Chitosan capping enzyme-responsive hollow mesoporous silica nanoplatforms for colon specific drug delivery

In summary, an enzyme-responsive colon specific drug delivery system was developed based on hollow mesoporous silica sphere (HMSS), in which the biodegradable chitosan (CS) was gated on the openings of HMSS through cleavable azo bonds (HMSS-N=N-CS). Doxorubicin (DOX) was encapsulated into the hollow cavity and mesopores of HMSS, and HMSS-N=N-CS/DOX showed a high loading amount of 35.2%. X-ray diffraction (XRD) experiment proved that the DOX loaded in the HMSS-N=N-CS was in a non–crystalline state. In vitro drug release experiments proved that HMSS-N=N-CS/DOX showed an enzyme-responsive drug release property. The grafted CS could increase the biocompatibility and stability of HMSS, and reduce the protein adsorption on the surface of HMSS. The gastrointestinal mucosa irritation and cell cytotoxicity results indicated the good biocompatibility of HMSS and HMSS-N=N-CS. The confocal laser scanning microscope (CLSM) and flow cytometry technique (FCM) results indicated that the cellular uptake of DOX was obviously increased after the HMSS-N=N-CS/DOX was preincubated with colonic enzyme mixture. Cell viability result indicated that HMSS-N=N-CS/DOX incubated with colon enzyme showed an increased cytotoxicity and the IC50 value was three time less than that of HMSS-N=N-CS/DOX group. The present work will lay the foundation for subsequent research on mesoporous carriers for oral colon-specific drug delivery. <o:p></o:p> TGA analysis on a a heating rate of 10 °C/min a nitrogen Power XRD was performed on a D5005 X-ray diffractometer with Cu-Kα cell medium was substituted by a serum–free medium containing different concentrations of nanoparticles. After incubation for 2 d, 50 µL of MTT solution (2 mg mL –1 ) was added and incubated for 4 h to measure the living cells. Then, MTT solution was removed and 150 μL DMSO was added to dissolve formazan. Subsequently, the absorbance was measured on a microplate reader (Tecan, Männedorf, Switzerland) at 570 nm. The cytotoxicity of free DOX, HMSS-N=N-CS/DOX and HMSS-N=N-CS/DOX pre-incubated with enzyme mixtures extracted from colonic microflora was measured using Caco-2 cells with the corresponding DOX concentrations of (0.1, 1, 5, 10 and 20 μg/mL). The incubation time was 48 h, and the other experiment processes were the same as above described.

Introduction media containing enzyme mixture was filtered through a 0.22 μm filter to remove all the cellular debris from the culture fluid. Subsequently, the filtrate was lyophilized to obtain the enzyme mixture in the powder form, which was used in the further study.
Drug loading process and enzyme-responsive release 25 mg DOX was dissolved in 5 mL pH 3.5 HCl solution. And 100 mg HMSS-N=N-CS was added in the DOX solution, and stirred at ambient temperature for 12 h. Subsequently, 0.2 M NaOH solution was used to adjust the pH of mixture to 7.0, and the suspension was stirred for another 12 h. Then, the DOX-loaded HMSS-N=N-CS (referred to HMSS-N=N-CS/DOX) was centrifuged and washed to remove the adsorbed DOX on the surface of HMSS-N=N-CS. The supernatant was gathered at each step to measure the DOX loading efficiency at 480 nm by UV-Vis spectrophotometry. The total mass of DOX loaded in HMSS-N=N-CS was calculated by subtracting the unloaded DOX after drug loading processes from the initial mass of DOX added.

BSA adsorption
The BSA adsorption amount was evaluated based on the published works [29,30]. BSA was added in pH 7.4 PBS (0.5 mg/mL). 5 mg HMSS and HMSS-NH 2 and HMSS-N=N-CS were added into 2.5 mL PBS (pH 7.4). And the equal volume BSA solution was supplied, and the suspension was placed in a shaker at 100 rpm. After 6 h, centrifugation was used to collect the upper solution. At last, the BSA concentration was measured at 595 nm after being stained with Coomassie brilliant blue solution.

Characterization
The mesoporous network structure and morphology of the HMSS nanoparticles were evaluated by TEM images (EM-208S, CSIS, USA). The surface area and pore size distribution of nanoparticles were characterized using nitrogen adsorption analysis analyzer (V-Sorb 2800P, Gold APP Instrument Corporation, China). The ξ potentials and particle sizes were characterized on a Nano-z90 Nanosizer (Malvern Instruments Ltd., Worcestershire, UK). TGA analysis was measured on a TGA-50 equipment (Shimadzu, Kyoto, Japan) with a heating rate of 10 °C/min under a nitrogen flow. Power XRD was performed on a Siemens D5005 X-ray diffractometer (Karlsruhe, Germany) with Cu-Kα radiation (λ = 1.5418 Å).

Cell culture and cell uptake experiment
Caco-2 cells were cultured in a medium supplemented with 10% FBS, 1% nonessential amino-acid, 1% (v/v) pyruvic acid sodium and 1% streptomycin. NIH-3T3 cells were cultured in DMEM with 1% streptomycin and 10% FBS. The Caco-2 cells uptake of the nanocarriers was characterized using FCM and CLSM. Caco-2 cells were seeded into 24-well plates. After culturing for overnight, free DOX, HMSS-N=N-CS/DOX and HMSS-N=N-CS/DOX pre-incubated with colonic enzyme nanoparticles (equal to the concentration of 5 μg/mL DOX) were added to corresponding wells. After continued incubation for 2 h, the cell medium was removed and washed thoroughly with PBS. Then, the cells were fixed by 4% formaldehyde and stained by Hoechst 33258 for CLSM observation. FCM was used to obtain a quantitative evaluation of cellular uptake. Caco-2 cells were seeded in 6-well plates and further incubated for 24 h. After washing with PBS, the Caco-2 cells were incubated with free DOX, HMSS-N=N-CS/DOX and HMSS-N=N-CS/DOX pre-incubated with colonic enzyme nanoparticles (equal to the concentration of 5 μg/mL DOX) in serum-free DMEM for 2 h. Then, the Caco-2 cells were rinsed with cold PBS, trypsinized and re-suspended in 0.5 mL PBS. The DOX fluorescence in cells was measured using a FACSCanto flow cytometer (Becton, Dickinson, USA).

In vitro cellular proliferation assay
The cytotoxicity of HMSS and HMSS-N=N-CS blank carriers towards NIH-3T3 and Caco-2 cells were testified by MTT assay [31,32]. Briefly, Caco-2 cells and NIH-3T3 cells were separately seeded in 96well plates and further incubated for overnight. The old cell medium was substituted by a serum-free medium containing different concentrations of nanoparticles. After incubation for 2 d, 50 µL of MTT solution (2 mg mL -1 ) was added and incubated for 4 h to measure the living cells. Then, MTT solution was removed and 150 μL DMSO was added to dissolve formazan. Subsequently, the absorbance was measured on a microplate reader (Tecan, Männedorf, Switzerland) at 570 nm. The cytotoxicity of free DOX, HMSS-N=N-CS/DOX and HMSS-N=N-CS/DOX pre-incubated with enzyme mixtures extracted from colonic microflora was measured using Caco-2 cells with the corresponding DOX concentrations of (0.1, 1, 5, 10 and 20 μg/mL). The incubation time was 48 h, and the other experiment processes were the same as above described.

Toxicity studies
The gastrointestinal mucosa irritation tests are vital for the evaluation of oral drug delivery in vivo biosafety. Male Sprague-Dawley rats (180 ± 10 g) were randomly divided into three groups (three rats for each group). Rats were administrated saline, HMSS and HMSS-N=N-CS nanoparticles with a dose of 100 mg/kg for each day. After 7 d, all the rats were sacrificed, and the tissues were collected and examined by histopathological examination (H&E). To evaluate the biosecurity of HMSS and HMSS-N=N-CS nanoparticles, the body weights of BALB/c mice (18-20 g) were recorded after oral administration at a dose of 100 mg/kg for every other day.

Results And Discussion Preparation and characterization of HMSS-N=N-CS
The HMSS was prepared based on previous works with minor changes [26]. Firstly, the solid SiO 2 nano-spheres were prepared, and the mesoporous shell was coated on the surface of the solid silica nanospheres containing CTAB template. Then, the Na 2 CO 3 was used to selectively etch the solid SiO 2 nano-spheres, while the mesoporous shell was protected by template CTAB. The prepared process of HMSS-N=N-CS with CS as a "gatekeeper" by the azo linkage is described in Fig. 1A. First, the surface of HMSS was modified with APTES as an alkyl coupling reagent to become amino-functionalized HMSS (HMSS-NH 2 ) by the post-modified method. Subsequently, the HMSS-N=N-COOH was prepared by an amidation reaction between the amino groups of HMSS-NH 2 and the carboxyl groups of azobenzene-3,3'-dicarboxylic acid. Then, CS was covalently modified onto the surface of HMSS nanoparticles by an amidation reaction between the carboxyl groups of HMSS-N=N-COOH and the amino groups in CS.
As displayed in the transmission electron microscopy (TEM) image in Fig. 1A, the average diameter of HMSS was 250 nm with a uniform hollow structure and highly ordered mesoporous shell. The average mesoporous shell thickness was ca. 90 nm. Compared to the smooth surface of HMSS, the surface of grafted polymer HMSS-N=N-CS ( Fig. 1B) was rough, indicating that the CS was covered on the HMSS carrier.
The surface areas and the pore distributions of mesoporous materials played a crucial role in loading and delivering host molecules for controlled release. The pore size distribution curves and isotherms were measured by N 2 adsorption and desorption analysis (Fig. 2). The detailed parameters of Brunauer-Emmett-Teller (BET) surface area (S BET ), total pore volume (V P ) and pore size distribution (D P ) are displaced in Table 1. The S BET and V P of pure HMSS were 810.7 m 2 /g and 0.969 cm 3 /g, respectively, and the D P was about 3.8 nm. The D P of HMSS-NH 2 was almost the same as that of HMSS after amination, indicating that the mesopores were not blocked after amino-functionalization. While S BET and V P of HMSS-N=N-CS were markedly decreased after modification of azo compound and CS coating, indicating that CS had coated on the surface of the HMSS. The cumulative release of DOX was improved to more than 20% within the same period. Additionally, the release amount of DOX was dramatically increased to nearly 40% in presence of concentrated enzyme. The relatively low drug release percentage was owing to the electrostatic interaction between the negatively charged HMSS and the positively charged DOX [34]. The above results proved that the release of DOX from HMSS-N=N-CS/DOX was markedly accelerated by extracted enzymes from microflora in colonic regions. And the enzyme-responsive release mechanism could be attributed to the degradation of azo bonds in HMSS-N=N-CS by enzyme, causing the detachment of CS from the surface of HMSS and the fast release from HMSS. Since azo bonds have been reported to be cleaved by enzymes secreted by colonic microflora [35,36].
Additionally, to further evaluate the enzyme-responsive release from HMSS-N=N-CS/DOX in the mimetic GIT environment, the HMSS-N=N-CS/DOX nanoplatforms were initially dispersed in SGF for 2 h, and then further dispersed in SIF for 6 h, and finally the carriers were added in pH7.4 PBS containing 1mg/mL extracted enzyme. As shown in Fig. 5B, in simulated gastric juice, the release of DOX showed a relatively fast release rate and the cumulative amount was up to 15% within 2 h. The relatively fast release was due to the weaker interaction between HMSS-N=N-CS and DOX in the acid conditions [1]. Then, the release rate of DOX was slowed down in SIF for 2-8 h. However, after the HMSS-N=N-CS/DOX was incubated with extracted enzymes in pH7.4 PBS, the release of DOX was continued to increase markedly, and the cumulative release amount was more than 50% within 24 h.
The incomplete DOX release form HMSS-N=N-CS/DOX was due to the strong interaction between the positively charged DOX and negatively charged HMSS.

The protein adsorption and stability of HMSS-N=N-CS
For oral administration, the surface properties of nanocarriers will unavoidably affect the drug release behaviors and bioadsorption [37]. The protein adsorption assay on the surface was used to evaluate the effect of grafted CS on the surface of HMSS. As displayed in Fig. 6A, bare HMSS nanocarriers had a dramatic BSA adsorbance up to 16.5% attributed to the large surface area and hollow cavity of HMSS, strong adsorption ability and nonspecific interactions between silanol groups of HMSS and BSA [29,30]. In addition, HMSS-NH 2 similarly had a relatively high adsorbed BSA amount

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. Scheme 1.docx