic bioactivity post‐lyophilization [ 17 , 38 ]. The selection of suitable excipient‐buffer combinations requires consideration of molecular compatibility, concentration‐dependent effects, solution viscosity, and physicochemical interactions. Elevated viscosity, often induced by high concentrations of disaccharides or polysaccharides, can hinder heat and mass transfer during freeze‐drying, affecting drying uniformity and reconstitution efficiency. Moreover, excessive viscosity can prolong the residence time in a low‐viscosity intermediate state during freezing, which is associated with increased risk of aggregation in colloidal systems such as polyplexes and nanoparticle formulations [ 39 ]. These effects are particularly critical when employing nanoporous delivery platforms such as FNB 2 , where excipient viscosity may influence nanoparticle–matrix interactions, diffusivity, and intrapore penetration. At specific freeze‐concentration levels, high local viscosity can restrict cargo mobility even above Tg′, thereby impacting loading efficiency and uniform distribution within nanopores [ 20 ]. To determine an optimal formulation, we systematically evaluated excipients—sucrose, dextran, and mannitol—and buffers including phosphate‐buffered saline (PBS), Tris‐HCl, and 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid (HEPES). Their effects on nanoparticle stability and protein activity were then assessed post‐lyophilization (Figure 2a ). Horseradish peroxidase (HRP) served as the model protein for optimizing freeze‐drying conditions, immobilized on NB 2 . Enzyme activity was assessed post‐lyophilization using a 3,3′,5,5′‐Tetramethylbenzidine (TMB)‐based colorimetric assay (Figure 2a ; Figure S2a ). Excipients were tested at concentrations ranging from 0.005% to 0.2% (w/v). A combination of sucrose (5%), trehalose (2.5%), mannitol (5%), and dextran (2.5%) achieved the highest retention of enzymatic activity post‐freeze‐drying (Figure 2b ). Among these, the sucrose–dextran formulation preserved HRP activity at levels approximately eightfold higher than the unstabilized control (Figure 2c ). This synergistic effect is attributed to their complementary stabilization mechanisms. Sucrose facilitates vitrification, immobilizing protein molecules within an amorphous matrix to prevent protein‐protein interactions and structural destabilization during freezing. Through a water‐replacement mechanism, sucrose hydroxyl groups establish hydrogen bonds with protein polar residues, substituting for lost water–protein interactions during drying [ 40 ]. Dextran contributes through an exclusion volume effect, forming a macromolecular network that minimizes aggregation and maintains the structural integrity of both proteins and the NB 2 matrix [ 41 , 42 ]. Sequential incorporation of sucrose and dextran into HRP‐loaded FNB 2 was confirmed by energy‐dispersive X‐ray spectroscopy (EDAX) and thermogravimetric analysis (TGA), validating their presence and distribution post‐lyophilization (Figure S2b,c ). For enzymes such as HRP and Cas9, divalent cations (e.g., Mg 2 + ) present in buffers like PBS and Tris‐HCl may form inhibitory complexes that impair enzymatic activity [ 43 ]. Replacing these with HEPES, a zwitterionic buffer with low metal‐binding affinity, resulted in a 2.5–3‐fold increase in activity retention following lyophilization (Figure 2d ; Figure S2d ). Based on these findings, the final optimized freeze‐drying formulation, comprising 20 mM HEPES, 5% (w/v) sucrose, and 2.5% (w/v) dextran was adopted for all experimental evaluations. FIGURE 2 Optimization of freeze‐drying conditions and excipient combinations for enzyme activity preservation using FNB 2 . (a) Schematic illustration of the freeze‐drying optimization process, including stabilizer screening and buffer selection to enhance nanoparticle integrity and enzyme stability. (b) Screening of excipients (sucrose, trehalose, mannitol, and dextran) at varying concentrations (0.5–20 w/v%) to identify the optimal formulation for maximizing HRP activity retention post‐lyophilization. (c) Comparison of excipient combinations with the highest retention to determine the optimal formulation for stabilizing NB 2 . Data are presented as mean ± SD ( n = 3 independent experiments). Statistical significance was evaluated using one‐way ANOVA followed by Tukey's post hoc test. (d) Effect of buffer systems (PBS, Tris‐HCl, and HEPES) on enzymatic activity retention, with pH adjustments performed during the freeze‐drying process. Data are presented as mean ± SD ( n = 3 independent experiments). Statistical significance was assessed using one‐way ANOVA followed by Tukey's post hoc test. To further elucidate the hypothesized protective mechanisms by which FNB 2 enhances Cas9‐RNP preservation and delivery efficiency, we conducted spectroscopic and thermal analyses. These include: (i) physical confinement of the RNPs within nanopores that reduce conformational mobility, (ii) electrostatic interactions that buffer local charge perturbations, and (iii) formation of a vitrified matrix via excipients that mimics the hydration layer and preserves protein structure. These processes effectively suppress secondary protein structure collapse during freeze‐drying, thereby preserving biological activity upon rehydration. In the FT‐IR spectra, Cas9‐RNP@FNB 2 exhibited pronounced broadening of the OH‐stretching band around 3200–3400 cm − 1 , indicative of enhanced hydrogen‐bonding interactions (Figure S3a ). This observation suggests that the excipients facilitated the formation of an extensive hydrogen‐bonding network, thereby contributing to the structural stabilization of the protein complex during lyophilization. Differential scanning calorimetry (DSC) further corroborated this stabilization effect. The Cas9‐RNP@FNB 2 formulation with excipients exhibited a thermal denaturation profile (T m ∼ 90°C–100°C) comparable to that of native RNP, demonstrating preservation of structural integrity [ 44 ]. Moreover, the elevated glass transition temperatures (T g ) observed in the presence of excipients indicate that the protectants and the silica matrix formed a vitrified solid matrix that immobilized the RNPs and effectively enhanced their thermal stability (Figure S3b ) [ 45 ]. Circular dichroism (CD) spectroscopy further confirmed the preservation of the secondary structure of Cas9‐RNP following lyophilization. The CD spectrum of Cas9‐RNP@FNB 2 displayed characteristic α ‐helix signals at 208 and 222 nm, along with a distinct β ‐sheet signature around 217 nm, closely resembling the native Cas9‐RNP in solution. In contrast, the spectrum obtained from freeze‐dried Cas9‐RNP formulated with lipid nanoparticles (LNPs) exhibited a marked loss of these structural features, indicating significant protein denaturation and loss of conformational integrity (Figure S3c,d ). We next quantitatively assessed the cooling profiles of aqueous formulations and the shift in freezing temperature (TF) to elucidate the enhanced interfacial interactions between water molecules and the nanoparticle matrix (Figure S4 ). Water confined within the nanoscale pores of NB 2 exhibits delayed crystallization, remaining in a supercooled liquid state at temperatures lower than those required for bulk water to freeze. This nano‐confinement effect leads to a noticeable suppression of ice nucleation [ 46 , 47 ]. As shown by DSC measurements, the TF of the FNB 2 formulation was −11.8°C, which was elevated compared to NB 2 (−13.2°C), LNP (−22.3°C), and freeze‐dried LNP (FLNP, −21.4°C) (Figure S4a ). Notably, this corresponds to an increase of ∼1.4°C, 10.5°C, and 10.9°C, respectively, while bulk water froze at −25.9°C under identical conditions. These results suggest that steric hindrance imposed by the cryoprotectant matrix restricts water mobility and consequently suppresses ice crystal formation. Furthermore, NB 2 and FNB 2 subjected to repeated freeze–thaw cycles (one and two cycles) displayed consistent TF values (−13.2°C and −11.8°C, respectively), with no significant deviations observed (Figure S4b,c ). This consistency supports the structural robustness of FNB 2 during thermal stress. These results indicate that FNB 2 delays freezing through two complementary mechanisms: primary suppression via nanopore confinement and secondary stabilization through matrix formation with excipients. This dual‐mode inhibition of ice crystallization reduces freezing‐induced stress and contributes to the enhanced structural and functional preservation of loaded RNPs during freeze–thaw processing. Taken together, these findings suggest that the mesoporous structure of FNB 2 provides a stabilizing environment, likely via electrostatic interactions between the RNP cargo and the internal pore surface, which help preserve the native protein conformation during freeze‐drying. Furthermore, including cryoprotective excipients (e.g., sucrose and dextran) was essential to maintaining the structural integrity, as evidenced by the enhanced spectral preservation in FNB 2 co‐formulated with these agents. These results highlight the critical role of the FNB 2 platform, both in its nanostructure and excipient compatibility, in protecting Cas9‐RNP from lyophilization‐induced structural damage. 2.3 Assessment of FNB 2 ‐Mediated Gene Editing Efficiency via RNP Delivery in Cells The FNB 2 platform was optimized by combining NB 2 with cryoprotectants, thereby resulting in a unified system that is capable of efficient gene delivery. In contrast to lipid‐based nanoparticles (LNPs), which exhibit heterogeneous ice crystal formation and surface ice accumulation even in the presence of cryoprotectants [ 48 , 49 ], NB 2 effectively mitigates these issues. The porous structure of NB 2 facilitates the infiltration of Cas9‐RNPs by capillaries, ensuring the effective protection of the encapsulated biopharmaceuticals. The utilization of appropriate cryoprotectants during freeze‐drying processes prevents the formation of internal ice crystals and preserves the structural integrity of NB 2 , thereby mitigating these effects [ 39 , 50 , 51 ]. Consequently, NB 2 enables superior preservation of biotherapeutics compared to lipid‐based carriers. Furthermore, upon rehydration, the carrier rapidly restores its original state, immediately resuming its function as a gene delivery vector. This highlights NB 2 as a system that not only preserves biopharmaceuticals effectively but also enables efficient gene editing applications. To evaluate gene editing performance, GFP‐expressing HeLa (GFP–HeLa) cells were used as a model. Cas9‐RNP complexes targeting the GFP gene were loaded into freeze‐dried Cas9‐RNP@FNB 2 nanoparticles, and knockout efficiency was assessed (Figure 3a ). Commercial lipid‐based reagents, Lipofectamine (Lipo), and LNPs were used as commercial controls under conditions optimized for gene delivery and viability (Figure S5a,b ). The biocompatibility of FNB 2 was assessed across a concentration range of 0 to 640 µg/mL. NB 2 and FNB 2 exhibited excellent biocompatibility, with cell viabilities exceeding 90% across all concentrations. In contrast, cell viability with LNP and Lipo was below 25% across conditions, demonstrating the cytotoxicity of these systems (Figure S5c ). FIGURE 3 In vitro evaluation of GFP knockout efficiency in naïve and freeze‐dried states for various nanoparticle formulations. (a) Schematic illustration of GFP gene disruption in GFP‐HeLa cells using Cas9‐RNP@FNB 2 . (b) Confocal laser scanning microscopy (CLSM) images showing GFP expression in cells treated with Cas9‐RNP@LNP, Cas9‐RNP@Lipo, and Cas9‐RNP@NB 2 , in both their naïve and freeze‐dried (FD) states. Scale bar is 100 µm. (c) Quantification of GFP‐negative cell populations using fluorescence‐activated cell sorting (FACS), demonstrating gene knockout efficiency of Cas9‐RNP@LNP, Cas9‐RNP@Lipo, and Cas9‐RNP@NB 2 in both naïve and freeze‐dried states. (d) Gene deletion frequency analysis of GFP‐targeted Cas9‐RNP in GFP‐HeLa cells using next‐generation sequencing (NGS) ( n = 3). After confirming that the Cas9‐RNP enters the nucleus by NB 2 ‐mediated intracellular uptake to be able to exert the role of gene editing (Figure S6 ), GFP knockout efficiency was assessed by fluorescence microscopy and flow cytometry. Cas9‐RNP@FNB 2 treatment led to a substantial reduction in GFP fluorescence, indicating effective gene disruption. Notably, the freeze‐dried formulation retained comparable activity to its non‐lyophilized counterpart (Cas9‐RNP@NB 2 ) (Figure 3b ; Figure S7a ). GFP‐negative cells quantified by fluorescence‐activated cell sorting (FACS) demonstrated knockout efficiencies of 84% for Cas9‐RNP@NB 2 and 58.5% for freeze‐dried Cas9‐RNP@FNB 2 , exceeding those of untreated controls. Notably, both LNP and Lipo exhibited reduced activity after freeze‐drying, in contrast to the retained high efficiency of FNB 2 (Figure 3b,c ; Figure S7b ). In addition, western blotting confirmed the decreased GFP expression in treated cells (Figure S7c ). Genomic DNA was extracted two days post‐treatment, and targeted deep sequencing was performed. Before freeze‐drying, indel frequency rates in the GFP gene were 49.5% for NB 2 , 39.5% for Lipo, and 21.2% for LNP, while those of post‐freeze‐drying were 44.8% for FNB 2 , 0.3% for Lipo, and 14.8% for LNP (Figure 3d ). To extend validation beyond the GFP model, we targeted the clinically relevant PCSK9 locus using ABE8e, confirming therapeutic effects through base editing rather than knockout, with efficient A‐to‐G editing (47% for NB 2 , 37% for FNB 2 ) accompanied by downregulation of PCSK9 and upregulation of LDLR (Figure S8 ) [ 1 , 52 , 53 , 54 ]. These results highlight the superior stability and efficacy of FNB 2 ‐mediated delivery compared to Lipo and LNP under freeze‐drying conditions. We evaluated more formulations to see how the nanoporous carrier matrix and cryoprotectants affect gene editing during lyophilization. Cas9‐RNP alone or with NB 2 but without cryoprotectants lost activity during freeze‐drying, with indel frequencies below 10%. Only the fully integrated formulation—Cas9‐RNP@FNB 2 —kept high editing efficiency, with an indel frequency of 44.8% upon rehydration. These results show that the combination of NB 2 ’s architecture and cryoprotectants is vital for Cas9‐RNP's bioactivity under lyophilization conditions (Figure S7d ). Collectively, these data highlight FNB 2 as a biocompatible and highly stable delivery platform capable of supporting long‐term storage and transport of protein–nucleic acid therapeutics. Its retained functionality, low cytotoxicity, and scalable lyophilization compatibility position FNB 2 as a promising system for clinical gene editing applications and translational research. 2.4 Prolonged Storage of FNB 2 with Sustained Bioactivity The FNB 2 platform demonstrated exceptional long‐term stability and functional retention, thereby addressing critical limitations in the storage and scalability of gene delivery systems. Utilizing an optimized lyophilization protocol, FNB 2 was converted into a powder formulation and stored at room temperature in sealed vials along with desiccant. This configuration was utilized to evaluate the physicochemical and biological stability of the formulation over time (Figure 4a,b ). In contrast to lipid‐based vectors such as LNP and Lipo, which exhibited rapid degradation of transfection efficiency in both solution and freeze‐dried forms, FNB 2 retained over 20% gene editing efficiency and showed a marked reduction in GFP expression in GFP–HeLa cells even after 90 days of storage (Figure 4c,d ; Figure S9 ). This extended stability underscores the resilience of the FNB 2 architecture in maintaining the activity of encapsulated biomolecules. The sustained performance can be attributed to the intrinsic robustness of the inorganic nanoparticle framework and the protective effects of lyoprotectants. Disaccharides such as sucrose, known to interact with cationic silica surfaces, prevent irreversible nanoparticle aggregation during dehydration and storage, thereby preserving structural integrity and bioactivity [ 20 , 55 ]. In contrast, organic‐based formulations frequently encounter challenges in maintaining function over time due to sensitivity to temperature, moisture, and degradation pathways [ 56 ]. FNB 2 ’s durable, solid‐state composition provides a protective microenvironment for Cas9‐RNP complexes, preserving both structural and functional integrity throughout the freeze‐drying and storage processes. This feature differentiates FNB 2 from conventional systems and offers a practical, scalable solution for the stabilization of protein–nucleic acid therapeutics. FIGURE 4 Prolonged storage and sustained bioactivity of FNB 2 . (a) Schematic representation of the evaluation of FNB 2 as a therapeutic platform, emphasizing its potential for long‐term storage and transport while preserving bioactivity. (b) Stability assessment of FNB 2 stored in freeze‐dried powder form under mild conditions for 90 days. (c) Indel frequency analysis in GFP‐HeLa cells treated with FNB 2 after 90 days of storage. (d) Representative fluorescence microscopy images of GFP‐HeLa cells treated with FNB 2 stored for 90 days. Scale bar is 100 µm. (e) Quantitative comparison of indel frequencies for Cas9‐RNP stored at −80°C, −20°C, room temperature, and FNB 2 preserved at room temperature. (f) Comparison of indel frequencies between −80°C cryopreservation and FNB 2 ‐based room temperature preservation after extended storage. Data are presented as mean ± SD ( n = 3 independent experiments). The capacity of FNB 2 to retain high gene editing activity after extended storage highlights its potential for stockpiling and distribution under non‐refrigerated conditions. Such performance is particularly valuable in settings requiring rapid deployment of biopharmaceuticals, including global health emergencies and pandemic response. The stability and scalability of FNB 2 make it well‐suited for next‐generation therapeutic platforms, enabling reliable transport, storage, and delivery of gene editing agents in both clinical and field environments. For comparison with conventional cryopreservation, RNPs stored at −80°C exhibited the highest activity, whereas those at −20°C or RT showed reduced efficiencies. Notably, FNB 2 preserved comparable editing efficiency to –80°C storage even after 30 days at RT, underscoring its utility as a practical alternative to the gold standard (Figure 4e,f ). 2.5 Scalable In Vivo Gene Editing via Injectable and Microneedle‐Based Delivery of Cas9‐RNP@FNB 2 To enable clinical translation and field deployment, the FNB 2 platform was engineered to support flexible administration routes, including both microneedle (MN)‐based and injectable delivery. We first assessed the noninvasive potential of FNB 2 by incorporating freeze‐dried Cas9‐RNP@FNB 2 into dissolvable MN arrays for localized dermal gene editing. Functional performance and scalability were evaluated using a stop‐EGFP reporter mouse model, in which EGFP expression is suppressed by a premature stop codon placed upstream of a β ‐globin polyadenylation signal under the control of a CAG promoter (Figure 5a,b ; Figure S10 ). For optimal MN fabrication, Cas9‐RNP@FNB 2 was freeze‐dried into a powder and incorporated into MNs at a concentration of 100 mg/mL, enabling full dermal penetration (Notes S1 and S2 and Videos S1 and S2 ). FIGURE 5 MN‐mediated delivery of Cas9‐RNP@FNB 2 for enhanced genome editing in Stop‐EGFP‐reporter mice. (a) Schematic representation of gene editing efficiency in Stop‐EGFP reporter mice following MN application. (b) Activation of EGFP expression in the skin of reporter mice via knockout‐mediated gene editing. The injection sites and the genomic structure of the reporter gene cassette are illustrated. The gRNA target sequence and PAM motif are highlighted in green and blue, respectively, while the EGFP stop codon is indicated in red. (c) Representative in vivo IVIS images of EGFP expression in mice treated with MN alone, Cas9‐RNP‐loaded MN (Cas9‐RNP@MN), Cas9‐RNP@FNB 2 ‐loaded MN (Cas9‐RNP@FNB 2 @MN), and Cas9‐RNP‐loaded NB 2 (Cas9‐RNP@NB 2 ) ( n = 3). (d) Quantitative analysis of IVIS fluorescence signals corresponding to the mice shown in (c), expressed as average radiant efficiency within the region of interest ( n = 3 mice per group). (e) Quantification of GFP‐positive cells in skin tissues by fluorescence‐activated cell sorting (FACS) at day 15 post‐treatment. (f) Representative flow cytometry analysis of GFP expression in skin cells at day 15 post‐treatment. (g) Targeted deep sequencing analysis of edited GFP sequences in skin cells following treatment with Cas9‐RNP@NB 2 and Cas9‐RNP@FNB 2 @MN ( n = 6). Statistical significance was assessed using one‐way ANOVA followed by Tukey's post hoc test. Two MN formulations were generated: Cas9‐RNP‐loaded MN (Cas9‐RNP@MN) and FNB 2 ‐loaded MN (Cas9–RNP@FNB 2 @MN). An intradermal injection of Cas9‐RNP@NB 2 served as a positive control. Mice received MN treatments on days 1, 5, and 10, and EGFP expression was monitored by in vivo imaging through day 15. While all groups exhibited some background fluorescence at the application site, robust EGFP expression was detected throughout the dorsal skin in mice treated with Cas9‐RNP@FNB 2 @MN (Figure 5c,d ). Representative IVIS images are shown for three mice per group, and fluorescence intensities were quantified from the same animals to directly relate imaging signals to quantitative measurements. Quantification of excised skin on day 15 by flow cytometry revealed EGFP‐positive populations in 37% of cells for the Cas9–RNP@NB 2 group and 34% for the Cas9‐RNP@FNB 2 @MN group, with negligible expression observed in control groups (Figure 5e,f ). Flow cytometry and targeted deep sequencing analyses were additionally performed using a larger number of animals ( n = 6 per group) to confirm the reproducibility of the observed gene editing outcomes. Targeted deep sequencing confirmed gene editing efficiencies of 42.11% for Cas9‐RNP@FNB 2 @MN, closely matching the 57.14% achieved by the NB 2 ‐injected control (Figure 5g ; Figure S14 and Table S1 ). Expanded IVIS, flow cytometry, and sequencing results obtained from six animals are provided in Figure S14 . These data demonstrate that FNB 2 supports efficient, noninvasive delivery of genome editors via MNs, with retention of biological activity following lyophilization and rehydration. This MN‐based delivery strategy offers notable advantages in terms of patient comfort, self‐administration potential, and logistical simplicity. Combined with the scalability and ambient stability of freeze‐dried formulations, the approach is well‐suited for use in low‐resource settings and for the decentralized deployment of gene editing or immunization technologies. To expand the utility of FNB 2 , we next assessed its performance following conventional injectable delivery. The platform's ability to be stored at ambient temperature and reconstituted on demand positions it as a viable candidate for rapid clinical deployment. We evaluated gene editing efficacy in a subcutaneous tumor xenograft model using GFP‐expressing HeLa cells implanted in BALB/c nude mice (Figure 6a ). Mice bearing established tumors were treated with a single intratumoral injection of Cas9‐RNP@FNB 2 . By day 3 post‐treatment, a substantial reduction in GFP fluorescence was observed. Compared to PBS controls, LNP and freeze‐dried LNP (FLNP) formulations resulted in moderate decreases in GFP signal (17.75% and 21.1%, respectively), while Lipo and freeze‐dried Lipofectamine (FLipo) demonstrated minimal efficacy (0.05% and undetectable changes). In contrast, NB 2 and FNB 2 treatments yielded greater reductions in fluorescence intensity (20.85% and 26.47%, respectively) (Figure 6b ; Figure S15a,b ). Ex vivo fluorescence imaging was used to evaluate tissue‐level changes in GFP signal following intratumoral administration, while flow cytometry was subsequently performed as a complementary analysis to quantify GFP disruption at the cellular level. Flow cytometry confirmed these findings, revealing GFP‐negative cell increases of 46.24% in the NB 2 group and 35.9% in the FNB 2 group relative to controls (Figure 6c ; Figure S15c ). As indel formation at the GFP locus was independently validated in vitro and in microneedle‐mediated in vivo models, functional GFP disruption was used as a surrogate efficacy metric for injectable delivery. Analysis of excised tumor tissues further demonstrated marked reductions in fluorescence, with quantitative imaging showing 84.49% and 63.31% decreases in EGFP signal for NB 2 and FNB 2 groups, respectively (Figure 6d,e ; Figure S15d ). These results validate the efficacy of Cas9‐RNP@FNB 2 for in vivo gene editing via injection, with performance comparable to non‐lyophilized controls. Comparable trends were reproduced in an independent set of experiments performed with six animals per group, as shown in Figure S16 . FIGURE 6 Injectable delivery of Cas9‐RNP@FNB 2 in GFP‐HeLa tumors. (a) Schematic illustration of in vivo gene editing to GFP in GFP‐HeLa tumors mediated by Cas9‐RNP@FNB 2 . (b) Ex vivo fluorescence images of GFP‐HeLa tumors harvested from nude mice 3 days after intratumoral injection of Cas9‐RNP formulated in LNP, Lipo, or NB 2 in both naïve and freeze‐dried states ( n = 3). (c) Flow cytometric quantification of GFP‐negative tumor cells following intratumoral administration of the formulations described above. (d) Confocal laser scanning microscopy (CLSM) images of GFP‐HeLa tumor sections following the treatments described above ( n = 3). Scale bar is 50 µm. (e) Quantitative assessment of GFP fluorescence intensity in excised tumor tissues and corresponding sections. To ensure therapeutic efficacy was accompanied by favorable tolerability, we performed a systemic biocompatibility evaluation following intravenous administration of FNB 2 . No significant changes in body weight were observed over the 14‐day study period (Figure S17 ), suggesting minimal systemic toxicity. Hematological analysis showed red and white blood cell counts within normal physiological ranges, with no signs of hemolysis, anemia, or immune activation (Figure S18a and Table S2 ). Serum biochemical parameters, including hepatic enzymes (ALT, AST) and renal markers (BUN, creatinine), remained comparable to PBS controls (Figure S18b and Table S2 ). Histological examination of major organs revealed no signs of inflammation, necrosis, or fibrosis (Figure S18c ). These findings confirm the in vivo safety of FNB 2 and demonstrate that inclusion of cryoprotectants does not compromise physiological compatibility following lyophilization and rehydration. In addition to hematological and biochemical assessments, systemic evaluation following intravenous administration of FNB 2 was performed. IVIS imaging and flow cytometry demonstrated hepatic accumulation of nanoparticles [ 57 , 58 , 59 ] and efficient EGFP activation in hepatocytes. Importantly, targeted deep sequencing confirmed robust on‐target editing at the reporter locus without detectable indels at predicted off‐target sites (Figure S19 and Table S3 ), thereby supporting both the efficacy and genomic safety of systemic FNB 2 delivery. Taken together, these data establish FNB 2 as a robust and biocompatible platform for the delivery of genome editors. Both MN‐based and injectable modalities achieved potent gene editing activity in vivo, with the freeze‐dried formulations maintaining functional integrity upon rehydration. The superior performance of FNB 2 relative to lipid‐based systems, particularly following lyophilization, highlights its translational potential for gene therapies requiring ambient storage, rapid activation, and flexible administration. Importantly, the combination of long‐term stability, modular formulation, and minimal toxicity supports the deployment of FNB 2 in diverse therapeutic contexts. The platform is particularly suited for applications where cold‐chain infrastructure is limited, including global health interventions, emergency response scenarios, and point‐of‐care therapeutics. 3 Conclusions In this study, we introduce FNB 2 , a cryoprotectant‐compatible nanoporous platform for the stabilization and delivery of biopharmaceuticals, with a focus on CRISPR‐Cas9 ribonucleoproteins (RNPs). The platform overcomes the limitations of conventional delivery systems, such as lipid nanoparticles, by offering enhanced structural integrity, ambient temperature stability, and extensive biocompatibility. The hierarchical porous architecture of FNB 2 facilitates the encapsulation, protection, and controlled release of Cas9 RNPs, thereby preserving bioactivity during both storage and in vivo delivery. To support post‐lyophilization functionality, we developed an optimized freeze‐drying protocol incorporating sucrose, dextran, and a HEPES‐based buffer. This formulation maintained the physicochemical properties and gene editing activity of FNB 2 ‐loaded Cas9 RNPs following rehydration, even after extended storage at ambient temperature. These characteristics permit cold chain independent distribution and long‐term stockpiling. Furthermore, FNB 2 also supports diverse administration routes. When integrated with dissolvable MN, the platform enabled minimally invasive, localized delivery while retaining gene editing efficacy, thus advancing patient‐centric applications in resource‐limited settings. In vivo toxicological evaluation across multiple dosing regimens revealed no evidence of systemic toxicity, with hematological and biochemical parameters within physiological ranges and no histopathological abnormalities in major organs. The modularity and scalability of FNB 2 make it broadly applicable across therapeutic classes. Previous studies on lyophilized mRNA–lipid nanoparticles demonstrated improved stability compared to liquid formulations but remained strongly dependent on optimized cryoprotectants, buffers, and excipients for maintaining functionality [ 10 , 60 , 61 , 62 ]. In contrast, although FNB 2 also incorporates an optimized cryoprotectant and buffer system, its hierarchical nanoporous architecture inherently stabilizes fragile macromolecules, thereby reducing formulation dependency and broadening applicability to diverse biologics. Its compatibility with fragile macromolecules, including nucleic acids and proteins, positions it as a promising vehicle for gene editors, mRNA vaccines, biologics, and other next‐generation therapeutics. Future studies will assess its utility in disease‐specific models and evaluate process scalability for clinical translation. Additionally, pairing FNB 2 with emerging modalities such as base editors or programmable protein systems may expand its therapeutic scope. In conclusion, FNB 2 can be regarded as a versatile platform for robust, cold chain‐independent delivery of sensitive biologics, with strong potential to enhance access to advanced therapies in both conventional and austere healthcare environments. 4 Experimental Section 4.1 Materials Cetyl trimethylammonium bromide (CTAB) was purchased from Acros Organics (MA, USA). 3‐Aminopropyltriethoxysilane (APTES), toluene, 1‐butanol, and tetraethyl orthosilicate (TEOS) were obtained from Sigma‐Aldrich (MO, USA). Phosphate‐buffered saline (PBS, 10X), Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), and Penicillin/Streptomycin (P/S, 100X) were acquired from WELGENE (Korea). C12‐200 was sourced from ChemOne (Korea). 1,2‐Distearoyl‐sn‐glycero‐3‐phosphocholine (DSPC, 18:0 PC) and DMG‐PEG (MW 2000) (DMG‐PEG2000) were procured from Avanti Polar Lipids (Alabaster, AL, USA). Cholesterol was obtained from Sigma‐Aldrich. Primers were purchased from Bioneer (Korea) or Cosmogenetech (Korea). Taq and pfu DNA polymerase for PCR were sourced from Biofact (Korea). Cas9 protein and sgRNAs were obtained from Bioneer (Korea). 4.2 Preparation of the NB 2 CTAB were dissolved in 60 mL of deionized water under stirring at 70°C for 1 h. Subsequently, toluene:1‐butanol: TEOS (20:1:1 v/v) were quickly added to the reaction mixture and then further stirred at 75°C for 12 h to promote the formation of the NB 2 . After aging for 12 h at room temperature (RT), products were isolated by centrifugation, followed by washing with deionized water and ethanol to remove any residual reactants. For surface functionalization with a positive charge, the dried NB 2 were suspended in a solution of APTES and subjected to reflux at 120°C overnight. 4.3 Optimization and Characterization of Cas9‐RNP@NB 2 and FNB 2 To prepare Cas9‐RNP‐loaded NB 2 , Cas9 protein and sgRNA were incubated for 5 min at RT to form the Cas9‐RNP complex. The resulting complex was combined with NB 2 dispersed in PBS and allowed to interact for 1 h at RT. After conjugation, the mixture was centrifuged at 8000 rpm for 10 min, and the Cas9‐RNP@NB 2 conjugates were collected as a pellet. The pellet was then resuspended in a lyophilization matrix containing HEPES buffer, sucrose, and dextran (0.5–20 w/v%) to ensure proper dispersion through pipetting. The samples were frozen at −80°C, annealed at −20°C for 30–60 min, and subsequently re‐frozen at −80°C for 30 min. Primary drying was performed at −60°C under 8 mTorr vacuum for 48 h. For secondary drying, the vials were transferred to a room‐temperature vacuum desiccator and held overnight (∼12 h) to further reduce residual moisture. The structural and physicochemical properties of NB 2 and FNB 2 were assessed using nitrogen adsorption/desorption isotherms on a surface area analyzer (Nova 600, Anton Paar). Morphological features were visualized through Transmission Electron Microscopy (TEM, Tecnai) and Field Emission Scanning Electron Microscopy (FE‐SEM, Inspect F50). To confirm the successful loading of Cas9‐RNP onto NB 2 , changes in zeta potential and hydrodynamic size were measured using a Zetasizer Nano ZS90 (Malvern, UK). To evaluate the release profile of sgRNA and Cas9 protein, the Cas9‐RNP@NB 2 were incubated in PBS at 37°C for 4 and 16 h, and the supernatant was collected by centrifugation. The released sgRNA was quantified using UV–Vis NanoDrop spectrophotometry (ThermoFisher Scientific), while the released Cas9 protein was quantified using the BCA protein assay (ThermoFisher Scientific). The loading capacity (LC) of Cas9‐RNP were calculated using the following equations: LC = (weight of the Cas9 protein or sgRNA in the NB 2 )/(total weight of the Cas9‐RNP@NB 2 ) × 100. 4.4 Mechanism Study To elucidate the stabilization mechanisms conferred by the FNB 2 formulation, we conducted a series of spectroscopic and thermal analyses. Fourier‐transform infrared (FT‐IR) spectroscopy (Thermo Scientific Nicolet iS50) was used to evaluate hydrogen bonding interactions and matrix formation between cryoprotectants and the protein cargo. Spectra were collected in the range of 4000–400 cm − 1 with a resolution of 4 cm − 1 and 32 scans per sample. Differential scanning calorimetry (DSC; TA Instruments Q2000) was used to determine freezing temperature (TF) and glass transition temperature (Tg), comparing FNB 2 , NB 2 , and LNP formulations. Approximately 10 mg of each freeze‐dried formulation was hermetically sealed in aluminum pans and cooled from 25°C to −40°C at a rate of 5°C/min, held for 5 min, and then heated to 120°C at the same rate. CD spectroscopy was performed on a Jasco J‐1500 spectropolarimeter using a 1 mm quartz cuvette. Spectra were acquired from 190 to 260 nm at 25°C, with a scanning speed of 50 nm/min, data pitch of 1.0 nm, and response time of 2 sec. Samples were prepared in deionized water at a protein concentration of ∼0.2 mg/mL. All samples were analyzed before and after freeze‐drying, and results were used to correlate nanoporous confinement, electrostatic interaction, and excipient matrix formation with protein structural integrity. 4.5 Enzyme Activity Using FNB 2 To optimize the freeze‐drying process for NB 2 , various excipients, including sucrose (S0389, Sigma), trehalose (PHR1344, Sigma), D‐mannitol (M4125, Sigma), and dextran 70 (D1449, TGI), were prepared at concentrations ranging from 0.5 to 20 w/v (%) and screened for their ability to stabilize nanoparticles. To identify the most suitable buffer for maintaining pH stability during freeze‐drying, buffering systems including phosphate‐buffered saline (PBS, P2101‐100, GenDEPOT), Tris‐HCl (648315, Calbiochem), and HEPES (CA011‐010, GenDEPOT) were systematically evaluated. Horseradish peroxidase (HRP) was utilized as a model protein to evaluate enzymatic stability throughout the freeze‐drying process. Enzymatic activity was quantitatively determined using a colorimetric assay with TMB (3,3′,5,5′‐tetramethylbenzidine) as the substrate, and the optical density was measured at 450 nm using a microplate spectrophotometer (GM3000, Promega). 4.6 In Vitro Transfection GFP‐HeLa (human cervical cancer) were cultured in DMEM (LM001‐05) supplemented with 10% FBS (F0600‐050, WelGENE), and 1% penicillin/streptomycin (LS202‐02) at 5% CO 2 and 37°C. Cells used in this study were determined to be negative for mycoplasma using Hoechst 33342 solution (62249l, Thermo Fisher Scientific). GFP‐HeLa (40 000 cells per well) were seeded on 24‐well cell culture plates (30024, SPL). After 24 h, the cells were treated with Cas9‐RNP@(F)Lipo or Cas9‐RNP@(F)LNP, or Cas9‐RNP@(F)NB 2 in serum‐free medium for 4 h. After this, the medium was replaced with a complete medium. Lipofectamine 2000 (11668019, Invitrogen) was used as a transfection control. 4.7 Cytotoxicity Test GFP‐HeLa cells (10,000 cells per well) were seeded on 96‐well cell culture plates (30096, SPL) and cultured in DMEM containing 10% FBS and 1% penicillin/streptomycin(p/s) at 5% CO 2 and 37°C for 24 h. Then, LNP, Lipo, and NB 2 were added to each well in various concentrations ranging from 0.1–640 µg/mL in serum‐free medium for 4 h. The total volume in each well was 100 µL. After incubation for 4 h, the medium was replaced with DMEM with 10% FBS and 1% p/s and incubated at 37°C for 24 h in a humidified, 5% CO 2 atmosphere. After this, the medium was replaced with 100 µL serum‐free medium with 10% MTS assay reagent (G3580, Promega) followed by 1 h incubation in conventional cell culture conditions. The absorbance was then recorded at 490 nm using a 96‐well plate reader (GM3000, Promega). The experiment was performed in triplicate. 4.8 Western Blotting GFP‐HeLa cells (4 × 10 4 cells per well) were seeded in 24‐well cell culture plates. After 24 h, the cells were treated with Cas9‐RNP@FNB 2 in serum‐free medium for 4 h. Following 48 h of incubation, the cells were washed with PBS and trypsinized. The collected cells were then washed again with PBS and harvested by centrifugation. Cell pellets were lysed using RIPA buffer (R4200‐010, GenDEPOT) supplemented with a 100× protease inhibitor cocktail and incubated on ice for 1 h. The lysates were centrifuged at 12 000 rpm for 15 min at 4°C. The resulting supernatants were transferred to 1.5 mL microcentrifuge tubes and stored at −80°C. Protein concentration was determined using the Pierce BCA Protein Assay Kit (#23228, Thermo Fisher Scientific) according to the manufacturer's instructions. Equal amounts of protein (20 µg) were mixed with 5× loading buffer (S2002, Biosesang, Korea) and loaded onto a 10% SDS‐polyacrylamide gel for electrophoresis. Protein separation was performed in Tris‐glycine SDS buffer at 80 V for 150 min. Proteins were transferred onto a 0.45 µm nitrocellulose membrane (#1620145, Bio‐Rad) using 1× Tris‐glycine transfer buffer (HT2028, GenDEPOT, Korea) at 50 V for 120 min at 4°C. The membrane was blocked with 5% BSA to prevent nonspecific binding and incubated overnight at 4°C with the following primary antibodies: anti‐GAPDH (ab9485, Abcam; 1:2000) and rabbit anti‐GFP (LF‐Pa0043, Santa Cruz; 1:2000). The membrane was then incubated for 1 h at room temperature with a secondary antibody, goat anti‐rabbit IgG H&L (HRP) (ab205718, Abcam; 1:5000). Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system (W3652‐020, GenDEPOT, LLC). 4.9 Targeted Deep Sequencing For the analysis of mutations at endogenous on‐target and predicted off‐target (OT) sites, genomic DNA was extracted from cells or tissues using the AccuPrep Genomic DNA Extraction Kit (Bioneer, Korea) according to the manufacturer's instructions. Candidate OT sites were identified using the web‐based tool [], and those with the highest cutting frequency determination (CFD) scores were selected for validation. The extracted DNA was subjected to three rounds of PCR amplification using 2× Pfu‐PCR Master Mix (PD302‐50 h, BioFACT) to generate sequencing libraries with Truseq DNA‐RNA CD Indexes (Illumina). NGS libraries were purified using the DNA Clean & Concentrator kit (Zymo Research, D4004) and sequenced on iSeq100 or MiniSeq sequencer (Illumina). Indel frequencies were analyzed with Cas‐Analyzer, CRISPR RGEN Tools [ http://www.rgenome.net/cas‐analyzer ]. NGS sequencing data are available on NCBI (SUB15242599). 4.10 Fabrication of the MN Patch and Time‐Lapse Fluorescence Microscopy of Release Behavior 1‐Vinyl‐2‐pyrrolidinone, acrylic acid (anhydrous), and α‐ketoglutaric acid were purchased from Sigma‐Aldrich and used without further purification. Customized polydimethylsiloxane MN molds were used. The MN mold contains 100 cavities, each with a base diameter of 500 µm and a height of 650 µm. MN patch synthesis was done in the glove box to avoid moisture. The MN patch contacting nanoparticle was synthesized as follows. First, a monomer solution was prepared by mixing vinyl pyrrolidone and acrylic acid in a 5:1 volume ratio. Next, α‐ketoglutaric acid, a biocompatible photoinitiator, was added at a concentration of 3 wt.% to the monomer solution. The nanoparticle (NB 2 ) was dispersed in the monomer solution at a concentration of 100 mg/mL using a magnetic stirrer. Then, the MN cavities in the mold were filled with a solution containing NB 2 , and any excess solution outside the cavities was carefully removed. Once the cavities were filled, the MN mold was then filled with a monomer solution containing only the monomer and photoinitiator, free of nanoparticles. Then, it was radiated with the UV light (BLX365, Vilber Lourmat) for 90 min, taken out of the glove box, and placed in the vacuum oven for 48 h at 30°C. After fabrication, MNs were stored under a vacuum at room temperature. Time‐lapse fluorescence microscopy imaging of both FITC and Cy5 samples was performed using a laser scanning confocal microscope (C2 plus / Ti2‐E Microscope, Nikon) and NIS‐Elements software. For the case involving nine release points, a 4X objective lens (Plan Apo λ 4X / NA: 0.20, Nikon) was used, whereas a 10X objective lens (Plan Apo λ 10X / NA: 0.45, Nikon) was employed for single‐point release measurements. FITC signals were excited using a 488 nm laser, and emission was collected in the 490–535 nm range. For Cy5 signals, excitation was achieved using a 640 nm laser, with emission detected in the 650–700 nm range. Laser power and gain settings were optimized individually for each fluorophore. The time‐lapse interval was set to 100 ms. Image acquisition continued until the fluorescent signal decreased to a level comparable to the background, indicating complete release. All video recordings and corresponding color‐coded map images were processed for enhanced visibility using ImageJ software. 4.11 Animal Studies All experimental procedures were approved and reviewed by the Korea Institute of Science and Technology (Approval No. KIST‐IACUC‐2024‐012‐1). Transgenic mice C57BL6/J (CAG‐STOP‐EGFP) and BALB/c nude mice (male, 6–8 weeks old, 18–30 g) were utilized in this study. Mice were anesthetized with 2% isoflurane with 1 L min −1 oxygen. 4.12 Non‐Invasive In Vivo Application Prior to treatment, the fur at the application site was carefully shaved to ensure optimal MN insertion. To evaluate the therapeutic efficacy of Cas9‐RNP@FNB 2 ‐loaded MN, the MNs were applied three times on days 1, 5, and 10. On day 15, the treated tissue at the MN insertion site was harvested for further analysis. On day 15, GFP expression was assessed using an imaging system (IVIS Spectrum, PerkinElmer), and cells were harvested from skin tissue for FACS analysis. GFP‐positive cells were then sorted and subjected to targeted deep sequencing. This schedule allowed for a comprehensive assessment of the gene editing effects of the Cas9‐RNP@FNB 2 formulation delivered via MNs. 4.13 Injectable Delivery of Cas9‐RNP@FNB 2 in Xenograft Mouse Model A GFP‐HeLa tumor xenograft model was established by subcutaneous injection of GFP‐HeLa (5 × 10 6 ) cells suspended in a 1:1 mixture of Matrigel and PBS into BALB/c nude mice. To evaluate the in vivo delivery efficiency of Cas9‐RNP formulations, Cas9‐RNP@(F)LNP, Cas9‐RNP@(F)Lipo, and Cas9‐RNP@(F)NB 2 were intratumorally injected ( n = 3 ) (formulation containing Cas9 protein: sgRNA = 2:1, NB 2 : 20 mg/kg). Ten days after tumor formation, the formulations were intratumorally injected. Three days post‐administration, the mice were imaged using an in vivo imaging system (IVIS Spectrum, PerkinElmer), and tumors were subsequently excised for further imaging. Following excision, tumor tissues were collected for analysis of GFP‐negative cells using flow cytometry or sectioned for GFP fluorescence imaging by confocal laser scanning microscopy (CLSM). 4.14 Statistical Analysis All images were obtained from at least three independent experiments with similar results. Also, all data were shown as mean corrected values ± s.d. of at least three independent experiments unless otherwise stated. Data were analyzed using Prism software version 8 (GraphPad, CA, USA). Floating bars represent the minimum and maximum values, with a line at the mean. Individual data points are overlaid. Graphical data are presented as means ± s.d. Statistical significance of differences between two groups was determined using Student's unpaired t ‐test or one‐way analysis of variance. p ‐values were considered statistically significant following stars: * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001; ns, not significant. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting File 1 : adma72768‐sup‐0001‐SuppMat.docx. Supporting File 2 : adma72768‐sup‐0002‐VideoS1.mp4. Supporting File 3 : adma72768‐sup‐0003‐VideoS2.mp4.