In Vitro Antimicrobial Studies of Mesoporous Silica Nanoparticles Comprising Anionic Ciprofloxacin Ionic Liquids and Organic Salts

The combination of active pharmaceutical ingredients in the form of ionic liquids or organic salts (API-OSILs) with mesoporous silica nanoparticles (MSNs) as drug carriers can provide a useful tool in enhancing the capabilities of current antibiotics, especially against resistant strains of bacteria. In this publication, the preparation of a set of three nanomaterials based on the modification of a MSN surface with cholinium ([MSN-Chol][Cip]), 1-methylimidazolium ([MSN-1-MiM][Cip]) and 3-picolinium ([MSN-3-Pic][Cip]) ionic liquids coupled with anionic ciprofloxacin have been reported. All ionic liquids and functionalized nanomaterials were prepared through sustainable protocols, using microwave-assisted heating as an alternative to conventional methods. All materials were characterized through FTIR, solution 1H NMR, elemental analysis, XRD and N2 adsorption at 77 K. The prepared materials showed no in vitro cytotoxicity in fibroblasts viability assays. The minimum inhibitory concentration (MIC) for all materials was tested against Gram-negative K. pneumoniae and Gram-positive Enterococcus spp., both with resistant and sensitive strains. All sets of nanomaterials containing the anionic antibiotic outperformed free ciprofloxacin against resistant and sensitive forms of K. pneumoniae, with the prominent case of [MSN-Chol][Cip] suggesting a tenfold decrease in the MIC against sensitive strains. Against resistant K. pneumoniae, a five-fold decrease in the MIC was observed for all sets of nanomaterials compared with neutral ciprofloxacin. Against Enterococcus spp., only [MSN-1-MiM][Cip] was able to demonstrate a slight improvement over the free antibiotic.


Introduction
Throughout recent decades, an exponential increase in multidrug resistant bacterial infections has been observed, which has hampered the efficacy and proficiency of current antibiotics and therapies. According to a Centers for Disease Control and Prevention (CDC)

Description of the Synthesis
Pristine mesoporous silica nanoparticles (MSNs-OH) were prepared following the procedure described by Bouchoucha et al. [67] with the minor modifications reported previously [66]. Triethoxysilane Cholinium, 1-Methylimidazolium and 3-Picolinium-based ionic liquid derivatives were prepared by microwave irradiation, as described below. MSNs-OH were then functionalized with these ionic liquids, also using microwave-assisted procedures.       Figure 3 for proposed numbering a 3-(chloropropyl)triethoxysilane (8 mL, 33.3 mmol, 1.08 equivalents) and 3-picoline (3 mL, 30.8 mmol) were added to a wide-neck glass vial. The reaction mixture was heated under microwave irradiation for 60 min at 120 • C until the formation of a single-phased dark orange and viscous oil. The crude was then washed with diethyl ether and dried under vacuum to give 9.68 g (91%) of the pure final product designated as Si-   Figure 3 for proposed numbering as   These materials were prepared by the following general procedure: calcined MSN-OH (0.50 g) was heated at 150 • C under reduced pressure for 2 h to remove the physiosorbed water. After cooling down to room temperature, an excess of the ionic liquid derivative (0.40 g) in toluene (8 mL), or in a mixture of ethanol and acetonitrile in the case of Si- [Chol][Cl] (due to its solubility), was added and the mixture was stirred under microwave irradiation at 100 • C for 1 h. The solid was separated by centrifugation (5000 rpm, 15 min) and washed with ethanol four times. The resultant material was dried overnight at 80 • C.
MSN- [Chol][Cl]: The data characterization of this material is in accordance with that previously described [66].

Antimicrobial Studies
The antimicrobial activity of the materials was tested against strains of sensitive and resistant Klebsiella pneumoniae (HS31 and HS16, respectively) as well as sensitive and resistant strains of Enterococcus spp. (FC89-8R e FC89-10R, respectively). K pneumoniae samples come from nosocomial infections, while Enterococcus spp. originate from infections of free-range chickens. The antimicrobial susceptibilities for all isolates were determined using the Kirby-Bauer disk diffusion method in accordance with EUCAST standards (2022) [68]. Briefly, the bacteria were seeded on BHI agar and incubated at 37 • C for 24 h to promote their growth. Then, using a swab, 1 to 2 colonies were picked and diluted in 3 mL of sterile saline solution (0.9% NaCl) until reaching a concentration equivalent to 0.5 on the McFarland scale. Subsequently, a homogeneous smear was made on an MH agar plate (Mueller-Hinton agar). Finally, antibiotic disks were properly spaced on the plate using forceps. The plates were then stored in an incubator at 37 • C for 24 h. After this time, an inhibition halo was observed around each disk, resulting from the effect that the antibiotic had on the bacteria under study. The diameters (mm) of the halos were measured in order to compare the results with regulatory standards and determine the resistance exhibited by the bacteria to each antibiotic.
A total of 13 antibiotics were tested for Enterococcus spp.  Table S2.
We utilized the microdilution technique to ascertain the minimum inhibitory concentration (MIC) of ciprofloxacin. Initially, each strain was introduced onto Brain Heart Infusion agar and allowed to incubate at 37 • C for 24 h. Subsequently, the bacterial cells were subcultured in tubes containing Mueller-Hinton broth and subjected to overnight incubation at 37 • C with 150 rpm using the ES-20/60 Orbital Shaker-Incubator from Biosan in Riga, Latvia. The overnight culture was then diluted in fresh MH broth to attain a turbidity standard of 0.5 McFarland, employing a spectrophotometer. Next, ciprofloxacin was diluted in sterilized distilled water, generating various concentrations ranging from 0 µg/mL to 8182 µg/mL. To assess bacterial growth, 75 µL of each ciprofloxacin concentration was dispensed into a polystyrene flat-bottom 96-well plate. Simultaneously, 75 µL of the bacterial suspension was placed in the same 96-well microtiter plate. The plates were subsequently incubated for 24 h at 37 • C with 150 rpm. Bacterial growth was quantified at 490 nm using a microplate reader, specifically the BioTek ELx808U from BioTek in Winooski, VT, USA. The MIC was determined as the lowest concentration of ciprofloxacin that effectively impeded bacterial growth. Triplicates were performed for every different concentration tested on each nanomaterial. simplicity. All experiments undertaken at 37 • C, 5% (v/v) CO 2 and 99% (v/v) humidity in the dark.

Antiproliferative Studies
Cells were seeded in a 96-well plate at a density of 7500 cells/well, and after 24 h were submitted to a concentration range between 0.1-50 µM of MSN-API-OSILs. After 48 h incubation, cellular viability was inferred with the Cell Titer 96 ® Aqueous One solution cell proliferation assay (Promega, Madison, WI, USA) according to the manufacturer's instructions and procedures previously described [69]. The absorbance at 490 nm of the produced Formazan was quantified with a Tecan microplate reader, Infinite M200 (Tecan, Mannedorf, Switzerland), and the analysis of the dose-response curves to determine the relative IC 50 was performed with GraphPad Prism 8 software (GraphPad Software, La Jolla, CA, USA). The cytotoxicity studies were conducted in triplicate for all compounds, and each concentration tested in duplicate.

Synthesis and Characterization
Our group has recently reported the functionalization of pristine mesoporous silica nanoparticles (MSNs) with a covalently linked choline derivative cation and ciprofloxacin as anion [66]. In this work, the effect of the cation functionalization, such as choline (for comparison purposes), methylimidazolium and picolinium derivatives, on the antimicrobial activity of ciprofloxacin (Cip) is evaluated. More sustainable synthetic methodology based on microwave heating for the preparation of the cation's precursors as well as their functionalization on MSNs enabled faster reactions, higher yields and less solvent required.       The N2 adsorption-desorption isotherms of nanomaterials and corresponding por size distributions ( Figure S3) present features as in previous work, namely those of th The loading of functionalized MSN materials was determined through elemental analysis, and later the results were compared with the NMR information. The values of the loading obtained through elemental analysis are summarized in Table 1. For [MSN-Chol]Cl, the C/N molar ratio obtained was 8.6 (the theoretical value for all ethoxy groups bounded to the surface is 7), which suggests that the anchorage occurs mainly through two ethoxy groups; the content of choline derivatives moiety was 1.0 mmol g −1 based on nitrogen value, a higher content than the one obtained for the analogous material described previously (0.73 mmol g −1 [66]). For materials [MSN-1-MiM]Cl and [MSN-3-Pic]Cl the C/N ratios were 3.9 (theoretical = 3.5) and 9.6 (theoretical = 8), respectively, indicating that the picolinium cation derivative is bounded to the surface also mainly through two ethoxy groups. In relation to the organic cations content, a higher value for the picolinium derivative (1.7 mmol g −1 ) was obtained in comparison with the choline (1.0 mmol g −1 ) and imidazolium derivatives (1.1 mmol g −1 ). Considering these loadings of organic cations, the C/N molar ratios and N content in the final hybrid materials, it is possible to estimate the content of Cip as counter-ion in the following materials:  (Table 1). These results mean that around 79% of picolinium cation derivatives contain Cip as counter-ion, while for choline and imidazolium cations this value is around 60%. The loadings of anchored guests determined through NMR spectroscopy were lower than those obtained through elemental analysis: 0.78 mmol g −1 for [MSN-Chol]Cl, 0.92 mmol g −1 for [MSN-1-MiM]Cl and 1.51 mmol g −1 for [MSN-3-Pic]Cl. This difference can be justified by experimental error of the procedure. However, in the final materials, comparing the ratio between the integration of the ciprofloxacin moiety protons and cation protons grafted on the surface, it is possible to confirm that the correlation approaches the same proportion already denoted through elemental analysis. Table 1. Values for the loading of organic content of the functionalized nanomaterials obtained by elemental analysis. In the case of the materials functionalized with ciprofloxacin, the value refers to the amount of ciprofloxacin expected to be present in the material.

Material
Loading (mmol g −1 ) [ The N 2 adsorption-desorption isotherms of nanomaterials and corresponding pore size distributions ( Figure S3) present features as in previous work, namely those of the pristine MSN are typical of mesoporous nanomaterials with reasonably uniform mesopore diameter inside small nanoparticles. The results of the analysis of the N 2 adsorption/desorption isotherms by the BET method, using criteria recommended by IUPAC [71] and by NLDFT using the Quantachrome software ASiQwin (Table 2), are in agreement with the higher organic loading obtained for [MSN-3-Pic]Cl, since this material revealed more pronounced changes in the textural properties than [MSN-Chol]Cl and [MSN-1-MiM]Cl. In the case of [MSN-Chol]Cl, the changes are more marked than in our previous work [66] where the final content of the choline derivative cation was lower. In all cases, there is an additional decrease of A BET with the ionic exchange compared with their precursors. Table 2. Values of A BET , D p and V p for the nanomaterials 1 .

Material
A BET (m 2 g −1 ) D p (nm) V p (cm 3 g −1 ) Pristine MSN (Chol) The XRD diffraction patterns for the precursor and for the functionalized nanomaterials are shown in Figure S4, and are typical of mesoporous silica nanoparticles, with the main peak broader and less intense than those obtained for bigger particles and with comparable average pore diameters, reflecting the usual lesser ordering of the mesopores inside the small nanoparticles.

Antimicrobial and Cytotoxicity Studies
The minimum inhibitory concentration of the nanomaterials was determined against Gram-positive Enterococcus spp. and Gram-negative Klebsiella pneumoniae. To determine the lowest concentration that inhibits bacterial growth, a range of solutions of different concentrations of the nanomaterials were incubated with the bacterial suspensions. The assays started at the lowest concentration of the nanomaterials, following each subsequent assay with a solution twice as concentrated as the one used in the previous assay. Therefore, the MIC was determined as the lowest concentration found for each active nanomaterial that effectively impeded bacterial growth. Upon finding such concentration, the loading of anchored guests and anionic ciprofloxacin in each individual set of nanomaterials (Table 1) is taken into account. As such, the value for the MIC is normalized regarding the amount of anionic ciprofloxacin estimated in each respective set of nanomaterials. The results are presented in Figures 6 and 7, with all data being compared to ciprofloxacin, used as positive control. ]Cl-were also tested against sensitive and resistant strains of K. pneumoniae and Enterococcus spp. and no noticeable effect was detected on the bacterial growth. On the upper limit, concentrations of up to 1 mg/mL were tested and no inhibition on bacterial growth was observed, which is consistent with the fact that only the addition of the antibiotic to the final material showed antibacterial activities.
The values of the minimum inhibitory concentrations (MIC, in µM) for the most active compounds are displayed in Table 3. The results point towards a correlation between the presence of the nanoparticle and the increase in the potency of the antibiotic, even if the improvement in results could not be fully extended to the tests against Gram-positive bacteria. This latter result can be somewhat correlated to the larger and more complex bacterial wall present in Grampositive bacteria, which might hinder the permeability of the antibiotic as well as making it more difficult for nanoparticle system to interact with the microorganism. However, due to the nature of the antibacterial studies performed, we are not able to make a thorough assessment on the overall effect of the different organic cations grafted on the surface. Simultaneously, because the MICs were calculated by doubling the concentration of the active materials in every subsequent assay, more detailed studies are needed to reach the most precise measurement of the MICs of each compound. Nonetheless, because in this study we could not determine the lowest possible concentration of inhibition, we postulate that more detailed studies-spawning shorter ranges of concentrations-should only improve the results, revealing an even lower concentration needed to hamper the bacterial growth.
Cytotoxicity assays were performed using the three prepared mesoporous silica nanoparticles loaded with the antibiotic, as well as the precursor materials and free ciprofloxacin. The selection of human primary dermal fibroblasts is considered a good surrogate model test system for human healthy cells. In general, none of the compounds displayed any in vitro cytotoxicity at high concentrations (50 µM). The low cytotoxicity at such high concentration demonstrates that this approach seems to be a good nanoplatform for future pharmaceutical applications.

Conclusions
In this work, a protocol for the synthesis of ionic liquids using microwave-assisted heating was employed. In comparison with conventional organic synthesis procedureswhich typically require several hours at reflux temperatures-the present work successfully provides a greener alternative, with reliable results, satisfactory yields, faster reactions and easier purifications. The obtained ionic liquids were grafted on the surface of mesoporous silica nanoparticles to functionalize the surface with organic cations, combining with ciprofloxacin as anion. The materials were extensively characterized through various techniques, including FTIR, solution 1 H NMR, elemental analysis, XRD and N 2 adsorption at 77 K to prove the desired structures as well as important parameters.
All prepared compounds are non-toxic according to the results of in vitro cytotoxicity using human fibroblasts.
The antimicrobial activities were determined against Gram-negative K. pneumoniae and Gram-positive Enterococcus spp. both with resistant and sensitive strains. Against the Gramnegative bacteria, all sets of nanomaterials were able to outperform free ciprofloxacin, with significant reduction in the minimum inhibitory concentration. The material [MSN-Chol][Cip] showed a tenfold increase in the antibacterial activity against sensitive K. pneumoniae, compared to free ciprofloxacin. For the case of resistant Gram-negative strains of K. pneumonia, all sets of nanomaterials showed a five-fold decrease in the concentration needed to hamper bacterial growth. Against resistant and sensitive Enterococcus spp. strains, only the material [MSN-1-MiM][Cip] was able to demonstrate a slight improvement over free ciprofloxacin. Further studies should be conducted to evaluate the MICs with more precision, as well as any possible correlation between the choice of IL/antibiotic, which might allow the tuning of specific properties assisting the development of more potent formulations and directed therapies.  Table S1. Phenotypical profile of K. pneumoniae strains of bacteria and their respective resistance profile obtained. S-means sensitivity to the antibiotic; R-means resistance to the antibiotic; Table S2. Phenotypical profile of Enterococcus spp. strains of bacteria and their respective resistance profile. S-means sensitivity to the antibiotic; R-means resistance to the antibiotic.