Helicobacter pylori is a microaerophilic, Gram-negative, spiral-shaped bacterium, capable of colonizing the acid environment existing in the stomach and taking on a virulent profile capable of triggering injuries in the digestive system [1], [2]. Its helical shape helps the movement driven by the flagella, crossing the protection of the gastric mucosa and escaping from the low pH in the stomach lumen (pH ∼ 2). Thus, associated with the release of cytokines and the chronic inflammatory process, it can evolve into serious pathologies, such as chronic active gastritis, peptic ulcer, gastric atrophy, mucosa-associated lymphoid tissue (MALT), and gastric adenocarcinoma. [3].

Several mechanisms of virulence have been identified, including the production of urease, the gene products of pathogenicity (cagPAI – cag pathogenicity island), and the vacuolizing cytotoxin (vac A) [4]. The adhesion profile on surfaces, or combined with the biofilm formation, are also very important virulence factors that make drug therapy difficult.

Antibiotics have been extensively used in the eradication of H. pylori, however, its resistance to these drugs and other obstacles, such as low aqueous solubility, stability problems in the gastrointestinal environment, and difficulty in controlling the release of the drug or its absorption, are factors that make drug therapy for this bacteria very tough [5], [6], [7].

Thus, it is worth searching for new antibiotic-free compounds with anti-Helicobacter pylori action. Trans-resveratrol (3,4,5-trihydroxystilbene) (RESV) is a flavonoid polyphenol found predominantly in natural edible products such as grapes (Vitis spp), peanuts (Arachis spp), blueberries and cranberries (Vaccinium spp) [8]. In the literature, it was observed that RESV can inhibit the growth and proliferation of different H. pylori strains [8]. Zhang et al. (2015) demonstrated that RESV inhibits the growth of H. pylori and has a significant effect against oxidative stress, with a potent anti-inflammatory action in mice gastric mucosa infected with H. pylori [9].

Despite its promising antibacterial activity reported in the literature, RESV low water solubility and photosensitivity hinder its use in clinical research [10], [11]. To overcome the limitations of RESV, this compound was encapsulated onto chitosan (CS) nanoparticles (NP) to improve RESV bioavailability and provide efficient action at the target site (near the gastric mucosa, where the H. pylori stands) [12], [13].

Indeed, one of the most efficient ways to treat gastric diseases is through the use of polymeric NP that establish mucoadhesive interactions with the stomach mucosa, which can prolong the contact time between the drug and the absorbing tissues, resulting in a high concentration of the drug in the site of action and a high drug flux through the absorbing tissues. Among the most explored polymers for the build of mucoadhesive NP, CS stands out. CS is a natural cationic polymer that has been extensively studied due to its biodegradability, non-toxicity, biocompatibility, and antibacterial properties. CS NP can be prepared using sodium tripolyphosphate (TPP) as a cross-linking agent for improving the biological and pharmacokinetic properties of the incorporated drugs [14]. To date, there are no studies employing polymeric NP composed of biocompatible polymers as a strategy to circumvent RESV problems for use against H. pylori.

Here we developed NP of CS-containing trans-resveratrol (RESV-NP) and tested their antibacterial activity against H. pylori.

Both blank and RESV-loaded NP were obtained by the ionotropic gelation method with minor modifications [14], [15]. Chitosan (CS; 2.0 mg/mL) (Mw ≈ 50,000–190,000 Da; deacetylation degree of 75–85 %, Sigma-Aldrich, USA) was dispersed overnight in acetic acid, (glacial PA 0.1 M, Quemis, BRA) and the pH was adjusted to 4.5 using sodium hydroxide 0.1 M (NaOH, Vetec, BRA). Afterward, the trans–resveratrol (RESV) (5.0 mg/mL, Galena, BRA) was dissolved in absolute ethanol (Êxodo Científica, BRA) and then added to the CS dispersion 1:5 v/v (RESV-NP). For the blank NP (from here simply called B-NP), the same volume of water was added instead of RESV solution. The sodium tripolyphosphate (TPP, Sigma-Aldrich, USA) was solubilized in Milli-Q® Water (Type I water,18.2 MΩ·cm at 25 °C; Merck Millipore) at a concentration of 1.0 mg/mL and added slowly over the CS dispersion under magnetic stirring for 40 min at room temperature (RT), at a mass ratio of 2.5:1.0 (CS: TPP) to the formation of B-NP and RESV-NP.

After the development of the particles, the size and polydispersity index (PDI) of the B-NP and RESV-NP were determined using a Malvern Zetasizer Nano ZS® instrument (Malvern Instruments, UK) at an angle of 173° and a temperature of 25 °C. For the ZP analysis, the same equipment was used to determine the electrophoretic light scattering. B-NP and RESV-NP were analyzed after 24 h of being produced. All measurements were performed in triplicate [16].

B-NP and RESV-NP number size (hydrodynamic diameter) distribution and concentrations were measured by the NanoSight NS300 instrument of Malvern (Malvern, UK) in its standard script (SOP Standard Measurement), equipped with a green laser (λ = 532 nm) and NTA Build 3.0 software for tracking Brownian motion. B-NP and RESV-NP were diluted in ultra-pure water. The samples were measured in triplicate at 25 °C [17].

B-NP and RESV-NP were lyophilized in a freeze-dryer model Freezone (Labconco) at 50 °C for 24 h and Pmin ∼ 0,100 mBar. After that pellets were prepared by crushed blending B-NP, RESV-NP, and RESV (2 mg) with 200 mg of potassium bromide previously dried for 24 h at 105 °C (KBr; Specac, USA). A FTIR spectrometer (Perkin-Elmer System 2000 FTIR spectrometer) at the i3S Biointerfaces and Nanotechnology scientific platform was used for the analysis. The spectra were obtained with the wavenumber ranging from 4000 cm−1 to 400 cm−1 employing 32 scans (4 cm−1 spectral resolution) [18].

Cryo-TEM was carried out based on Marena et al. 2022 [16], with minor modifications. Lacey carbon-coated grids were glow-discharged for 50 s. Next, the vitrobot system (22 °C and 100 % humidity) was filled with liquid nitrogen and ethane until achieving a liquid state (≃180 °C) for sample vitrification. B-NP and RESV-NP (3 μL) were uniform (vortex 10x) and placed into grids. After that, the grids were immersed in ethane and moved to the grid box, followed by the analysis in the workstation at the Brazilian Nanotechnology National Laboratory (LNNano − Campinas, São Paulo, Brazil) on a Talos Arctica G2 microscope (Thermo Fisher Scientific) operating at 200 kV.

To evaluate RESV incorporation into the NP, an Eppendorf® 5810R centrifuge was used and the RESV-NP were centrifuged at 4000 g for 30 min at 4 °C using an ultrafiltration centrifuge tube (100 kD, Amicon Ultra4, Millipore, USA). The supernatant (500 µL) was filtered through a 0.45 µm cellulose acetate membrane (Sigma-Aldrich, USA) and appropriately diluted in methanol (Sigma-Aldrich, USA). After that, RESV-NP were analyzed by HPLC (Agilent Technologies 1290 Infinity I equipment coupled with a diode array detector DAD/UV–visible). The quantity of RESV detected by the equipment is the quantity of non-incorporated drugs [19]. A calibration curve was performed at concentrations ranging from 1 to 100 µg/mL and then EE of RESV-NP was quantified. The analytical quantification methodology used was adapted from [20]. Briefly, xBridge C18 column (50 x 4.6 mm, 3.5 µm), consisting of ultrapure water containing 2 % glacial acetic acid (mobile phase A, Sigma-Aldrich, USA) and methanol (mobile phase B, Sigma-Aldrich, USA) in a ratio of 1:1 (v/v). The flow rate was 0.4 mL/min. The encapsulation efficiency (EE) was determined by indirect method in triplicate from the equation 1:

Equation 1:EE(%)=TheoreticalconcentrationofRESV-concentrationinthesupernatantTheoreticalconcentrationofRESVx100

To evaluate the in vitro release profile of RESV from NP, an in vitro dissolution test was performed according to [21] adapted method. Briefly, in an incubator with orbital agitation (SteadyShakeTM 575), at 37 ± 0.1 °C, under agitation at 100 rotations per minute (rpm).

The test was conducted in simulated gastric fluid pH = 1.2 (SGF), composed of of 0.1 M hydrochloric acid (HCl, Sigma-Aldrich, USA), 0.2 M sodium chloride (NaCl, Sigma-Aldrich, USA), and pepsin (0.32 %, Sigma-Aldrich, USA) with 1.5 % Tween® 80 (Dinâmica, BRA) for 24 h. According to sink conditions, a volume of RESV-NP dispersion corresponding to 200 μg of RESV was added to the respective dissolution medium, in Eppendorf ® tubes, and incubated under the described conditions. At different time points, samples were collected, filtered (0.45 µm cellulose acetate membrane filter); and the amount of RESV released was quantified using the HPLC analytical method described in section 2.3. As a control of free drug release, 200 µg of RESV (powder) was added to an equivalent volume of the dissolution medium (also containing tween) and incubated under the same conditions as the RESV-NP. The experiment was carried out in triplicate.

To determine the interaction of the mucins and B-NP under SGF, porcine mucin-type II (Sigma-Aldrich, USA) was dispersed in the SGF (pH 1.2) in 3 different concentrations (100, 250, and 500 µg/mL) overnight. Then, the B-NP and the RESV-NP were scattered in the mucin solutions mentioned above followed by incubation in agitation (50 rpm) at 37 ˚C for 30 min. Finally, the ZP of B-NP and RESV-NP was analyzed by ZetaSize Nano ZS at 37 ˚C [22].

The human gastric adenocarcinoma cell line AGS (ATCC® CRL-1739™) and MKN-74 (ATCC® CRL-2947) were grown in Roswell Park Memorial Institute (RPMI 1640) + glutamax (Gibco, Invitrogen, UK) and, complemented with 10 % heat inactivated fetal bovine serum (FBS, Gibco), 10 µg/mL of streptomycin (Biowest, FR) and penicillin (Biowest, FR), at 5 % CO2 and temperature of 37 °C in a humidified atmosphere. Cells (1 × 104 cells per well) were seeded (96-well plates) under the same conditions mentioned above and allowed to adhere. After 24 h, the RESV-NP (125 and 3.9 µg/mL), RESV (125 µg/mL), and the controls (B-NP and DMSO 5 %, Synth, BRA) were diluted in the medium and were placed in contact with the cells for 24 h. The concentrations used were based on the MIC of the samples and, RESV-NP was also tested at the MIC concentration of free RESV. Cells with 1 mM H2O2, (Merck, USA) were used as a positive control of cytotoxicity, while only cells with the medium as a negative control. Next, 100 uL resazurin (20 % v/v; Sigma-Aldrich, USA) was added to the plates (96-well, Greiner Bio-one, DE) and incubated in the dark for 4 h. Finally, the solution was placed in the black plates (96-well) and fluorescence was measured (λex = 530 nm, λem = 590 nm) in a microplate fluorometer (Spectra Max GeminiXS, Molecular Devices). To determine cell viability, the average of three replicates was used and presented as the percentage of metabolic activity of treated cells in relation to cells only in RPMI supplemented. [5]. The cell viability was calculated with Equation 2:

Equation 2:Cellviability(%)=treatedcellscontrolcellsx100

Samples were considered cytotoxic if the cell viability was below 70 %, according to ISO standard 10993–5 [23].

Galleria mellonella larvae (weighting from 200 to 400 mg; n = 10 larvae/group) were used to determine the acute in vivo toxicity following the protocol described by [24]. RESV-NP, free RESV, and controls [B-NP, DMSO 5 %, 100 % methanol (death control), and only orifice (orifice control)] were injected (10 μL/larva) in the last proleg using a Hamilton Microliter™ (Hamilton Company, Reno, USA) syringe. RESV-NP dose used was determined based on the maximum RESV concentration possible to incorporate in the system (500 µg/mL). RESV dose used was determined based on the maximum concentration of which RESV (was completely soluble in the vehicle (DMSO 5 %)).

The larvae were placed into a Petri dish at 25 °C and were evaluated at different time points (24, 48, and 72 h). The behavior (loss of mobility, larvae color, presence of melanization) and death of the larvae were evaluated. After non-reaction to physical stimulation, the larvae were considered dead. This experiment was performed in triplicate.

Human blood buffy coats obtained from consenting healthy volunteers were used (Centro Hospitalar de São João, EPE, Porto, Portugal). RBC were obtained using Histopaque 1077 (Sigma-Aldrich, USA) through centrifugation over a density gradient following the producer's instructions. After that, the upper layer part of the plasma was removed, and the lower (containing the RBC) was washed three times in Phosphate Buffered Saline 01.M (PBS) at 250 g for 10 min. The purified RBC were diluted to 2x108 cells/mL in PBS and placed in contact with RESV (250 and 125 µg/mL), RESV-NP (7.8 and 3.9 µg/mL) and respective controls (DMSO 5 % and B-NP), followed by incubation at 37 °C under 5 % CO2 for 3 h. The concentrations used were based on the MIC of the samples. After, samples were centrifuged (900 g, 10 min) and the supernatant was placed on black (96-well) plates (Greiner Bio-one, DE) for absorbance reading. The absorbance values of the released hemoglobin were evaluated at 380, 415, and 450 nm in a microplate reader (Synergy Mx, Biotek, USA). Wells containing cells treated with 0.2 %(v/v) Triton X-100 were used as positive controls and, untreated cells were used as negative control [25]. The amount of hemoglobin (Hb) was calculated according to Equation 3:

Equation 3:Hbvalueofsample=2xA415-(A380+A450)x1000EWhere:

E = 79.46.

Helicobacter pylori human strain J99 (ATCC® 700824) was grown in solid medium Blood Agar (BA; Thermo Scientific, USA) + 10 % defibrinated horse blood (Probiológica, PT) and, 6.25 g/L Vancomycin (Sigma-Aldrich, USA), 0.155 g/L Polymixin B (Sigma-Aldrich, USA), 3.125 g/L Trimethoprim (Sigma-Aldrich, USA) and, 1.25 g/L Amphotericin B (Sigma-Aldrich, USA) under microaerophilic conditions (<5% O2; GenBox System, BioMérieux, FR) for 48 h at a temperature at 37 °C. Subsequently, some colonies were spread in BA and incubated for another 48 h at 37 °C under microaerophilic conditions. Next, bacteria were placed in a T-flask with Brucella Broth medium (BB, Oxoid, FR) + 10 % FBS with an optical density (OD) of 0.1 (λ l = 600 nm; UV/VIS spectrophotometer, Lambda 45, Perkin Elmer, USA) which is approximately 3.3 × 107 colony-forming units per mL (CFU/mL) to prepare the overnight pre-inoculum (18–20 h) under the same temperature and oxygen conditions mentioned above and at 150 rpm [18]. This pre-inoculum was then be used for the all-bacterial assays.

The MIC was determined using the microdilution technique in plates (96 wells) following the methods described in the M7-A6 standard of the Clinical Manual and Laboratory Standards Institute [26]. The bacterial inoculum was set to 0.03 (OD = 600), which is approximately 1 × 107 colony-forming units per mL (CFU/mL) and the B-NP and RESV-NP were added at different concentrations (ranging from 250 µg/mL to 3.9 μg/mL). The MIC was qualitatively determined 24 h after incubation using the (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) (MTT, Sigma-Aldrich) assay. For the MBC, each well of the microplate was plated in BA and incubated as described [27]. Three days later, the number of CFUs was counted for the evaluation of viable bacteria. The experiment was performed in triplicate with duplicates.

H. pylori (1x107 CFU/mL, OD600 = 0.03) was incubated with RESV-NP, B-NP, and free RESV at MIC and 2xMIC concentrations for 24 h into BB + 10 % FBS under microaerophilic conditions (<5% O2), 150 rpm and, 37 °C. At different times (2, 6, 12, and 24 h), 20 µL of each culture was serially diluted and, plated on BA. Three days later, the number of CFUs was counted for the evaluation of viable bacteria [28].

The effect of NP on H. pylori morphology was evaluated by Scanning Electron Microscopy (SEM). The H. pylori J99 strain was incubated with RESV-NP, RESV, and respective controls (B-NP and DMSO 5 %), as described above (Section 2.3.2.). 24 h later, samples were washed (3000 g, 5 min, in PBS) and bacterial pellets fixed for 30 min at RT (glutaraldehyde solution 2.5 % and sodium cacodylate buffer 0.14 M; Merck, DE). Next, bacteria were allowed for 2 h to be fixed to coverslips (Ø=13 mm; VWR, USA). Then, an increasing solution of ethanol/water gradient [50 % (v/v) to 99 % (v/v)] was used for dehydration and submitted to critical point drying (CPD 7501, Polaron, GB). Then, for SEM observation, samples were placed on carbon tape and, covered with a gold/palladium thin film by sputtering (SPI Module Sputter Coater equipment) with a 15 mA current for 80 s. SEM analyses were carried out at Centro de Materiais da Universidade do Porto (CEMUP) and used a high resolution (Schottky) Environmental Scanning Electron Microscope with X-ray microanalysis and Electron Backscattered Diffraction analysis (FEI Quanta 400 FEG ESEM/ EDAX Genesis X4M). Images were obtained at 20000x magnification [18].

For the biofilm activity assay, the H. pylori J99 inoculum (109 CFU/mL) was prepared in BB with 5 % FBS as described by [29], added to a flat-bottom plate (24-well, Falcon®, USA) and incubated under agitation at 100 rpm and using the above-mentioned settings for temperature and atmosphere, to form biofilm at the air–liquid interface. After 48 h, the biofilm was centrifuged (3000 g, 5 min), and culture media was removed and replaced with fresh media. 24 h later, the RESV-NP, RESV, and the controls (B-NP and, DMSO 5 % respectively) were added to the biofilms and, on the next day, bacterial viability was accessed by CFU determination on biofilm. For that, the biofilms were washed twice and, mechanically disrupted (up-down) in 1 mL of PBS with a micropipette and, plated in BA. Plates were incubated under microaerophilic conditions at 37 °C. Three days later the number of viable bacteria was evaluated by CFU counting. The experiment was performed in triplicate [30].

High-content confocal imaging was evaluated using an Opera Phenix (Perkin Elmer), using a 40 × water immersion lens for imagining H. pylori biofilms. The Filmtracer™ LIVE/DEAD™ Biofilm Viability Kit (LL10316; Invitrogen™, USA) was used for the staining according to the manufacturer instructions. Nine fields of view (equating to 0.4 mm2) were observed for each well (3 z-stacks per field at 0.5-μm intervals), ensuring comprehensive visualization of the bacterial monolayer. Triplicate biological and technical were carried out for all experiments. Image analysis was evaluated using Harmony (v4.9) [29], [31].

To carry out the in vitro infection assay, H. pylori J99 and the gastric cell line MKN-74 were used. MKN-74 cells were routinely cultured as described in section 2.6.1. After 2 passages, MKN-74 cells were seeded (105 cells per well) in Transwell® inserts (12-well, 0.4 μm pore size, Millicell® Hanging Cell Culture Inserts) containing only RPMI supplemented with 10 % heat inactivated FBS for 7 days to obtain functional tight junctions. Every 2 days, the RPMI was removed and fresh medium was replaced. After 7 days, the formation of cell monolayers was performed using an epithelial EVOM2® (around 150 Ω.cm2) to determine the transepithelial electric resistance (TEER). Then, the medium on the apical side was removed and was replaced by porcine mucins type II (100 μL of mucins at 1 % w/v, Sigma, USA) for 5 min at 50 rpm in an orbital shaker, followed by 25 min static incubation. Next, 400 μL of a bacterial suspension (1.5 × 108 bacteria/mL) in RPMI supplemented with 10 % FBS were placed on the apical side and incubated under 5 % of CO2 at 37 °C, to enable the bacterial adhesion and migration through mucins to infect the gastric cells. 24 h later, 100 μL of the apical side were withdrawn and replaced by 100 μL of RESV and RESV-NP (4xMIC). After 48 h of the infection, the number of infected gastric cells was evaluated using 150 μL of trypsin (Gibco, USA) for 15 min at 37 °C and, 50 rpm. After that, mechanical disruption of the cells through scraping was done. The samples were diluted in PBS (pH 7.4) and plated in BA. After 3 days at 37 °C, the number of CFUs were counted [32].

Numerical data were expressed as the mean ± standard error and one-way analysis of variance (ANOVA) was performed to determine statistical significance. In cases where ANOVA showed significant differences (p ≤ 0.05), Tukey or Dunnet tests were performed. All the in vitro tests were performed in triplicate. Statistics and graphs were performed using the GraphPad Prism 8 software (GraphPad Software Inc., USA).