Novel Collagen-Polyphenols-Loaded Silica Composites for Topical Application

Lesions can affect skin functions and cause a simple issue, such as dehydration, or more challenging complications, such as bacterial infections. The purpose of this study was to design composites for topical application that can prevent and/or assist in bacterial infections and support cell regeneration using natural components. A polyphenolic extract obtained from Salvia officinalis was embedded in functionalized mesoporous silica nanoparticles for better stability, followed by their distribution into a collagen porous scaffold. The resulting polyphenols-loaded MSN exhibited enhanced antibacterial activity and good cytocompatibility. Improved thermal stability of the collagen porous scaffold was obtained due to the presence of the functionalized MSN. For the first time, collagen-polyphenols-loaded silica composites were reported in the literature as potential wound dressings. The newly developed composites showed excellent sterility.


Introduction
The skin is the largest human organ, and, in case of injury, it is the first and most affected. It has three main functions: protection, thermoregulation, and sensation. Lesions caused to the skin affect these functions and can lead to multiple complications, such as bacterial infections, hypothermia, impairment of the immune system, dehydration, and wounds that, in the long-term, turn into scars [1]. Among skin lesions, burn wounds represent one of the most complex and traumatizing injuries associated with high mortality rates, causing not only physical impairment for the patient but also mental distress [2]. The World Health Organization estimates that, every year, 180,000 burn injuries have a fatal outcome, most of them occurring in low-and middle-income countries [3].
Collagen is used in wound dressing applications in various forms depending on the manufacturing technique: porous scaffolds, electrospun matrices, films, hydrogels, or microspheres [4]. The interconnected pore array of porous scaffolds allows cells access into the scaffold and ensures their adhesion to the collagen sheets promoting cell proliferation [5]. Furthermore, the porous structure enables bidirectional transport: (i) oxygen passes through the scaffold toward the epithelial tissue, preventing hypoxia and thus enhancing the healing process, and (ii) the exudates and metabolic wastes are absorbed from the wound [6]. Marine sources can be used to obtain collagen and its derivatives, such as gelatin or peptides [7]. Collagen obtained from marine sources is considered a better alternative to mammalian ones due to the high risks associated with the latter, such as initiation of immune reactions or transmission of zoonotic diseases (e.g., spongiform encephalopathy Aldrich), 3-(triethoxysilyl)propionitrile (CETES, 97%, Sigma-Aldrich), ammonium chloride (Sigma-Aldrich), toluene (Riedel de Haen, Honeywell Riedel-de Haën, Seelzer, Germany), acetone (Sigma, Merck Group, Darmstadt, Germany), absolute ethanol (Sigma-Aldrich), 36.5-38% hydrochloric acid, sulfuric acid 98% (Merck), acetic acid (>99.7%, Sigma-Aldrich Co.), Folin-Ciocalteu (Sigma-Aldrich), 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS, Sigma-Aldrich Co.), and potassium persulphate (K 2 S 2 O 8 , Sigma-Aldrich Co.) were used as received. Ultrapure water (Millipore Direct-Q3 UV water system, version Q3 UV, product no. C9185, Merck Group, Darmstadt, Germany, equipped with a Biopack UF cartridge) was used for all experiments. Collagen was provided by the National Institute of Research and Development for Biological Sciences, Bucharest, Romania and prepared as previously described [36].

Mesoporous Silica Carriers Synthesis
Pristine mesoporous silica nanoparticles (MSN) were prepared through a newly reported sol-gel synthesis. Tetraethyl orthosilicate (TEOS) and triethanolamine (TEA) were first heated at 90 • C for 20 min, under static conditions, and then 3.5% CTAC aqueous solution was added to the mixture under vigorous magnetic stirring. Last, a PEG aqueous solution (50 mg/mL) was added dropwise, and the reaction mixture was heated at 100 • C overnight. The TEOS: TEA: CTAC: PEG: H 2 O molar ratio was 1: 0.2: 10.4: 0.05: 130. The structure-directing agent was removed first by extraction in an ethanolic ammonium chloride solution, followed by calcination at 550 • C for 5 h.
The functionalized mesoporous silica nanoparticles were prepared by grafting organic functional groups on pristine MSN through the post-synthesis approach using MSN: organosilane molar ratio of 1: 0.2. The sample denoted MSN-NH 2 was prepared using (3aminopropyl)triethoxysilane and the method previously described [11]. The MSN-COOH support was obtained by condensation reaction between 3-(triethoxysilyl)propionitrile and silanol groups of pristine MSN, followed by hydrolysis of the cyanide groups to carboxylic ones in 50% wt of aqueous sulfuric acid solution [11].

Polyphenols-Loaded MSN Preparation
Polyphenolic extract was obtained from Salvia officinalis (common sage) by ultrasoundassisted extraction (Bandelin Sonorex Digitec ultrasonic bath-Berlin, Germany) at 40 • C, in absolute ethanol, at a vegetal material/solvent ratio of 1/30 (w/v). The extraction was carried out in three 15 min. steps, with solvent replacement in each stage. The extract was evaporated using a rotary evaporator (RE100-Pro, DLAB Scientific, Beijing, China) until it reached a constant mass and then was redissolved at a desired concentration. After preparation, Salvia officinalis polyphenolic extract (denoted S) was adsorbed into functionalized MSN mesopores through impregnation, followed by solvent evaporation under vacuum. First, the mesoporous carriers were dried under vacuum, at room temperature, for 12 h, and then the corresponding volume of polyphenolic extract (15 mg/mL) was added to the support to obtain composites with a theoretical content of 25 %wt extract. The polyphenolic-extract-loaded MSN were denoted S@MSN-x, where x = NH 2 or COOH.

Radical Scavenger Activity Testing
Radical scavenger activity (RSA) of polyphenols-loaded MSN was assessed against 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)diammonium salt (ABTS) according to the method developed for solid samples. The detailed method is presented elsewhere [37]. In summary, 2.95 mL ABTS˙+ solution was added to 50 µL free polyphenolic extract, to the corresponding amount of polyphenols-loaded MSN, or to the corresponding amount of silica carriers. The experiment was carried out in duplicate under dark conditions in glass vials under continuous shaking using an orbital shaker (Grant bio-PSU-10i, Grant Instruments, Cambridge, United Kingdom) for 6 h. The solids were separated, and the absorbance of the solution was measured at 753 nm (Shimadzu UV-1800, Shimadzu Corporation, Kyoto, Japan). For the free extract, results were presented as Trolox equivalents

Antimicrobial Activity Testing
Antimicrobial activity was tested against both Gram-positive and Gram-negative bacterial strains, P. aeruginosa (ATCC 27853) and S. aureus (ATCC 25923), respectively. Quantitative analysis of the antibacterial potential of the extract and encapsulated extracts was carried out by using the method of serial microdilutions in Mueller-Hinton broth in 96-well plates. The aqueous suspensions were serially 2-fold diluted; negative control was represented by culture medium, while positive one by bacterial cultures. Plates containing fresh standardized bacterial inoculum and the tested extracts were incubated at 37 • C for 24 h. The minimum inhibitory concentration (MIC) was established macroscopically as the last concentration at which the turbidity was observed, and it was further confirmed spectrophotometrically. The minimum bactericidal concentration (MBC) was established by further cultivating on Mueller-Hinton agar the samples incubated with the established MIC for each tested sample. The concentration that managed to kill 99.9% of the bacterial cells after 24 h was confirmed as MBC. Experiments were performed in duplicate.

Cytocompatibility Testing
The biological evaluation of samples was performed in agreement with the law for ethics in research 206/2004, completed with OUG 28/2011, harmonized with European specifications in force. Standardized cell lines were used for the in vitro experiments by following the manufacturer guidelines. No primary cell cultures or animal studies were used in the experimental set-up.
Human normal keratinocytes (HaCat) were cultivated in RPMI supplemented with 10% FBS and 1% PSN antibiotic mixture at 37 • C and 5% CO 2 . For the cytocompatibility assay, the cells were seeded in 96-well culture plates at a density of 5 × 10 4 cells/mL and incubated in standard conditions to allow cell adhesion. After 24 h, the culture medium was replaced with fresh medium and the tested aqueous suspensions at various concentrations, 5, 30, 50, and 100 µg/mL. The evaluation was carried out at 24 h and 72 h, and the control was considered the untreated cell culture. Results were expressed as mean value ± standard deviation (n = 3). Statistical analysis of the data was performed using two-tailed, twosample equal variance Student's t-test (Microsoft Office 365 Excel software) on each pair of interest.

Cytotoxic Effect
Human Mel-Juso skin carcinoma cells were cultivated in RPMI supplemented with 10% FBS and 1% PSN antibiotic mixture at 37 • C and 5% CO 2 . To determine the cytotoxic potential, cells were seeded and treated under the same conditions as the normal human keratinocytes, and the MTT test was performed at the same time points.

Apoptotic Effect
Flow cytometry analysis was conducted as previously described [38]. Mel-Juso cells were seeded in 12-well culture plates, allowed to attach for 24 h, and then treated with the indicated concentrations of samples for 72 h at 37 • C. Cells were trypsinized, washed with phosphate buffered saline (PBS), and collected by centrifugation at 1500 rpm for 5 min at room temperature. Apoptosis was analyzed using Dead Cell Apoptosis Kit. Aliquots of cells (1 × 10 5 cells) were double stained with fluorescein-isothiocyanate-labeled Annexin-V and propidium iodide for 15 min at room temperature in the dark according to the manufacturer's instructions (Invitrogen). Immediately after staining, samples were analyzed on a LSR II BD flow cytometer (Beckton Dickinson, San Jose, CA, USA).

Preparation of Collagen Porous Scaffold Composites Preparation
Collagen porous scaffolds (C) were prepared by freeze-drying the collagen solution. Thus, pristine collagen porous scaffolds were prepared from 0.5 mL collagen acidified (0.1 M acetic acid) solution (1.06 mg/mL) and 0.5 mL water. Samples were lyophilized at −40 • C for 24 h (Martin Christ Alpha 2-4 LSCbasic, Martin Christ Gefriertrocknungsanlagen GmbH, Osterode am Harz, Germany).
The collagen porous scaffold-polyphenols-loaded silica composites were prepared from 0.5 mL collagen acidified solution (1.06 mg/mL) and 0.5 mL S@MSN-x suspension in water. Thus, common sage extract-loaded functionalized MSN were dispersed in water assisted by ultrasounds and then the suspension was added to the collagen solution. Before lyophilization, the mixture was placed on an orbital shaker for 1 h to allow a good dispersion of the S@MSN-x (10 or 20 mg/mL) into the collagen solution.

Sterility Testing
Sterility testing of collagen composites was performed by direct inoculation, as described in the European Pharmacopoeia [39]. Thus, collagen porous scaffolds and collagenpolyphenols-loaded silica composites were placed on two types of solid media (nutrient agar and yeast extract peptone agar) and incubated for 14 days at 37 • C.

Characterization Methods
The texture parameters of the functionalized mesoporous carriers were computed from the nitrogen adsorption/desorption isotherms acquired at 77 K on a Quantachrome Autosorb iQ 2 gas sorption analyzer (Quantachrome Instruments, Boynton Beach, FL, USA). Pore size distribution and average pore diameter were calculated with Non-Local Density Functional Theory (NLDFT), specific surface area was determined in the range of relative pressure, p/p 0 , 0.02-0.25 with Brunauer-Emmett-Teller (BET) method, and total pore volume was assessed at p/p 0 = 0.99.
Structural evaluation of MSN carriers was carried out by small-angle X-ray diffraction (XRD) on a Rigaku MiniFlex II diffractometer (Rigaku Corporation, Tokyo, Japan) using Cu-Kα radiation, in the range of 1.2-6 degrees (scanning speed 0.5 deg/min).
The chemical composition of the collagen-polyphenol-loaded silica composites was assessed by FTIR spectroscopy using the KBr technique. The FTIR spectra were recorded between 4000 and 400 cm −1 , with 64 scans, on a Bruker Tensor 27 spectrometer (Bruker Corporation Optik GmbH, Bremen, Germany).
The content of functional groups on the MSN carriers and common sage polyphenolic extract content was evaluated by thermogravimetric analysis (TGA, Netzsch STA 2500 Regulus, Selb, Germany) between 30 • C and 800 • C, heating rate 10 • C/min. All TGA measurements were performed under synthetic air.
The thermal stability of collagen and its composites was evaluated by differential scanning calorimetry on a Mettler Toledo (Greifensee, Switzerland) DSC 823e calorimeter operated at an N 2 flow of 80 mL/min and a heating rate of 5 • C/min between 20 • C and 200 • C.
The morphology of MSN and composites was evaluated through scanning electron microscopy (SEM) and transmission electron microscopy (TEM). SEM images were collected on a Tescan Vega 3 LMH microscope (Brno, Czech Republic) coupled with an energy dispersive X-ray (EDX) detector, and TEM micrographs were recorded on a TECNAI F30 G2 S-TWIN microscope (Hillsboro, Oregon, USA) with a maximum accelerating voltage of 300 kV with a field emission electron gun.

Results
Prior to development of collagen-polyphenols-loaded silica composites, all components were thoroughly characterized. The structural and textural characteristics of MSN were evaluated by small-angle XRD, nitrogen adsorption/desorption isotherms, SEM, TEM, TGA, and FTIR.

MSN Carriers Characterization
The first step in design of novel composites consisted of preparation of MSN, which were used as a carrier for the polyphenolic extract. Pristine MSN were prepared by the sol-gel method, and, after template agent removal, they were further used to obtain functionalized MSN. Structural, textural, and morphological characteristics for pristine and functionalized silica carriers were evaluated.
The small-angle XRD pattern of the pristine MSN (Figure 1a, red curve) indicated a pseudohexagonal disordered pore array displaying an intense broad diffraction peak corresponding to the Bragg reflection and a small shoulder that can be attributed to the overlapped reflections (110) and (200), which are characteristic of the hexagonal pore array. Introduction of functional groups reduces the intensity of the main diffraction peak, as shown for MSN-NH 2 (blue curve) and MSN-CN (green curve). Hydrolysis of nitrile groups to form carboxylic acid moieties, in the presence of sulfuric acid, determined a more pronounced effect on the main diffraction peak reduction because of the hydrolysis reaction that produced a collapse of the mesostructure.
A decrease in textural parameter values (specific surface area, pore volume, and average pore diameter) of functionalized samples compared to pristine MSN is also noted ( Table 1). The synthesis procedure yielded nanoparticles with high porosity, calculated from nitrogen adsorption/desorption isotherms ( Figure 1b). Thus, pristine MSN have a large specific surface area of 977 m 2 /g and a total pore volume of 1.43 cm 3 /g, which enabled further modification of their surface with organic groups. Attachment of functional groups through post-synthesis method reduced the specific surface area and pore volume for MSN-NH 2 and MSN-COOH carriers to 512 m 2 /g and 388 m 2 /g and 0.42 cm 3 /g and 0.47 cm 3 /g, respectively. The average pore diameter of the functionalized carriers also decreased compared to that of pristine MSN The content of functional groups was determined from thermogravimetric analy ( Figure 2 and Table 1), disregarding the mass loss up to 100 °C, corresponding to the f endothermic event, which is assigned to physically adsorbed water on the silica surf of the mesoporous carriers. The mass loss between 100 °C and 650 °C was attributed the thermal decomposition of the organic groups grafted on the surface of the silica. T content of organic functional groups (FG) on silica surface is listed in Table 1. The content of functional groups was determined from thermogravimetric analysis ( Figure 2 and Table 1), disregarding the mass loss up to 100 • C, corresponding to the first endothermic event, which is assigned to physically adsorbed water on the silica surface of the mesoporous carriers. The mass loss between 100 • C and 650 • C was attributed to the thermal decomposition of the organic groups grafted on the surface of the silica. The content of organic functional groups (FG) on silica surface is listed in Table 1.
The content of functional groups was determined from thermogravimetric analysis ( Figure 2 and Table 1), disregarding the mass loss up to 100 °C, corresponding to the first endothermic event, which is assigned to physically adsorbed water on the silica surface of the mesoporous carriers. The mass loss between 100 °C and 650 °C was attributed to the thermal decomposition of the organic groups grafted on the surface of the silica. The content of organic functional groups (FG) on silica surface is listed in Table 1.  SBET-specific surface area; dDFT-average pore diameter; V-total pore volume; Vd < 10 nm-pore volume for pores with diameter less than 10 nm; FG-functional groups.
The morphology of MSN was assessed by SEM and TEM. The acquired SEM image ( Figure 3a) shows that the nanoparticles have a spherical shape and uniform dimensions. The diameter of the nanoparticles was determined from SEM images by measuring at least 100 nanoparticles, and the average diameter was 105 ± 13 nm. The TEM image ( Figure 3b) revealed more details of the nanostructure of the mesoporous carriers. The well-defined spherical nanoparticles present a mesopore array consisting of worm-like and cylindrical channels, consistent with the XRD results ( Figure 1a).  S BET -specific surface area; d DFT -average pore diameter; V-total pore volume; V d < 10 nm -pore volume for pores with diameter less than 10 nm; FG-functional groups.
The morphology of MSN was assessed by SEM and TEM. The acquired SEM image ( Figure 3a) shows that the nanoparticles have a spherical shape and uniform dimensions. The diameter of the nanoparticles was determined from SEM images by measuring at least 100 nanoparticles, and the average diameter was 105 ± 13 nm. The TEM image ( Figure 3b) revealed more details of the nanostructure of the mesoporous carriers. The well-defined spherical nanoparticles present a mesopore array consisting of worm-like and cylindrical channels, consistent with the XRD results ( Figure 1a).   Successful functionalization of MSN with organic moieties was demonstrated by FTIR spectroscopy (Figure 4). All carriers show the vibrations of physically adsorbed water molecules (1634 cm −1 ) and the vibration bands specific to the silica matrix: symmetric and asymmetrical stretching vibrations of the Si-O-Si bond (800 cm −1 and 1086 cm −1 , respectively), Si-OH stretching vibrations (966 cm −1 ), Si-O symmetric bending vibration (468 cm −1 ), and the wide band corresponding to O-H stretching vibrations (2700-3800 cm −1 ). The characteristic vibrations of the functional groups grafted on the silica surface are observed: bending and wagging vibrations of the amine group (1502 cm −1 and 692 cm −1 , respectively) and the carbonyl stretching vibration of the carboxylic acid group (1716 cm −1 ). The complete hydrolysis of the nitrile group to carboxylic moiety is confirmed by replacement of the nitrile band (2260 cm −1 ) identified in the MSN-CN spectrum with the C=O stretching vibration (1716 cm −1 ) in the MSN-COOH spectrum (Figure 4).

Polyphenols-Loaded MSN Composites Characterization
The source of the polyphenolic extract was the Salvia officinalis plant. The details of extract characteristics are provided in the Supplementary Materials. The chromatogram (HPLC-PDA, Shimadzu Nexera 2, Shimadzu Corporation, Kyoto, Japan) of the polyphenolic extract at 330 nm, the wavelength corresponding to the quantification of rosmarinic acid, is presented in Figure S1. Similar to previously reported S. officinalis extracts [34,40], the polyphenolic ethanolic extract obtained by ultrasound-assisted extraction at 40 °C was rich in rosmarinic acid, along with small quantities of caftaric acid, chlorogenic acid, caffeic acid, and rutin hydrate (Table S1). The presence of common sage polyphenolic extract into the mesopores of MSN carriers was evidenced in the FTIR spectra ( Figure S2).

Polyphenols-Loaded MSN Composites Characterization
The source of the polyphenolic extract was the Salvia officinalis plant. The details of extract characteristics are provided in the Supplementary Materials. The chromatogram (HPLC-PDA, Shimadzu Nexera 2, Shimadzu Corporation, Kyoto, Japan) of the polyphenolic extract at 330 nm, the wavelength corresponding to the quantification of rosmarinic acid, is presented in Figure S1. Similar to previously reported S. officinalis extracts [34,40], the polyphenolic ethanolic extract obtained by ultrasound-assisted extraction at 40 • C was rich in rosmarinic acid, along with small quantities of caftaric acid, chlorogenic acid, caffeic acid, and rutin hydrate (Table S1). The presence of common sage polyphenolic extract into the mesopores of MSN carriers was evidenced in the FTIR spectra ( Figure S2).
Total polyphenolic content (TPC) was determined by Folin-Ciocalteu method, and it was expressed as gallic acid equivalents (GAE) according to the method described previously [34]. The value of TPC for the polyphenolic extract was 163.78 ± 1.09 mg GAE /g extract .
The amount of polyphenolic extract loaded into the mesopores of the functionalized MSN was determined by thermogravimetric analysis (Figure 2). Thermal decomposition of the polyphenolic extract (black curve) occurred in the temperature range of 30-650 • C, after which the mass reached a plateau corresponding to a 2.84% residual mass consisting of inorganic compounds present in the extract. The results indicated 23.9% extract content for S@MSN-COOH and 26.6 % for S@MSN-NH 2 .
Radical scavenger activity (RSA) of Salvia officinalis extract was evaluated against ABTS˙+ radical, and polyphenolic extract RSA was 294 ± 3 mg TE /g extract . Among our previously reported extracts obtained from S. officinalis [34,40], the one used for developing collagen-polyphenols-loaded silica composites had the highest RSA value quantified through ABTS assay. The same method was applied for RSA evaluation of extract-loaded MSN. Mesoporous carriers alone were also tested to assess their contribution to RSA. The results obtained for the polyphenols-loaded MSN were compared with the ones for the same amount of mesoporous carrier and free extract, respectively, in the extract-loaded MSN ( Figure 5).
Antibacterial activity of the common sage extract in comparison to that of the embedded extracts was assessed ( Table 2). Stronger microbial growth inhibition was observed for the encapsulated extract, lower concentrations being necessary to interfere with the bacterial multiplication, the lowest MIC (3.9 mg/mL) being observed for S@MSN-COOH on S. aureus, while the lowest MBC obtained for the same sample on the same bacterial strain for S@MSN-COOH and 26.6 % for S@MSN-NH2.
Radical scavenger activity (RSA) of Salvia officinalis extract was evaluated against ABTS˙+ radical, and polyphenolic extract RSA was 294 ± 3 mg TE/g extract. Among our previously reported extracts obtained from S. officinalis [34,40], the one used for developing collagen-polyphenols-loaded silica composites had the highest RSA value quantified through ABTS assay. The same method was applied for RSA evaluation of extract-loaded MSN. Mesoporous carriers alone were also tested to assess their contribution to RSA. The results obtained for the polyphenols-loaded MSN were compared with the ones for the same amount of mesoporous carrier and free extract, respectively, in the extract-loaded MSN ( Figure 5). Antibacterial activity of the common sage extract in comparison to that of the embedded extracts was assessed ( Table 2). Stronger microbial growth inhibition was observed for the encapsulated extract, lower concentrations being necessary to interfere with the bacterial multiplication, the lowest MIC (3.9 mg/mL) being observed for S@MSN-COOH on S. aureus, while the lowest MBC obtained for the same sample on the same bacterial strain was 7.81 mg/mL. Among the bacterial strains, S. aureus showed the highest susceptibility to the tested extract.
Cytocompatibility and antiproliferative activity of free and embedded polyphenolic extract was evaluated by MTT assay after 24 h and 72 h incubation. The tested extract is cytocompatible, whether it is free or embedded into functionalized MSN, with cell viability values greater than 70% for normal human keratinocytes for both time points tested (Figure 6a). After 72 h of treatment, the percentage of cell viability in the group treated with Salvia-officinalis-embedded extract was comparable to that registered in the control group, the free S. officinalis group. The percentages of viable cells varied from 100 % to 71 %. All tested concentrations, including the highest 100 µg/mL, proved not to dramatically interfere with cellular proliferation. Table 2. Antibacterial activity of free common sage extract and embedded in functionalized MSN.  Cytocompatibility and antiproliferative activity of free and embedded polyphenolic extract was evaluated by MTT assay after 24 h and 72 h incubation. The tested extract is cytocompatible, whether it is free or embedded into functionalized MSN, with cell viability values greater than 70% for normal human keratinocytes for both time points tested (Figure 6a). After 72 h of treatment, the percentage of cell viability in the group treated with Salvia-officinalis-embedded extract was comparable to that registered in the control group, the free S. officinalis group. The percentages of viable cells varied from 100% to 71%. All tested concentrations, including the highest 100 µg/mL, proved not to dramatically interfere with cellular proliferation. The ability of Salvia officinalis extract encapsulated in mesoporous silica matrices to inhibit the cell proliferation process of Mel-Juso cancer cells was investigated. The common sage-extract-loaded functionalized MSN were tested on skin carcinoma cell line in order to extend the possible applications of developed composites for cancer-suffering people for a synergistic effect: anticancer and healing of skin-wounds. Based on MTT results, S@MSN-NH2 statistically significantly decreased (p < 0.05) the number of viable cells (62.09 ± 0.72%) at a concentration of 100 µg/mL at 24 h post-treatment (Figure 6b). Cell proliferation was also decreased in the case of S@MSN-COOH at the same tested concentration (66.36 ± 1.14%). The trend was maintained after 72 h of treatment and had an even higher impact on the ability of cells to proliferate, beginning with 50 µg/mL for the free extract. In terms of embedded extract, the number of viable cells has already been significantly reduced from 30 µg/mL for both S@MSN-COOH and S@MSN-NH2, with percentages of viable cells in the range of 69.37 ± 0.86% and 68.63 ± 0.75%, respectively. The proportion of apoptotic cells with increasing sample dose of treatment added to the cultured Mel-Juso cells is shown in Figure 7. An increased proportion of apoptotic cells was detected compared to the control, 41.9%, 43.1%, and 46.2% when cultured Mel-Juso cells were incubated with samples S, S@MSN-COOH, and S@MSN-NH2, respec- The ability of Salvia officinalis extract encapsulated in mesoporous silica matrices to inhibit the cell proliferation process of Mel-Juso cancer cells was investigated. The common sage-extract-loaded functionalized MSN were tested on skin carcinoma cell line in order to extend the possible applications of developed composites for cancer-suffering people for a synergistic effect: anticancer and healing of skin-wounds. Based on MTT results, S@MSN-NH 2 statistically significantly decreased (p < 0.05) the number of viable cells (62.09 ± 0.72%) at a concentration of 100 µg/mL at 24 h post-treatment (Figure 6b). Cell proliferation was also decreased in the case of S@MSN-COOH at the same tested concentration (66.36 ± 1.14%). The trend was maintained after 72 h of treatment and had an even higher impact on the ability of cells to proliferate, beginning with 50 µg/mL for the free extract. In terms of embedded extract, the number of viable cells has already been significantly reduced from 30 µg/mL for both S@MSN-COOH and S@MSN-NH 2 , with percentages of viable cells in the range of 69.37 ± 0.86% and 68.63 ± 0.75%, respectively.
The proportion of apoptotic cells with increasing sample dose of treatment added to the cultured Mel-Juso cells is shown in Figure 7. An increased proportion of apoptotic cells was detected compared to the control, 41.9%, 43.1%, and 46.2% when cultured Mel-Juso cells were incubated with samples S, S@MSN-COOH, and S@MSN-NH 2 , respectively, at concentrations of 50 µg/mL. At a concentration of 100 µg/mL, there was a progressive increase in percentages of apoptotic cell populations, with values of 66.0%, 68.5%, and 69.3% for samples S, S@MSN-COOH, and S@MSN-NH 2 , respectively. A statistically significant increasing proportion (p < 0.05) of apoptotic cells were identified in the late apoptosis stage for the encapsulated extract at 100 µg/mL concentration.

Collagen Porous Scaffold Composites Characterization
Round-shaped collagen porous scaffolds were obtained after lyophilization with 15 mm diameter and 7 mm thickness (Figure 8a). The morphology of the collagen porous scaffold was assessed by SEM. Pristine porous scaffold exhibited a multilayered structure formed by sheets, with interconnected pores with diameters ranging from 0.05 to 0.6 mm (Figure 8b). Functionalized MSN were introduced into the collagen porous scaffold prior to lyophilization in various proportions to assess the structural features of the resulting collagen-MSN composites. The samples were denoted C@MSN-x (x = NH2 or COOH). Thus, the two types of functionalized MSN were added to the collagen suspension in various concentrations (10 or 20 mg MSN/mL collagen solution). At a higher concentration, MSN started to aggregate on collagen sheets, causing collapse of the scaffold structure under the weight of MSN (Figure 8e,f). At a concentration of 10 mg/mL, both types of MSN showed good integration into the collagen porous structure, being very well dispersed into the collagen scaffold (Figure 8c,d). The distribution of MSN in the collagen porous scaffold was examined by SEM coupled with energy dispersive X-ray analysis ( Figure S3). The image demonstrated the MSN uniform distribution into the scaffold.

Collagen Porous Scaffold Composites Characterization
Round-shaped collagen porous scaffolds were obtained after lyophilization with 15 mm diameter and 7 mm thickness (Figure 8a). The morphology of the collagen porous scaffold was assessed by SEM. Pristine porous scaffold exhibited a multilayered structure formed by sheets, with interconnected pores with diameters ranging from 0.05 to 0.6 mm (Figure 8b). Functionalized MSN were introduced into the collagen porous scaffold prior to lyophilization in various proportions to assess the structural features of the resulting collagen-MSN composites. The samples were denoted C@MSN-x (x = NH 2 or COOH). Thus, the two types of functionalized MSN were added to the collagen suspension in various concentrations (10 or 20 mg MSN/mL collagen solution). At a higher concentration, MSN started to aggregate on collagen sheets, causing collapse of the scaffold structure under the weight of MSN (Figure 8e,f). At a concentration of 10 mg/mL, both types of MSN showed good integration into the collagen porous structure, being very well dispersed into the collagen scaffold (Figure 8c,d). The distribution of MSN in the collagen porous scaffold was examined by SEM coupled with energy dispersive X-ray analysis ( Figure S3). The image demonstrated the MSN uniform distribution into the scaffold.
( Figure 8b). Functionalized MSN were introduced into the collagen porous scaffold prior to lyophilization in various proportions to assess the structural features of the resulting collagen-MSN composites. The samples were denoted C@MSN-x (x = NH2 or COOH). Thus, the two types of functionalized MSN were added to the collagen suspension in various concentrations (10 or 20 mg MSN/mL collagen solution). At a higher concentration, MSN started to aggregate on collagen sheets, causing collapse of the scaffold structure under the weight of MSN (Figure 8e,f). At a concentration of 10 mg/mL, both types of MSN showed good integration into the collagen porous structure, being very well dispersed into the collagen scaffold (Figure 8c,d). The distribution of MSN in the collagen porous scaffold was examined by SEM coupled with energy dispersive X-ray analysis ( Figure S3). The image demonstrated the MSN uniform distribution into the scaffold. FTIR spectra of proteins have specific spectral regions [41] where vibrations of the amide bonds can be identified. Therefore, the FTIR spectra of the collagen porous scaffolds (Figure 9a) showed amide I vibration (1651 cm −1 ) corresponding to C=O stretching FTIR spectra of proteins have specific spectral regions [41] where vibrations of the amide bonds can be identified. Therefore, the FTIR spectra of the collagen porous scaffolds (Figure 9a) showed amide I vibration (1651 cm −1 ) corresponding to C=O stretching vibration, with contributions from C-N stretching and N-H bending vibrations; amide II (1542 cm −1 ) assigned to C-N stretching band combined with N-H in-plane bending vibration; amide III (1244 cm −1 ) belonged to N-H bending vibration coupled with C-N stretching vibration; amide A (3300 cm −1 ) and amide B (3074 cm −1 ) ascribed to N-H stretching vibration and asymmetrical stretching of CH 2 . Aside from the specific amide bands, other vibrations of the skeletal structure of collagen were identified: CH 3 bending modes in skeletal proteins (1396 cm −1 ) and O-H bending vibration of COOH terminal groups that could overlap the C-H bending bands (1449 cm −1 ). In collagen-MSN composites, slight shifts in the bands of the collagen scaffold were noticed and displayed in Figure 9a.  The thermal stability of collagen-MSN composites in comparison with that of gen scaffold was assessed. DSC analyses of the samples (Figure 9b) showed an end mic event at 56.9 °C for the collagen porous scaffold, 71.2 °C for C@MSN-COOH, an °C for C@MSN-NH2. The DSC results showed an enhanced denaturation temperat collagen-MSN composites compared to that of pristine collagen porous scaffold.
After complete characterization of the collagen porous scaffold and establishi parameters regarding the amount of MSN that could improve the properties of the p scaffold, collagen-polyphenols-loaded MSN were prepared and characterized. FTIR tra showed intense amide I and amide II of the collagen porous scaffold (Figure 10 more intense bands in the 1350-1475 cm −1 wavenumber range due to C-H bending The thermal stability of collagen-MSN composites in comparison with that of collagen scaffold was assessed. DSC analyses of the samples (Figure 9b) showed an endothermic event at 56.9 • C for the collagen porous scaffold, 71.2 • C for C@MSN-COOH, and 74.2 • C for C@MSN-NH 2 . The DSC results showed an enhanced denaturation temperature of collagen-MSN composites compared to that of pristine collagen porous scaffold.
After complete characterization of the collagen porous scaffold and establishing the parameters regarding the amount of MSN that could improve the properties of the porous scaffold, collagen-polyphenols-loaded MSN were prepared and characterized. FTIR spectra showed intense amide I and amide II of the collagen porous scaffold (Figure 10), and more intense bands in the 1350-1475 cm −1 wavenumber range due to C-H bending vibrations that overlap bands of all three components of the composites. Moreover, in the range of 500-700 cm −1 , bands of C-H out-of-plane bending vibrations are distinguished.

Discussion
The developed collagen-polyphenols-loaded silica composites combine synthetic and natural components to obtain advanced materials. A polyphenolic extract was incorporated into synthetic mesoporous inorganic carriers and then included in a natural polymer matrix, a collagen porous scaffold.
MSN were prepared through sol-gel method using TEOS as silica precursor and CTAC as templating agent. The hydrolysis and condensation reactions of the silica precursor were catalyzed by TEA, which also assisted in achieving better particle size control. PEG is mostly used for coating of silica nanoparticles surface for biomedical applications because it shields the negative charge of the silica surface and helps in reducing the MSN interaction with biological molecules [42]. However, in the proposed synthesis procedure, PEG was not bound to the MSN surface; it was employed as an inhibitor of particles growth. The aim of this work was to obtain small silica nanoparticles that would enable good dispersion into the collagen porous matrix without influencing the microstructure of the porous scaffold. The MSN synthesis method yielded spherical nanoparticles with a diameter of around 100 nm and high porosity (SBET = 977 m 2 /g and V = 1.43 cm 3 /g).
Pristine MSN were further used for functionalized MSN synthesis. Moieties with acidic or basic nature able to form acid-base interactions and hydrogen bonding with the collagen matrix were grafted onto the MSN surface by a post-synthesis approach. Thus, 3-aminopropyl moieties were grafted on pristine MSN surface, creating MSN-NH2. For The morphology of both types of collagen-polyphenols-loaded silica composites was evaluated by SEM ( Figure 11). The S. officinalis-extract-loaded MSN were homogenously distributed into collagen sheets, although the functionalized MSN were slightly more agglomerated after extract loading in their mesopores.

Discussion
The developed collagen-polyphenols-loaded silica composites combine synthetic and natural components to obtain advanced materials. A polyphenolic extract was incorporated into synthetic mesoporous inorganic carriers and then included in a natural polymer matrix, a collagen porous scaffold.
MSN were prepared through sol-gel method using TEOS as silica precursor and CTAC as templating agent. The hydrolysis and condensation reactions of the silica precursor were catalyzed by TEA, which also assisted in achieving better particle size control. PEG is mostly used for coating of silica nanoparticles surface for biomedical applications because it shields the negative charge of the silica surface and helps in reducing the MSN interaction with biological molecules [42]. However, in the proposed synthesis procedure, PEG was not bound to the MSN surface; it was employed as an inhibitor of particles growth. The aim of this work was to obtain small silica nanoparticles that would enable good dispersion into the collagen porous matrix without influencing the microstructure of the porous scaffold. The MSN synthesis method yielded spherical nanoparticles with a Sterility testing of the collagen porous scaffolds and their analogue composites with polyphenols-loaded-silica revealed that, after 14 days, no yeasts or molds were developed on nutrient agar or yeast-extract peptone culture medium.

Discussion
The developed collagen-polyphenols-loaded silica composites combine synthetic and natural components to obtain advanced materials. A polyphenolic extract was incorporated into synthetic mesoporous inorganic carriers and then included in a natural polymer matrix, a collagen porous scaffold.
MSN were prepared through sol-gel method using TEOS as silica precursor and CTAC as templating agent. The hydrolysis and condensation reactions of the silica precursor were catalyzed by TEA, which also assisted in achieving better particle size control. PEG is mostly used for coating of silica nanoparticles surface for biomedical applications because it shields the negative charge of the silica surface and helps in reducing the MSN interaction with biological molecules [42]. However, in the proposed synthesis procedure, PEG was not bound to the MSN surface; it was employed as an inhibitor of particles growth. The aim of this work was to obtain small silica nanoparticles that would enable good dispersion into the collagen porous matrix without influencing the microstructure of the porous scaffold. The MSN synthesis method yielded spherical nanoparticles with a diameter of around 100 nm and high porosity (S BET = 977 m 2 /g and V = 1.43 cm 3 /g).
Pristine MSN were further used for functionalized MSN synthesis. Moieties with acidic or basic nature able to form acid-base interactions and hydrogen bonding with the collagen matrix were grafted onto the MSN surface by a post-synthesis approach. Thus, 3-aminopropyl moieties were grafted on pristine MSN surface, creating MSN-NH 2 . For the acid moieties, propionitrile groups were grafted onto pristine MSN, which were then hydrolyzed in acidic medium to form propionic acid moieties (MSN-COOH).
Natural compounds, such as polyphenols, can be used as antibacterial and antifungal agents [18]. We selected Salvia officinalis (common sage) for the composites' development as it was reported as a source of phytochemicals with biological activity. Literature data show that application of low-intensity ultrasound during the extraction process usually does not alter the vegetal material and its components, instead decreasing the extraction time, which is correlated with a higher yield in natural compounds compared to the conventional method [43]. For common sage extract, ultrasound-assisted extraction led to better recovery of natural compounds as a result of breaking down plant tissues into the solvent containing the vegetal material. The total polyphenol content (TPC) of the prepared common sage extract was determined using Folin-Ciocalteu reagent and expressed as gallic acid equivalents (163.78 mg GAE /g extract ). The ultrasound-assisted extraction yielded a high TPC value, larger than the previously reported values for the conventional method (138.11 mg GAE /g extract ) [40] or MW-assisted extraction (119.85 mg GAE /g extract ) in absolute ethanol [34]. Furthermore, a high concentration of rosmarinic acid was obtained through this extraction method (25.56 mg/g extract ), along with small quantities of caftaric acid, chlorogenic acid, and caffeic acid. Rosmarinic acid exhibits biological activities, such as antioxidant, anti-inflammatory, and antidiabetic, and also has antimicrobial properties [44].
Entrapment of polyphenolic compounds into mesopores of MSN has been shown to be beneficial in terms of maintaining the radical scavenger activity of an extract [37,45]. For this reason, this strategy was approached with the S. officinalis extract to ensure good stability over time. The results obtained from determination of RSA using ABTS assay indicated slightly better antioxidant activity of the embedded extract in MSN-COOH. For the S@MSN-NH 2 sample, the RSA value is lower than that of the free extract (S), but there are some interferences because of the carrier that seems to interact with ABTS radical.
The antibacterial activity of polyphenols-loaded MSN was tested against two bacterial strains, Staphylococcus aureus and Pseudomonas aeruginosa, and compared to the one of the free extract. The two bacterial strains are the most common pathogens representative of Gram-positive and Gram-negative bacteria classes that are associated with skin infections as a result of injuries, i.e., burns [46,47]. The results illustrated better inhibitory and bactericidal activities for the polyphenols-loaded MSN compared to the free extract. Moreover, there are differences between the two types of bacteria: lower values of MIC and MBC were demonstrated for Staphylococcus aureus, most likely due to the different composition and structure of cell envelope of the bacterial strains.
To further address the aesthetic aspect in long-term healing of skin wounds, a collagen porous scaffold was considered for development of complex composites. Natural protein fibers such as collagen promote cellular regeneration, and, since they are naturally occurring compounds found in the epithelial tissue, they have good biocompatibility [5]. The marinesource collagen used in these composites comes from Rapana venosa, which is a sustainable source of collagen.
The collagen structure consists mainly of glycine, proline, and hydroxyproline, and the chains of the triple helix are connected through glycine bonds. The positions of amide I and II in the FTIR spectra have been shown to be related with the secondary structure of proteins [48]. Thus, the position of the amide I band at 1651 cm −1 indicated the α-helical structure of the collagen. Minor shifts (4-8 cm −1 ) of the main amide bands of collagen in the case of collagen-MSN composites are only due to the overlap of the vibrations characteristic of the functional groups of MSN with those of the amino acids in the protein.
A disruption of the α-helix of the collagen structure would shift the position of the amide I band towards higher frequencies [49], but the position of amide I band of collagen-MSN composites did not reveal such a perturbation on the collagen secondary structure. Amide A and amide B superimposed the broad band of the silanol groups of MSN supports. In the FTIR spectrum of the collagen-MSN-NH 2 composite, the N-H stretching vibration band of MSN-NH 2 also overlapped the amide A and B band, and, therefore, it is difficult to assign its position.
DSC data were recorded to assess the thermal behavior of the collagen porous scaffolds. The endothermic event identified in the 30-100 • C range is attributed to the collagen denaturation temperature. The pristine porous scaffold showed a denaturation temperature of 56.9 • C, while the presence of functionalized MSN led to a higher denaturation temperature, above 70 • C, for both collagen-functionalized MSN composites. Denaturation temperature is strongly associated with proteins structure. The endothermic event is due to transformation of the collagen secondary structure from a triple-helix to a random coil structure [50]. This transformation occurs as result of elimination of water molecules strongly bound through hydrogen bonds, which stabilize the triple-helix structure. The DSC results indicated enhanced thermal stability of the collagen-MSN porous scaffold compared to the pristine one, probably due to the electrostatic interactions between the functional groups of MSN and the amino acids from the collagen structure.
The collagen-polyphenols-loaded silica composites retained the same microstructure as the pristine collagen porous scaffold, a structure with multiple layers and interconnected pores, that showed good distribution of polyphenols-loaded MSN on the sheets of the collagen matrix.
It is very well known that natural products stored in open air are always susceptible to fungal contamination; thus, molds and/or yeasts can easily grow. Prevention of such contamination is a critical aspect considering that the envisioned role of the composites is prevention of bacterial infection in wound healing applications. Both the collagen porous scaffold and its composites containing polyphenols-loaded silica exhibited good sterility, which demonstrated high stability against development of fungi for long-term storage of the composite materials.
Analysis of apoptosis showed that, for the S@MSN-NH 2 sample, a relatively higher percentage of apoptotic cell population was obtained, these results being correlated with that obtained from antiproliferative analysis by MTT assay. To conclude, the encapsulated common sage extract did inhibit tumoral cell growth after 72 h, with an impact on apoptosis induction, and had no effect on the viability of normal keratinocytes cells.

Conclusions
This work reported for the first-time development of collagen-polyphenols-loaded silica composites. The newly designed composites comprised polyphenolic extract with good antibacterial properties, which was embedded into mesopores of functionalized mesoporous silica carriers. Polyphenolic-extract-loaded MSN showed good preservation of radical scavenging activity and enhanced antibacterial activity on both Gram-negative and Gram-positive bacterial strains. The final collagen-polyphenols-loaded silica composites exhibited excellent stability against fungus development. Furthermore, introduction of MSN carriers into the collagen porous scaffold increased the denaturation temperature of the protein, improving the thermal stability of the composites.
The challenges of designing tissue engineering composites consist of reaching an equilibrium between the physicochemical characteristics and biological properties of the composites. The design of collagen-polyphenols-loaded silica composites represents a new approach to use of natural compounds for wound dressings. The advantages of using such composites are: (i) use of renewable sources for their design (polyphenolic extract prepared from S. officinalis plant and collagen scaffold obtained from an invasive snail, R. venosa), (ii) exploring new antibacterial agents as an alternative to typical antibiotics, (iii) high loading capacity of polyphenols and their protection against degradation provided by mesoporous silica carriers, (iv) support in cell regeneration offered by the collagen matrix, (v) good biocompatibility, and (vi) antiproliferative effect on skin carcinoma cells.
These composites that address multiple issues in one application represent a promising perspective on tissue engineering, but further knowledge in this research area is necessary to develop new biocompatible matrices with advanced properties.