siRNA delivery using intelligent chitosan-capped mesoporous silica nanoparticles for overcoming multidrug resistance in malignant carcinoma cells

Although siRNA is a promising technology for cancer gene therapy, effective cytoplasmic delivery has remained a significant challenge. In this paper, a potent siRNA transfer system with active targeting moieties toward cancer cells and a high loading capacity is introduced to inhibit drug resistance. Mesoporous silica nanoparticles are of great potential for developing targeted gene delivery. Amino-modified MSNs (NH2-MSNs) were synthesized using a modified sol–gel method and characterized by FTIR, BET, TEM, SEM, X-ray diffraction, DLS, and 1H-NMR. MDR1-siRNA was loaded within NH2-MSNs, and the resulting negative surface was capped by functionalized chitosan as a protective layer. Targeting moieties such as TAT and folate were anchored to chitosan via PEG-spacers. The loading capacity of siRNA and the protective effect of chitosan for siRNA were determined by gel retardation assay. MTT assay, flow cytometry, real-time PCR, and western blot were performed to study the cytotoxicity, cellular uptake assay, targeting evaluation, and MDR1 knockdown efficiency. The synthesized NH2-MSNs had a particle size of ≈ 100 nm and pore size of ≈ 5 nm. siRNA was loaded into NH2-MSNs with a high loading capacity of 20% w/w. Chitosan coating on the surface of siRNA-NH2-MSNs significantly improved the siRNA protection against enzyme activity compared to naked siRNA-NH2-MSNs. MSNs and modified MSNs did not exhibit significant cytotoxicity at therapeutic concentrations in the EPG85.257-RDB and HeLa-RDB lines. The folate-conjugated nanoparticles showed a cellular uptake of around two times higher in folate receptor-rich HeLa-RDB than EPG85.257-RDB cells. The chitosan-coated siRNA-NH2-MSNs produced decreased MDR1 transcript and protein levels in HeLa-RDB by 0.20 and 0.48-fold, respectively. The results demonstrated that functionalized chitosan-coated siRNA-MSNs could be a promising carrier for targeted cancer therapy. Folate-targeted nanoparticles were specifically harvested by folate receptor-rich HeLa-RDB and produced a chemosensitized phenotype of the multidrug-resistant cancer cells.

www.nature.com/scientificreports/ allocated to the N-H bending vibration at ~ 1500 cm −1 (Fig. 2D). The gel retardation assay revealed that the siRNA migration entirely stopped when the weight ratio of siRNA/MSN was 1:5 in electrophoresis and a loading capacity of approximately 20% w/w (Fig. 3A). The siRNA loading in MSN-NH 2 was further confirmed by shifting the zeta potential value from positive to negative after siRNA loading (Table 1).
Targeted chitosan coats and MSN-siRNA capping. Chitosan was conjugated with the sulfhydryl group of SH-folate and SH-cysteine-TAT using a PEG spacer to improve tumor specificity and MSN uptake. The SHfolate for such reaction was obtained using 2-Mercaptoacetic acid with an average synthesis yield of 70.7 ± 5.5%. A complete reaction procedure is depicted in Supplementary Fig. S1. TLC was used to confirm the reaction between Cys-TAT and MAL-PEG(3000)-NHS; then, using Ellman's assay, the conjugation efficiency of SH-folate and Cys-TAT to the maleimide group of MAL-PEG(3000)-NHS was determined to be 86.5% and 70.2%, respectively ( Supplementary Fig. S2). The average synthesis yields for those reactions were 82.2% and 65.4%, respectively. The NHS moieties in folate-PEG(3000)-NHS or TAT-PEG(3000)-NHS were substituted with the pri-    Table 1. Zeta-potential values of NH 2 -MSN structures were completely affected by the modifications. For example, the z-potential of MSN was − 10 mV due to silicon hydroxyl groups, whereas the z-potential of MSN-NH 2 nanoparticles was + 16 mV due to amine groups (Table 1). NH 2 -MSN-siRNA coating with double folate/TAT modified chitosan was also confirmed on TEM images through rough surfaces and the loss of mesoporous structures of the nanoparticles (Fig. 4). The particle sizes of NH 2 -MSN and coated NH 2 -MSN-siRNA with double folate/TAT modified chitosan were determined to be 70 ± 12 nm and 81 ± 10 nm, respectively, using SEM images (Fig. 4). Using a DLS, the sizes of NH 2 -MSN and double folate/TAT modified chitosan-coated NH 2 -MSN-siRNA were determined to be approximately 80 and 110 nm, respectively. The larger nanoparticles determined by DLS were caused by the hydrodynamic layer surrounding the particles in the solution ( Supplementary Fig. S5). Overall data obtained from the coated and uncoated NH 2 -MSN through TEM, SEM, and DLS confirmed the successful deposition of a functionalized chitosan on the nanoparticle surface. UV-spectroscopy at 260 nm was used to monitor siRNA leakage from nanoparticles for 4 days in PBS. The formulation exhibited no evidence of siRNA leakage and exhibited high stability over 4 days at 4 °C (Fig. 3).
Chitosan protection capacity on siRNA. The protective effect of chitosan on siRNA-MSNs was examined through agarose gel electrophoresis. Enzymatic cleavage of siRNA in the physiological fluid is a critical obstacle in gene therapy. Thus, NH 2 -MSN-siRNA and NH 2 -MSN-siRNA-chitosan were treated with RNase-A to evaluate siRNA protection capability. Naked siRNA was totally digested by RNase-A incubation before gel electrophoresis (  Biocompatibility and nanoparticle cytotoxic kinetics. The MTT assay was used to determine cellular toxicity following treatment with MSNs structures. Figure 5 shows that MSNs, NH 2-MSN, and any functional coated MSNs had insignificant toxicity on EPG85.257-RDB (p > 0.05). However, some coated particles influenced the viability of HeLa-RDB cells at very high concentrations. Microscopic images of the cells after 72 h nanoparticle treatments confirmed no toxic influences, and no discrepancy was found between microscopy images of the cells with or without nanoparticles treatment ( Supplementary Fig. S10). On the other hand, daunorubicin toxicity was significantly increased in siRNA delivered cells. Increased exposure to folate receptors rendered HeLa-RDB more susceptible to daunorubicin toxicity, owing to the final NH 2 -MSN-siRNA-chitosan-PEG(3000)-folate/TAT nanoparticles' greater silencing of the MDR1 (Table 2).
Cellular uptake and MDR1 gene silencing. Flow cytometry was used to assess the internalization ability of the functionalized chitosan-coated nanoparticles as an active delivery system in HeLa-RDB (high-expression folate receptor) and a passive delivery system in EPG85.257-RDB cells (low-expression folate receptor). Based on the flow cytometry results, the fluorescent intensities of TAT/folate modified NH 2 -MSNs were more potent in   6 and Supplementary Fig. S13).
The results indicate that TAT-conjugated nanoparticles significantly increase cellular uptake of nanoparticles in both cell lines when the control sample is considered to be PEG-chitosan-NH 2 -MSN treated cells. This phenomenon shows a positive role for the TAT in the interaction with the cell membrane and cell entry. PEGylation of chitosan resulted in decreased cellular uptake and transfection efficiency for various reasons, including inhibition of cellular uptake (repulsion from PEG) and inefficient endosomal escape. Targeting moieties (TAT or folate) were covalently bonded to the free ends of PEG to improve cellular uptake by allowing them to recognize the expressed receptors over the cell membrane, specifically. Cells' uptake of PEGylated nanoparticles is significantly increased due to a specific ligand-receptor interaction.  www.nature.com/scientificreports/ siRNA delivery through mesoporous silica nanoparticles led to a significant reduction of MDR1 transcripts and proteins after 48 h, where TAT/folate modified chitosan was the strongest structure, and HeLa-RDB was the most influenced line (p < 0.001; Fig. 7, Supplementary Table S1, and Supplementary Fig. S14). Real-time PCR analyses demonstrated that siMDR1-loaded NH 2 -MSNs-chitosan-PEG-TAT significantly reduces MDR1 transcription in both cell lines due to TAT's high cell membrane penetration and proton sponge properties. Gene silencing via siMDR1-loaded NH 2 -MSNs-chitosan-PEG-TAT/folate construct was great on HeLa-RDB due to the corresponding folate receptor on this cell and active transportation. In the western analyses, decreases in protein expression were observed in all samples, but the significance of these reductions varied. Hela-RDB treated cells with NH 2 -MSN-siRNA-CS-PEG-folate/TAT showed the greatest decrease (***p < 0.001). Western blotting results are almost consistent with the real-time PCR outputs.

Discussion
The current study's objective was to develop MSN-based nanocarriers for tumor-specific gene therapy and induce a chemosensitized phenotype in multidrug-resistant carcinoma cells. MSNs have been employed as siRNA delivery vehicles due to their favorable properties, including a network of hollow cavities, high loading capacity, large surface areas (ca. 1000 m 2 /g), high pore volumes (ca. 1 cm 3 /g), biocompatibility, and facile surface modification 18,19 . The fabrication of MSNs is cost-effective and straightforward. However, some limitations remain in the successful translation of this system to the bedside. Recently, several promising results demonstrating the remarkable potential of MSNs for drug delivery in cancer therapy have been gathered 20,21 .
NH 2 -MSNs were designed and synthesized using a base-catalyzed sol-gel method for siRNA loading. TEM images showed successful syntheses of NH 2 -MSNs and represented the bright and dark areas corresponding to the pores and the silica walls of NH 2 -MSNs, respectively. TEM also confirmed organized nanochannels in a hexagonal structure. The particle size of NH 2 -MSNs was approximately 70 ± 12 nm, as determined by the ImageJ software in scanning electron microscopy (SEM) images, consistent with the TEM result (Fig. 4). The polydispersity index (PDI) value of 0.5 for NH 2 -MSNs indicated that the nanoparticles were approximately monodisperse; consistent with electron microscopy images. The BET isotherm curve obtained from the NH 2 -MSNconfirmed the typical type-IV isotherm curve with a hysteresis loop and a noticeable step between 0.3 and 0.4 p/p0, indicating the presence of mesoporous structures 22 .
The hexagonal arrays of pores in the NH 2 -MSNs were confirmed by XRD and BET results. The XRD pattern (Fig. 2C) indicated a slight shift in the d 00 value, which could result from the addition of TMB as a micelle pore expander and the covalent binding of amine groups in the pores or on the walls of MSNs 23 . MSNs were aminated with APTES to facilitate siRNA loading, as tiny aminopropyl grafts on the mesopores enhance biocompatibility and adsorption capacity. The disappearance of the 960 cm −1 bands on FTIR spectra demonstrated that MSN surfaces were adequately aminated (Fig. 2D). Additionally, several peaks can be attributed to the C-H stretching vibrations of the aliphatic chain at 2900 cm −1 , implying that APTES is anchored to MSN surfaces 22,24,25 .
The ability of NH 2 -MSN to interact with siRNA was confirmed using gel assay (Fig. 3A), indicating that the loading capacity of siRNA is approximately 20% w/w, which is a high loading record for a modified medium-sized mesopore of approximately 5 nm ( Fig. 2A). Such pore size is freely available for a common siRNA around 2 nm 26 . While some researchers demonstrated that large-pore MSNs could adsorb more significant amounts of siRNA within the pores, they are susceptible to RNase diffusion and siRNA digestion due to the RNase's small size. Furthermore, the assembled gatekeeper or protective shield on the large-pore MSNs needs complex multistep www.nature.com/scientificreports/ preparation processes 27,28 . On the other hand, the positive charge and porous nature of NH 2 -MSN enable siRNA adsorption on the outer surface and loading within the pores after a specific orientation is adopted. Physically attached siRNAs on the nanoparticle surfaces may be easily degraded by an endonuclease, and prolonged incubation time may allow siRNAs to be released from pores before delivery to the target site. The NH 2 -MSN-siRNA surface was coated with chitosan to protect siRNA from enzymatic cleavage, and gel assay analyses revealed that chitosan coating could protect siRNA from enzymatic cleavage (Fig. 3B,C). The negative charges of the siRNA are needed for efficient electrostatic interactions of chitosan. In addition, the protective effect of chitosan coats facilitated the surface modification of NH 2 -MSN with multifunctional ligands. PEGylation of chitosan was used in this study to increase its water solubility, prevent the reticuloendothelial system from detecting the chitosanbased nanocarriers, and prolong their circulation time 29 . Besides, the medium-sized PEG created a suitable linker for the folate and TAT moieties (Supplementary Fig. S3). Conjugation of folate-PEG or TAT-PEG to chitosan covered the positive charge in modified chitosan-coated NH 2 -MSN-siRNA, which was in agreement with zeta-potential measurement results ( Table 1). The zeta potential value confirmed the presence of positive charges on both inner and outer surfaces of NH 2 -MSN. The zeta potential of MSNs was − 10.8 mV due to the silicon hydroxyl groups, whereas the zeta potential of NH 2 -MSN nanoparticles was + 16 mV due to the amine groups. The electrostatic interaction of negatively charged siRNA with cationic NH 2 -MSN groups resulted in a shift in the zeta potential value from positive to negative. Then, coating the NH 2 -MSN-siRNA with chitosan increased the zeta-potential values from − 15 to + 18, due to the primary amine groups in chitosan. The zeta-potential values of chitosan-PEG-coated NH 2 -MSN-siRNA were negative due to the presence of large hydroxyl groups on PEG(3000) and balanced the chitosan's positive charges 30,31 . Modification of chitosan with PEG(3000)-folate or PEG(3000)-TAT resulted in positive zeta-potential values (around 9 mV), confirming that the targeting moieties (ligands) are covalently bonded to the PEG free ends. The presence of folate or TAT can mask the negatively charged hydroxyl groups on PEG(3000), increasing zetapotential values. The zeta-potential of NH 2 -MSN-siRNA-chitosan-PEG(3000)-folate/TAT was 1.7 mV due to the presence of more PEG(3000) molecules on this coating, confirming the successful grating process [31][32][33] .
The final nanoparticle size was around 100 nm, which was ideal for EPR-induced tumor tissue accumulation. Although MSNs have been widely reported as safe and biocompatible for intracellular delivery, several studies have found that MSNs are toxic above 100 µg/ml when APTES is incorporated 34 . Others revealed that APTES-modified nanoparticles were not toxic to cells even at higher concentrations 35 . In this study, the presence of APTES or functionalized chitosan on the MSN surface did not affect cell viability in EPG85.257-RDB or HeLa-RDB cells in the microscopic view ( Supplementary Fig. S10); however, MTT results showed reduced viability of HeLa-RDB cells after treatment with NH 2 -MSN or functionalized chitosan-coated NH 2 -MSNs (Fig. 5). There is evidence in a report that treatment of HeLa cells with functionalized MSNs induces MTT exocytosis and decreases intracellular formazan crystals, but this is not associated with decreased cell viability 36 . PEG and chitosan are biodegradable polymers in their entirety, and their presence on the MSN structure has no effect on cellular viability at therapeutic concentrations 16 .
Final constructs were used as targeting agents for HeLa-RDB cells expressing high levels of folate receptors and EPG85.257-RDB cells expressing low levels of folate receptors. HeLa-RDB cells internalized DiI-NH 2 -MSNchitosan-PEG-folate/TAT at a much higher rate than DiI-MSN, indicating that folate conjugation enhanced HeLa-RDB cell uptake. In contrast, EPG85.257-RDB cells inefficiently internalized DiI-NH 2 -MSN-chitosan-PEGfolate/TAT due to low folate expression. Moreover, the cellular internalization of DiI-NH 2 -MSN-chitosan-PEGfolate/TAT by HeLa-RDB cells was blocked in the presence of excess free folates. Simultaneously, this structure was not taken up by EPG85.257-RDB cells, indicating that the folate receptor on HeLa-RDB cells promotes endocytosis via folate ligands on the corresponding nanostructures (Fig. 6). There are numerous studies on the benefits of folate receptors in cancer targeting [37][38][39][40] . This paper demonstrated that conjugating folate with a nanostructure increase targeted therapy effectiveness because it induces tumor cells to uptake nanoparticles, selectively. On the other hand, pegylation decreased chitosan charges and nanoparticle uptake, consistent with the previous research 41 . TAT-modified nanocarriers accumulated a higher degree of NH 2 -MSNs within the cells, agreeing with the flow cytometry data. The proton sponge effect of chitosan and TAT promotes endosomal escape and release of the encapsulated cargo 42 . This outcome confirmed the functionality of the nanocarrier in the targeted delivery of siMDR1. Even though the NH 2 -MSN-siRNA and NH 2 -MSN-siRNA-chitosan nanoparticles decreased the MDR1 transcript and protein in both cells, more silencing results were observed in HeLa-RDB cells in the presence of final constructs (Fig. 7). Targeting NH 2 -MSN-siRNA with folic acid and TAT (NH 2 -MSN-siRNA-CS-PEG-TAT/Folate) improved the siRNA delivery in folate receptor-rich HeLa-RDB cells in comparison with NH 2 -MSN-siRNA due to effective uptake of folate-targeted nanoparticles. EPG85.257-RDB gene silencing using NH 2 -MSN-siRNA-CS-PEG-Folate constructs did not significantly decrease MDR1 translation in comparison to the NH 2 -MSN-siRNA. This was due to the low expression of folate receptors on the surface of these cells, which led to nonspecific and lower uptake of nanoparticles. According to the literature, both TAT and chitosan have proton sponge properties and several studies reported that chitosan has a lower proton sponge capacity and transfection [43][44][45][46] . TAT and chitosan enter cells via a nonspecific uptake mechanism that is dependent on the nature of the cell membrane and varies from cell to cell. On the contrary, targeted dual nanostructures have a higher concentration of PEG molecules on their surface, which may hinder cell entry and endosomal escape in NH 2 -MSN-siRNA-CS-PEG-folate/TAT structure.

Conclusion
In this study, a dual tumor-targeting design of MSNs for gene delivery was successfully fabricated. The aminated MSNs with medium pore sizes of approximately 5 nm and a surface area of 512 m 2 /g demonstrated a high siRNA loading capacity. Folate and TAT were attached to chitosan using a small PEG spacer to target the cancer cells  MSN synthesis and surface amination. NH 2 -MSNs with large pores were prepared in the presence of TMB as a micelle swelling agent. Initially, 1 g CTAB was dissolved in 480 ml deionized water, the temperature of the mixture was increased to 60 °C, 7 ml TMB was added, and the mixture was stirred for 5 h. The solution was then poured with 3.5 ml of 2 M sodium hydroxide. Following that, the temperature was increased to 80 °C, and 5 ml TEOS and 0.2 ml APTES were simultaneously inset dropwise into the solution for 5 min, followed by another 2 h of stirring. After cooling, the white precipitate was washed with generous amounts of 96% ethanol, deionized water, and freeze-dried. By incorporating APTES into the silanization process, the pores and surface were functionalized with amine groups. Additionally, the outer surface of the nanoparticles was covalently aminated. Afterward, 1 g of nanoparticles was dispersed in 100 ml ethanol containing 2% v/v APTES and stirred for 12 h at 60 °C under reflux conditions. NH 2 -MSNs were collected with centrifugation at 14000 rpm for 15 min at ambient temperature. Finally, 1 g nanoparticle was resuspended in 200 ml 96% ethanol containing 2 g NH 4 NO 3 , and the suspension was refluxed at 60 °C for 12 h. The precipitate was collected via centrifugation, and the process was repeated at least three times to remove the surfactant from the pores completely. For future use, the surfactant-free NH2-MSNs were freeze-dried and stored in 96% ethanol 47 . A transmission electron microscope (TEM, Zeiss, Germany), a scanning electron microscope (SEM, JEOL, Japan), X-ray diffraction (XRD, STOE, Germany), and Fourier transform infrared spectroscopy were used to physiochemically examine the nanoparticles (FTIR, Bruker, Germany). A nitrogen adsorption analyzer was used to determine the surface area (Belsorp-mini, Japan). All samples' particle size and zeta potential were determined in deionized water using a DLS (Malvern, Worcestershire, England).
Chitosan modification and targeting. Four sequential syntheses were used to generate MSN targeting coats. Amine moieties on chitosan were determined using 1 H-NMR in a solution containing 2% deuterated acetic acid.The pH was adjusted to 6.0 after dissolving 10 mg chitosan (equivalent to 40 mol NH2 groups) in www.nature.com/scientificreports/ 2 ml of 2% acetic acid. Polyethylene glycol, folate, TAT, or double folate/TAT were used to conjugate chitosan. Around 0.8 mol of MAL-PEG(3000)-NHS, TAT-PEG(3000)-NHS, folate-PEG(3000)-NHS, or a mixture of TAT-PEG(3000)-NHS and folate-PEG(3000)-NHS was added to 2 ml chitosan solution (1:50 ratio to amine group) and stirred in the dark at room temperature. After 3 h, the pH was raised to 7 for another 24 h. For 48 h, the solution was dialyzed against deionized water (cutoff 12 kD, SLS, UK). The modified chitosans included chitosan-PEG(3000), chitosan-PEG(3000)-folate, chitosan-PEG(3000)-TAT, chitosan-PEG(3000)-folate/TAT and were collected after freeze-drying and stored at 4 °C. The conjugations were confirmed and quantified using NMR 49-51 . Formulation and evaluation of MSNs coated with functionalized chitosan. NH 2 -MSNs loaded with optimized RNA were briefly coated with various functionalized chitosans, and 5 µl of the MSN stock (0.1% w/v) plus 25ρmol of siRNA were stirred gently for 30 min to form the NH 2 -MSN-siRNA complex. Then, 2 µl of each functionalized-chitosan solution (0.5% w/v in 2% acetic acid, pH: 6) was added to the RNA-loaded NH 2 -MSNs and stirred for 24 h. The chitosan-coated NH 2 -MSN-siRNA were collected by centrifugation (14000 rpm, 15 min) and washed twice with deionized water 52 . Zeta potential was measured to verify the chitosan coating on the surface of the MSN structures. Finally, the protective role of chitosan coats was evaluated by incubating the nanostructures with 0.25% RNase-A at optimal conditions for varying times. The enzyme was inactivated at 60 °C for 5 min. Heparin (200 IU/µl) was added to the complexes for 10 min to release the siRNA, and agarose gel electrophoresis was performed to identify the siRNA degradation level 27,42,53 . For the stability of nanoparticles, siRNA leakage from nanoparticles was studied for 4 days in PBS by UV-spectroscopy at 260 nm. Briefly, the siRNA loaded nanoparticles were dispersed in PBS at 4 °C. At a predetermined sampling time (1, 2, 3, and 4 day), the suspension was centrifuged and samples were taken from the supernatant. siRNA quantities of the supernatant were determined by UV-spectroscopy at 260 nm.
HeLa transfection with multidrug-resistant pumps. Nanoparticle endocytosis and competition analysis. Cellular uptake was assessed by flow cytometry (CyFlow Space ® , Munster, Germany) using the FL-2 channel. HeLa-RDB and EPG85.257-RDB cells (2 × 10 5 cells per well) were seeded into 12-well plates. Initially, 10 μl of DiI fluorescent dye (1 mg/ml in ethanol) was loaded into 10 mg of NH 2 -MSN. The DiI-MSNs were washed three times with ethanol and centrifuged to collect them. The fluorescent nanosilica particles were then coated with various modified chitosans, and cells were treated for 2 h at 37 °C with 100 μg/ml of the modified-fluorescence nanovehicles 39 . After washing the cells with PBS, the mean intracellular fluorescence intensity indicated the constructs' endocytosis activity. The same procedure was used to conduct a competitive uptake assay in the presence and absence of 5 µM folate as the competitor. Data were analyzed using FlowJo software (version 7.6.1) and compared to the control samples (treated cells with uncoated DiI-NH 2 -MSN) 54 .
Molecular evaluation of gene silencing. Real-time PCR and Western blotting were used to determine the MDR1 level. In 6-well plates, HeLa-RDB and EPG85.257-RDB cells were seeded. After 24 h, cells were transfected with optimized si-MDR1-loaded nanovehicles (25 pmol si-MDR1/5 µg MSNs) coated with specific chitosans. After 48 h, RNA was extracted, cDNA was synthesized, and real-time PCR was performed according to the Pfaffl method on a Rotor-GeneQ instrument (Qiagen, Germany) using the SYBR Green kit (Qiagen kits, Germany). Simultaneously, proteins were extracted using a lysis buffer (8 M urea, 2 M thiourea, Tris 10 mM, pH = 8.0), and the Bradford method was used to quantify the proteins colorimetrically. Proteins were run in a 12% SDS-PAGE according to the Laemmli method and then electroblotted to the nitrocellulose membrane www.nature.com/scientificreports/ using a semi-dry Trans-Blot apparatus (Bio-Rad, Richmond, CA). Blots were briefly blocked by 5% skimmed milk, incubated with a primary anti-Pgp mouse monoclonal IgG (1:500 v/v) overnight at 4 °C, treated with an HRP-conjugated secondary antibody (1:5000 v/v) for 2 h, and photos were taken using the luminal system on a blot scanner (LiCor, Lincoln, NE). For quantification procedures, β-actin was considered as the internal normalizer for both quantifications 48 .

Statistical analyses.
Results were expressed as mean ± standard deviation of at least three independent experiments and analyzed using graph pad prism. P values < 0.05 were considered statistically significant.
Ethical statement. The authors have read and adhered to the journal's ethical standards for manuscript submissions. This article does not contain any studies with human participants or animals performed by any of the authors. We declare that the submitted manuscript does not contain previously published materials and is not considered for publication elsewhere. All the authors have contributed to conception and design, collection, analysis, and interpretation of data, writing or revising the manuscript, or providing guidance on the research's execution.

Data availability
The datasets generated or analyzed during the current study are available on request from the corresponding author.