Polyethylenimine‐grafted mesoporous silica nanocarriers markedly enhance the bactericidal effect of curcumin against Staphylococcus aureus biofilm

Abstract The recalcitrant nature of biofilms makes biofilm‐associated infections difficult to treat in modern medicine. Biofilms have a high vulnerability to antibiotics and a limited repertoire of antibiotics could act on matured biofilms. This issue has resulted in a gradual paradigm shift in drug discovery and therapy, with anti‐biofilm compounds being sought alongside new drug carriers. A potential solution to biofilm‐associated infections is to employ antibiofilm treatments, which can attack biofilms from many fronts. Nanocarriers are promising in this regard because they can be entrapped within biofilm matrix, target biofilm matrix, and provide local drug delivery to inhibit biofilm formation. In this study, curcumin as an herbal extract was loaded onto hyperbranched polyethylenimine‐grafted mesoporous silica nanoparticles (F‐MSN‐PEI/Cur) and antibiofilm investigations were performed. The F‐MSN‐PEI/Cur design has the potential to repurpose curcumin as an antibiofilm agent by increasing its solubility and lowering the required doses for the destruction of matured biofilms as well as suppressing biofilm development. Using imaging and spectroscopic techniques, we assessed the interaction of F‐MSN‐PEI/Cur with Staphylococcus aureus bacterial cells and determined the impact of F‐MSN‐PEI/Cur on eradicating matured biofilms and suppressing biofilm development. The F‐MSN‐PEI/Cur design is highly cytocompatible, as observed by the cytotoxicity screening investigations on L929 mouse fibroblast cell line. Our findings show that F‐MSN‐PEI/Cur design reduces the bacterial cell viability, inhibits biofilm formation, and induces biofilm eradication, which is attributed to F‐MSN‐PEI/Cur design having the potential to repurpose the antibiofilm activity of curcumin‐herbal extract.


| INTRODUCTION
Infectious diseases caused by bacteria have been life threatening due to increased resistance to antibiotics and the intricacy of the infection.
Disease progression and the response of the host to the infection reveal two different clinical conditions, acute versus chronic infection. 1 Chronic infections are highly associated with aggregated/ attached clusters of planktonic cells, termed the biofilm mode of growth, with a delayed healing process. 2 A biofilm is a structured bacterial community embedded in a three-dimensional matrix that attaches to a solid surface or aggregates into clusters. The self-produced, so-called extracellular polymeric substances (EPS) matrix is responsible for the protection and structural stability of bacteria cells which also results in tolerance to antimicrobial treatments and environmental stress. Antibiotic tolerance in biofilms may be due to (i) EPS components preventing antibiotic diffusion through biofilms, (ii) metabolic activity and growth rate gradients within the biofilm restricting antibiotic uptake by bacteria cells, and (iii) antibiotic degradation by enzymes secreted by biofilm bacteria. 3 Therefore, biofilms persist with antibiotic therapy up to 1000 times more than required for planktonic bacteria treatments. 4 The emergence of biofilmassociated infections has increased the demand for novel approaches for delivering bactericidal drugs through biofilms. One intriguing strategy is to use nanoparticles (NPs) to deliver not only antibiotics but also natural bactericidal agents to handle the issue of biofilmassociated infections, which account for more than 60% of chronic infections in people. 3 With the support of gained knowledge from research in oncology-related nanomedicine, NPs have become attractive for the treatment of infectious diseases. Among existing NP, mesoporous silica nanoparticles (MSN) can provide multifunctionality (i.e., inherently therapeutic, drug delivery, targeted delivery, precision in dosing) due to their modular design options. 5 The recent advancements in NPs aided antibacterial treatments and already well-established research on MSN-aided oncology treatments 6,7 have driven our interest in achieving MSN bactericidal delivery system for the treatment of biofilm-associated infections due to the unique conceptual similarities between biofilms and tumor microenvironments. 5 In comparison to the currently used antibacterial NPs for bacterial growth inhibition, modular MSN-based bactericidal carriers may offer distinct advantages, including entrapment and penetration within the biofilm matrix via attachment to the EPS outer surface followed by migration into the biofilm. Furthermore, silica chemistry can be exploited to provide surface modification on MSN surfaces once for the benefit of high cargo loading, penetration through the biofilm matrix and better interactions with bacterial cells. In this regard, the allocation of primary, secondary, and tertiary amines on MSN surfaces via polyethylenimine (PEI) surface grafting is a legitimate strategy to deliver permeabilizing effects and disintegrate the bacterial cell membranes. 8,9 As a result, MSN-PEI design can function as antibacterial agent carriers while also improving therapeutic efficacy by avoiding antimicrobial agent recognition and deactivation in the biofilm matrix, providing targeted delivery to cells, and increasing local concentration and antibacterial activity of bactericidal agents against bacterial cells. MSN-PEI drug carriers have the potential to improve the efficacy of bactericidal agents by repurposing existing medications and minimizing the development of drug resistance, primarily by lowering the drug concentration required for the treatment and clearance of biofilm infections. 8 Curcumin, the primary component of turmeric, has been shown to possess a variety of pharmacological properties and to exert antibacterial activity when combined with other antibacterial agents. [9][10][11][12][13] Although curcumin's bactericidal property is imposed in combination with other compounds, its use alone is hampered by its low water solubility, stability in physiological environment, and bioavailability. 14 It is classified in Class IV according to Biopharmaceutical Classification System (BCS) and its usage results in low absorption, inadequate distribution, and rapid elimination from living systems. [15][16][17] It was indicated once the curcumin is orally administered (up to 12 g/day) less than 1% of it culminated in human circulation system. 18 Moreover, curcumin has been noted as having very low water solubility of only 0.6 μg/ml and is susceptible to degradation particularly under alkaline conditions. 19 The use of curcumin is limited due to low water solubility under acidic or neutral conditions, high decomposition rate in alkaline media and photodegradation in organic solvent. 20 More in detail, the degradation of curcumin in series of pH conditions ranging from 3 to 10 has been shown and the result revealed that decomposition was pH-dependent and occurred faster at neutral-basic conditions. 21 In vitro studies showed that more than 90% of curcumin tends to degrade in 30 min under physiological pH conditions (0.1 M phosphate buffer solution, 37 C, pH 7.2). 22 Surfactants, polymer mixtures, and the formation of inclusion complexes encompass the most used and successful approaches employed in the literature. 23 Obvious approach to improve the biopharmaceutical properties of curcumin is to overcome water solubility limitations of curcumin by employing nanocarriers. 24 In light of our previous study wherein we have shown the loading of curcumin into MSN possessing different surface chemistries could improve the interaction of curcumin without degradation due to the presence of surface-grown PEI on MSN by providing strong interaction between PEI and curcumin. 25 As such, the use of polyethylenimine surface grafted MSN (MSN-PEI) as a carrier for curcumin is anticipated to be a promising strategy for repurposing curcumin as an antibiofilm agent by improving the water solubility and without providing combinatory antibiotic treatments.
In this study, we interpreted the potency of hyperbranched poly-   In order to solidify F-MSN-PEI nanocarriers-aided curcumin solubility enhancement whether it is due to the obtained differences in the solid state of the curcumin crystals on nanocarriers, differential scanning calorimetry (DSC) investigations were also performed for curcumin powder, F-MSN-PEI and F-MSN-PEI/Cur powders. All samples were scanned by a scanning rate of 5 K/min. An empty aluminum crucible was used as a reference sample and the temperature scale of the calorimeter. A quantity of 2-6 mg (total weight) was used for the DSC measurements.  In detail, S. aureus culture was grown overnight by inoculation of tryptic soy broth (TSB) from a single colony at 37 C and 150 rpm under aerobic conditions. The following day, the overnight grown cultures were diluted at 1:50 in TSB for 2 h to reach the exponential growth phase. Subsequently, the S. aureus culture concentration was adjusted to 10 8  To evaluate the degree of biofilm inhibition and eradication upon exposure to F-MSN-PEI/Cur suspensions and curcumin solutions, biofilm viability, and biomass were quantified using resazurin and safranin assays, respectively. To this end, after the treatments in Section 2.5.2 the supernatants in the wells were gently discarded, and biofilms were washed with PBS thrice to remove loosely attached bacteria. For biofilm viability analysis, 200 μl of 20 μM resazurin solution diluted in PBS was added to each well. The plate was incubated at room temperature and 150 rpm until a color change was observed in the negative control group. Fluorescence measurement was carried out at excitation/emission wavelength of 530/590 nm using CLARIOstar ® multimode reader. For quantification of biomass, safranin dye was employed due to its known ability to bind to negative charges and therefore target many different molecules of bacteria and EPS. 37,38 After the treatments in Section 2.5.2 the supernatants in the wells were gently removed, and the biofilms were washed with PBS thrice.

| In vitro cytotoxicity investigations of F-MSN-PEI/Cur
Thereafter, 200 μl of 0.1% (vol/vol) safranin solution was added to each well, and the plates were incubated at room temperature for where

| Data interpretation and statistical analysis
All experiments were carried out in triplicates and data are represented as mean ± SD. Statistical analyses were performed using Gra- with the following equation: where μ max and μ min are the maximum and minimum mean of assay signals, and σ max and σ min are the SDs of the maximum and minimum T A B L E 1 Interpretation of biofilm production.

Average OD Biofilm production
OD ≤ OD c Non-biofilm producer OD c < OD ≤ 2xOD c Weak biofilm producer 2xOD c < OD ≤ 4xOD c Moderate biofilm producer 4xOD c < OD Strong biofilm producer assay signals. The test method is interpreted as appropriate if the Z' factor is between 0.5 and 1, acceptable in some cases if it is between 0 and 0.5, and not acceptable if the Z' factor value is <0.

| Characterization of the F-MSN-PEI/Cur
The synthesis protocol yielded spherically shaped monodisperse F-MSN-PEI and F-MSN-PEI/Cur in the size range of 150-250 nm ( Figure 1A,B). 250 μg/ml F-MSN-PEI and F-MSN-PEI/Cur NP suspensions were dispersed in HEPES buffer (pH 7.2, 25 mM) for hydrodynamic size and net surface charge (ζ-potential) measurements, respectively ( Figure 1C). The hydrodynamic size and ζ -potential data are presented in Table 2. The F-MSN-PEI samples were fully dispersible in HEPES buffer after all processing steps, including template removal, surface modification of F-MSN with polyethylenimine (F-MSN-PEI) and further curcumin loading (F-MSN-PEI/Cur). The success of surface modification was ensured by the ζ-potential measurements as presented in Table 2. The re-dispersibility of F-MSN-PEI/ Cur was investigated. The results obtained (Table 2)  can be ascribed to a physical interaction between the excipients during processing. Almost 5-fold ΔH reduction was also observed as given in Figure S2 for F-MSN-PEI/Cur compared to curcumin powder.
This could be attributed to the disorganization of curcumin crystals after loading into F-MSN-PEI nanocarriers. 44 The slightly lower T m could be ascribed for this, while the ΔT 1/2 remains unchanged.

| In vitro cytotoxicity of F-MSN-PEI/Cur
The The Z 0 factor, which predict the robustness of an assay, was calculated for the optical density reading assays used for bacterial growth profiles, safranin assay for biomass determination and resazurin assay for cell viability to affirm the separation between the distributions of positive and negative controls. The calculations resulted in the Z' factor between 0.1 and 0.6 for safranin assays, which is indicated as a moderate assay. However, the Z factor was calculated as 0.94 for optical density investigations and 0.84 for the resazurin assay, indicating the robustness and reproducibility of the assays. 45 The biofilm forming capacity of the treated bacterial cells with ascending dosing of F-MSN-PEI/Cur were found that the treatment conditions completely eliminate the biofilm forming ability of S. aureus biofilms as presented in Table 3. 39 SEM images of S. aureus biofilms after exposure to the highest dose of F-MSN-PEI/Cur (1600 μg/ml) for both inhibition and eradication investigations and negative control group were presented in Figure 8 to predict the architectural changes of the biofilm matrix after treatments. Reduction in the biomass for both cases was F I G U R E 6 Investigation of the biofilm inhibiting potency of F-MSN-PEI/Cur and equivalent curcumin treatment by (A) safranin assay for total biomass and (B) resazurin assay for bacterial cell viability in biofilm. Statistical analysis was performed between treatment groups by applying two-way ANOVA followed by Sidak's multiple comparison test, ns: 0.1234, *p < .03, **p < .002, ***p < .0002, ****p < .0001.
obtained compared to untreated condition. In addition, the debris of the biofilm matrix could be also visualized in Figure 8C whereas proper bacterium structures without any destruction could be visualized in Figure 8B. is therefore of great concern to obtain the desired anti-biofilm effects since higher doses of the active ingredients is required against biofilms compared to planktonic bacteria.
In addition, the F-MSN-PEI carrier could aid in improving the biocompatibility of curcumin as a bactericidal agent. It was revealed that curcumin show distinct toxicity that is concentration dependent. 49,50 The incorporation of curcumin into F-MSN-PEI carriers did not cause any cytotoxicity and pungency, which could be due to both the com- design, PEI is well known for its cytotoxicity. Despite the toxicity consideration PEI functionalization is utilized commonly especially for gene delivery due to capability to employ reactive groups, increase stability, tune surface charge. It was indicated that PEI toxicity could be attributed to be molecular weight dependent. 55 However, Desai and coworkers demonstrated that PEI adsorbed silica NPs in different forms showed less toxicity than PEI alone even at high concentrations. Also, toxicity did not correspond to the particles concentrations but PEI/SiO 2 ratio. The study proved that incorporation of silica and PEI could aid to suppress the toxicity of PEI. 56 All in all, the in vitro cytotoxicity investigations did not lead reduction in cell viability especially for 24 h incubation which is also matching with the treatment period in the bactericidal investigations.
F-MSN-PEI drug carrier could aid in improving the bactericidal effect of curcumin. Due to the spatial confinement of drug molecules within the F-MSN-PEI mesoporous network, drugs with poor solubility can be transformed into their amorphous/disordered crystalline counterparts with higher solubility. 57 Also, it was shown that aminegrafted mesoporous silica employed a prolonged and adjustable release profile of curcumin. 58 In addition, the surface chemistry of the NPs has been shown to play a prominent role in altering the interaction between NPs and the biofilm matrix. This fact eventually determines the penetration ability, depth, and distribution of the NPs through the biofilm matrix. 59 63 In addition, the porous architecture of the silica matrix has been revealed to prevent the bacterial cell aggregations that take place as the initial step of biofilm formation, as also claimed in the literature studies. 64 On the other hand, free curcumin treatment of bacterial cells during biofilm formation steps has induced the biomass formation compared to control groups, which could be due to the secretion of different components and could cause relapsing of the biofilm formation in the long term. When the short-term treatment of S. aureus cells is taken into account, the viability of bacterial cells could already be reduced by 20% with 400 μg/ml F-MSN-PEI/Cur treatment compared to the negative control group, which could be depicted as that changes in the metabolic activity of bacterial cells could inhibit the biofilm formation potency of treated bacterial cells.
As presented in Table 3, F-MSN-PEI/Cur could play a role in inhibiting the biofilm-forming capacity of the bacterial cells by switching them to non-biofilm-forming bacterial cells. However, there was only an acceptable level of statistical difference in the reduction of biofilm viability in the matured biofilms in Figure 7B, whereas the statistical differences were significantly higher in Figure 6B

DATA AVAILABILITY STATEMENT
Data available on request from the authors. The data that support the findings of this study are available from the corresponding author upon reasonable request.