Repurposing the anthelmintic drug niclosamide to combat Helicobacter pylori

There is an urgent need to discover novel antimicrobial therapies. Drug repurposing can reduce the time and cost risk associated with drug development. We report the inhibitory effects of anthelmintic drugs (niclosamide, oxyclozanide, closantel, rafoxanide) against Helicobacter pylori strain 60190 and pursued further characterization of niclosamide against H. pylori. The MIC of niclosamide against H. pylori was 0.25 μg/mL. Niclosamide was stable in acidic pH and demonstrated partial synergy with metronidazole and proton pump inhibitors, such as omeprazole and pantoprazole. Niclosamide administration at 1 × MIC concentration, eliminated 3-log10 CFU of H. pylori adhesion/invasion to AGS cells. Interestingly, no resistance developed even after exposure of H. pylori bacteria to niclosamide for 30 days. The cytotoxic assay demonstrated that niclosamide is not hemolytic and has an IC50 of 4 μg/mL in hepatic and gastric cell lines. Niclosamide administration decreased transmembrane pH as determined by DiSC3(5) assay indicating that the mechanism of action of the anti-H. pylori activity of niclosamide was the disruption of H. pylori proton motive force. Niclosamide was effective in the Galleria mellonella-H. pylori infection model (p = 0.0001) and it can be develop further to combat H. pylori infection. However, results need to be confirmed with other H. pylori and clinical strains.

a known anthelmintic that is listed in the WHO essential medicines 20 , and we reported that the drug was active against H. pylori infection. This manuscript further investigated whether they possess activity against H. pylori strain 60190, both in vitro and in vivo.
The MIC of niclosamide was the lowest of the anthelmintics tested (0.25 μg/mL, Table 1), therefore we focused on this agent and time to kill assays were used to confirm the killing properties of niclosamide against H. pylori. Bacterial cells (10 9 /mL) exposed to the compounds at 4× MIC showed potent inhibition of the bacterial cell division relative to DMSO controls (Fig. 1). The finding that niclosamide is bacteriostatic against H. pylori is consistent with previous data reported by our research group claiming that niclosamide is bacteriostatic against MRSA 13 . Of note, clarithromycin, a commonly prescribed antibiotic for H. pylori treatment, is also bacteriostatic 22 . Niclosamide partially synergized with clinical molecules. Use of multiple antimicrobial agents can decrease bacterial resistance and even re-establish the clinical efficacy of certain antibiotics 23 . Checkerboard assays were carried out to determine whether niclosamide can act synergistically against H. pylori when paired with clinical antibiotics and PPIs. Paired combinations of compounds and their observed fractional inhibitory concentration indices (FICI) are listed in Table 2. Synergistic activity, where the combined antibacterial effect  of the 2 antimicrobial agents is more than the sum of their effects alone, is defined by FICI ≤ 0.5, antagonism by FICI > 4.0 and 'no interaction' by FICI > 0.5-4.0 24 . In our assays, antagonism was not observed for any of the combinations (amoxicillin, clarithromycin, metronidazole, omeprazole, and pantoprazole) with niclosamide. Niclosamide was partially synergistic with metronidazole, and the PPIs omeprazole and pantoprazole (Table 2). Notably, PPIs are known for their role in decreasing acid secretion, but they do exhibit mild anti-H. pylori activity 25,26 . The main purpose of PPIs in triple therapy (two antibiotics with one PPI) is to potentiate the antibacterial activity of antibiotics in a hostile gastric acid environment 27 . Taken together, our results indicate that niclosamide is highly active in acidic pH and can also partially synergize with omeprazole and pantoprazole (Table 2).
Inhibitory role on adhesion/invasion. Adhesion is the initial step in the H. pylori infection process 28 . In order to determine the role of niclosamide on adhesion, we allowed H. pylori to adhere to AGS cells for 2 h then administrated niclosamide for 24 h (2× MIC, 0.5 μg/mL). After incubation, we observed that the administrated concentration eliminated the adhered bacteria and reduced CFU/mL counts by 3-log 10 ( Fig. 2A). H. pylori is also reported as a facultative intracellular bacterium 29 and it can survive in the plasma membrane of infected epithelial cells 30 . To determine the role of niclosamide on H. pylori invasion, a gentamicin protection assay was performed and it revealed that niclosamide (2× MIC, 0.5 μg/mL) reduced 3-log 10 CFU in AGS cell lines (Fig. 2B). We found a contradiction between the killing kinetics and adhesion/invasion assays indicating that niclosamide appeared bacteriostatic during the killing kinetics and bacteriocidal during adhesion/invasion assays. We hypothesized that the variance in the efficacy of niclosamide could be due to a difference in bacterial cell concentration. The density of bacterial inoculum in the adhesion/invasion assays was lesser than the time-to-kill assay. A previously reported molecule, omeprazole, acted in a similar manner 31 and it appeared bacteriostatic in high density and bactericidal in a low density of bacteria. In comparison, amoxicillin treatment (0.1 μg/mL) eliminated bacterial adhesion and invasion completely.

Niclosamide treatment inhibits IL-8 secretion.
H. pylori infection induces IL-8 secretion 32 -which, in turn, initiates neutrophil chemotaxis and activation, ultimately causing mucosal damage due to the production of reactive oxygen radicals 33 and IL-8 secretion is an important factor in the immunopathogenesis of peptic ulcers and its relevant role in gastric carcinogenesis 34 . We carried out Real-Time (RT-PCR) to determine whether niclosamide has an inhibitory role in IL-8 secretion. As shown in Fig Stability of niclosamide in acidic pH. In H. pylori therapy, antibiotics need to be active in acidic pH 35,36 .
However, certain antibiotics lose their potency in low pH [37][38][39] . Broth microdilution assays in various acidic pHs were carried out to determine the stability of niclosamide. We found that antibacterial potential of niclosamide did not change in acidic pH (tested between 5.0 to 6.5) (Fig. 3A) and the MIC of niclosamide in low pH remained the same as in neutral pH (MIC 0.25 μg/mL). To determine the potential for resistance, we passaged H. pylori treated with either niclosamide or a standard clinical antibiotic (amoxicillin and clarithromycin) for 30 days. The MIC increased 2-fold for clarithromycin and amoxicillin, while that of niclosamide remained constant through at least 10 passages (Fig. 3B). Similarly, the MIC to amoxicillin and clarithromycin increased when H. pylori bacteria were exposed to these agents for after 15 days and 24 days, respectively. More specifically, during continuous passage with amoxicillin and clarithromycin, the MIC increased by 2-fold and 4-fold, respectively ( Fig. 3B), while MIC to niclosamide remained unchanged, indicating a smaller chance of emerging resistance to niclosamide during H. pylori treatment.
Niclosamide inhibits H. pylori motility at sub-MIC level. Initial colonization of the stomach mucosa by H. pylori is associated with the motility of bacteria 42 . In order to evaluate the effects, if any, of niclosamide on H. pylori motility, we performed a motility assay and found that motility decreased under niclosamide treatment at a dose-dependent manner (from 0.10 to 0.20 μg/mL). The swarming movement of H. pylori bacteria decreased gradually depending on the concentration of niclosamide. At sub-MIC (0.2 μg/mL) concentrations, the motility was abated (   Niclosamide role on H. pylori proton motive force (PMF). Many antimicrobial agents are known to act on bacterial membranes 46 and exposure to niclosamide can disrupt the membrane potential 47 . We hypothesized that the mechanism of action of niclosamide is likely to disrupt H. pylori PMF. The fluorescent probe DiSC 3 (5) was loaded with H. pylori bacteria, which accumulates in the cytoplasmic membrane. The dissipation of PMF can be determined by either increase in fluorescence (disruption of membrane potential) or decrease in fluorescence (dissipation of transmembrane pH) 48 . We found that niclosamide treatment decreased the transmembrane pH of H. pylori (Fig. 7B), as determined by decreased fluorescence. In this series of experiments, we used the known protonophore, CCCP, that disrupts PMF by decreasing transmembrane pH 49 as a positive control (Fig. 7D). Valinomycin (Fig. 7A), that can disrupt membrane potential, and amoxicillin ( Fig. 7C), that has effects on the membrane, were also included in this study. Finally, the inner-membrane permeabilization was also monitored by uptake of the membrane-impermeable DNA-binding dye Sytox Green into H. pylori by measuring the increase in fluorescence due to drug administration. Exposure of cells to niclosamide (64 μg/mL) did not show any changes in cellular fluorescence, indicating that niclosamide does not cause any physical damage to H. pylori membranes (Fig. 7E).
Cytotoxicity. Eventhough niclosamide is approved and widely used agent, we evaluated the cytotoxicity of niclosamide against a gastric cell line (AGS), as well as human red blood cells (h-RBC) and hepatic HepG2 cells. Serially diluted Triton-X (0.001 to 1%) was included as a positive control. As expected, niclosamide caused no hemolysis, while the IC 50 against the gastric cell line, was similar to that of hepatic cells (Fig. 8A-C). Notably, niclosamide is not absorbed through the intestinal wall of the host and in murine models, animals survived even after a dose of 5 mg/kg 50 . The U. S. Environmental Protection Agency has reported that the human oral LD 50 of niclosamide is 1 gm/kg and based on the IC 50 values of niclosamide in vitro 51 , it seems to be cytotoxic. However, according to US EPA, the human oral LD 50 is much higher 51 and niclosamide could eradicate H. pylori from the gastric region with low cytotoxic level.
The in vivo efficacy of niclosamide in the Galleria mellonella model. Giannouli (Fig. 9). The clarithromycin injected group served as the positive control (up to 80% survival at 5 days). Antibacterial susceptibility assays. In vitro antibacterial activity was tested using the broth microdilution assay 53 . Experiments were carried out in triplicate using Müller-Hinton broth (BD Biosciences, Franklin Lakes, NJ, USA) supplemented with 10% FBS in 96-well plates (BD Biosciences) at a total assay volume of 100 μL. Antimicrobials used in the pilot study was selected from our previous HTS assay 14,21,54 and tested against H. pylori. Anthelmentics (niclosamide, oxyclozanide, closantel, rafoxanide) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Two-fold serial dilutions were prepared between the concentration range 0.01-64 μg/mL. An initial bacterial inoculum was adjusted to OD 600 = 0.06 and incubated with test compounds at 37 °C for 3 days in humidified incubators under the 5% CO 2 atmosphere. OD 600 was measured and the lowest concentration of compound that inhibited bacterial growth was reported as the MIC 55 . The broth microdilution assay was used to   Killing kinetic assays. The antibacterial properties of niclosamide against H. pylori were further examined using killing kinetic assays, as previously described 13 . Briefly, agar grown H. pylori bacteria were suspended in fresh MHB with 10% FBS to a density of 10 9 cells/mL and placed into 10 mL tubes (BD Biosciences). Test compounds were then added at the 4× MIC and incubated with agitation at 37 °C under microaerophilic conditions. Aliquots were periodically drawn from the tubes, serially diluted and plated onto Brucella agar (BD Biosciences) supplemented with 10% FBS. CFUs were counted after a 3-day incubation and assays were carried out in duplicate.
Checkerboard assays. Antibacterial synergy was tested using the checkerboard assay 56  Adhesion and invasion assays. The AGS (Gastric adenocarcinoma cell lines) cell line was used to examine inhibition of adhesion and invasion of H. pylori by niclosamide, as described by Schmitt et al. 57 . The AGS cells were grown in DMEM (Gibco), supplemented with 10% fetal bovine serum (FBS) (Gibco) and 1% penicillin/ streptomycin (Gibco) and maintained at 37 °C in 5% CO 2 58,59 . Then, 5 × 10 5 cells in antibiotic and serum-free DMEM were seeded in 6-well plates 24 h prior to infection. H. pylori bacteria at a multiplicity of infection (MOI) = 100 were added and allowed to adhere to the surface of the AGS cells. Planktonic bacteria were removed after 2 h and DMEM with niclosamide (1× MIC) was then added to the wells containing the AGS. The mammalian cells were lysed by 0.1% saponin after 20 h of incubation and the adhered bacteria were then serially diluted, plated in Brucella agar supplemented with 10% FBS, and incubated as described earlier. To determine the bacterial invasion inhibition, AGS cells were treated with DMEM supplemented with 200 μg/mL gentamicin and incubated for 1.30 h to eliminate extracellular bacteria. Antibiotic and serum-free DMEM with and without test compounds were added and incubated in 5% CO 2 for 20 h. Cell lysate preparation, plating, and incubation were carried out as we described in the adhesion assay. Assays were carried out in triplicate 21,49 .

Gene expression assays.
To determine the effect of niclosamide in H. pylori VacA toxin expression, bacteria grown on agar were suspended in Brucella broth supplemented with 10% FBS (OD 600 = 0.06) in dilutions of niclosamide (0, 100, 150, or 200 ng/mL) and incubated for 3 days 60 . After 3 days, the bacteria were washed and treated with TRIZOL (Invitrogen, Carlsbad, CA, USA), and the RNA concentration was determined using a NanoVue Plus spectrophotometer (GE Healthcare, Fairfield, CT, USA). PCR amplifications were performed and the products were analyzed by gel electrophoresis (1.0% agarose) containing SYBR ® Safe (Invitrogen). Gel images were captured and analyzed using the Quantity One System. To test whether niclosamide treatment inhibits H. pylori-induced IL-8 secretion, MKN-28 cells were co-cultured with H. pylori (MOI = 100) in the presence or absence of niclosamide. At 24 hours, RNA was isolated from the infected host cells. cDNA was synthesized using the Verso cDNA (Invitrogen) synthesis kit according to manufacturer instructions. Quantitative real-time reactions were carried out in a Bio-Rad iTaq universal SYBR (Bio-Rad, Hercules, CA USA) one-step kit and a CFX98 real-time PCR cycler 61 . The relative gene expression was calculated from Cq values using a ΔΔCq method. The primer sequences and PCR conditions are listed in Supplementary Table S2. Multi-step emergence of resistance. To determine whether H. pylori can develop resistance against niclosamide, bacterial cells were suspended in MHB with 10% FBS at OD 600 = 0.06 in the presence of serially diluted niclosamide and incubated, as described in the antibiotic susceptibility assays sub-section. From the highest drug concentration allowing visible growth at sub-MIC concentration, aliquots were diluted (1:40) into fresh medium. These were used to inoculate the second set of serial drug dilutions as described by Dalhoff et al. 62 . After a 3 day incubation, H. pylori bacteria were passaged sequentially over 30 days and the OD was recorded.
Motility assays. The motility assay was performed as described previously 63 . In brief, Brucella agar with 10% FBS was prepared in 2 layers. The bottom layer contained pre-casted 1.5% agar and the softer top layer was comprised of 0.4% agar and niclosamide (75, 100, 150 or 200 ng/mL). Agar-grown H. pylori cells were sliced and the densely grown H. pylori agar slice was placed facing up towards the soft layer and incubated as described in bacteria and mammalian cell culture subsection 64 . Scanning electron microscopy. H. pylori bacteria were suspended in Brucella broth with 10% FBS in the presence or absence or niclosamide (1x ,4x ,8x MIC) for 2 h under microaerophilic conditions with agitation. After 2 h, the bacterial cells were harvested and adhered to coverslips using a 1% poly-L-lysine (Sigma-Aldrich) solution. Then, cells were fixed in a 5% glutaraldehyde (Sigma-Aldrich), 4% paraformaldehyde (Sigma-Aldrich), 0.1 M sodium cacodylate (Sigma-Aldrich) solution. After fixation, the cells were washed in a 0.1 M sodium cacodylate buffer and dehydrated by immersion in increasing concentration of ethanol (from 30 to 100%). The critical point drying method (CPD) was applied to dehydrate the samples, the coverslips were mounted on aluminum stubs, and then sputter coated with gold (Emitech K550, Ashford, Kent, UK). Images were taken on a Hitachi 2700 Scanning electron microscope.
Light microscopy imaging for vacuolation. AGS cells were harvested and seeded into a 6 well plate at a concentration of 5 × 10 5 cells, 24 h before experimentation. H. pylori bacteria were harvested and washed with sterile PBS and co-cultured with AGS at a MOI of 100. After incubation for 24 h, the cells were washed, magnified under a microscope, and examined for vacuolation.
Membrane permeabilization assays. Sytox Green (Life Technologies, Carlsbad, CA, USA) was used to probe the effects of niclosamide on H. pylori membrane permeabilization, as previously described 65 . Assays were carried out in duplicate in 96-well plates (Corning). Bacterial cells were re-suspended in phosphate buffered saline (PBS, pH 7.4) to OD 595 nm = 0.2. Sytox Green was added at a final concentration of 5 μM and cells were incubated in the dark for 30 minutes. Cell suspensions (50 μL) were added to 50 μL of compound (64 μg/mL in PBS), and the fluorescence intensity was measured (excitation 485 nm, emission 530 nm) periodically over 60 minutes. DMSO was included as the vehicle control. Membrane disruption effects of compounds were indicated by an increase in cellular fluorescence caused by the enhanced permeability of the membrane impermeable DNA staining dye.
Membrane potential. Membrane potential measurements were performed as described by Scott et al. 66 .
In brief, H. pylori cells (OD 595 nm = 0.2) were washed with a buffer (PBS, 100 mM KCl, 20 mM Glucose) and loaded with 5.0 μM of DISC3 (5) (Molecular Probes/Thermo Fisher Scientific, ON, Canada) dissolved in DMSO. The fluorescence was recorded (excitation 610 nm, emission 660 nm) periodically and valinomycin, amoxicillin, carbonyl cyanide m-chlorophenyl hydrazone (CCCP -a protonophore) were used as controls. The percentage of membrane potential change was calculated by comparing drug-treated cells with the untreated control, over all time points.
Mammalian cell cytotoxicity assays. HepG2 and AGS cells were used to test the cytotoxicity of niclosamide, as described by Kwon et al. 67 . Cells were grown in DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS) (Gibco) and 1% penicillin/streptomycin (Gibco) and maintained at 37 °C in 5% CO 2 . 5 × 10 4 cells in 100 μL were added to wells of 96-well plates. Compounds were serially diluted in serum and antibiotic-free DMEM and added to the monolayer and incubated at 37 °C in 5% CO 2 for 24 h. At 4 h prior to the end of the incubation period, 10 μL of 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2, 4-disulfophenyl)-2H-tetrazolium (WST-1) solution (Roche, Mannheim, Germany) was added to each well. The WST-1 reduction was monitored at 450 nm using a Vmax microplate reader. Assays were performed in triplicate and the percentage survival was calculated by comparing the compound-treated wells to the DMSO-treated vehicle controls.
Human blood cell (RBC) hemolysis assays. Human erythrocytes (Rockland Immunochemicals, Limerick, PA, USA) were used to measure the hemolytic activity of the compounds, as described by Isnansetyo et al. 68 Human erythrocytes (4%, in PBS, 50 μL) were added to 50 μL of serially diluted test compounds in PBS in 96-well plates. After incubating at 37 °C for 1 h, the plates were centrifuged at 500 × g for 5 min and 50 μL of the supernatant from each well was transferred to a second 96-well plate. Absorbance (540 nm) was used as a measure of hemolytic activity. Assays were carried in triplicate.
G. mellonella MRSA infection assay. Assays were performed as described by Giannouli et al. 52 . Twelve randomly selected G. mellonella larvae (Vanderhorst, Inc., St. Mary's, OH, USA) between 300-350 mg were used for each group in the experiment. H. pylori cells were washed with PBS and diluted to OD 600 nm = 0.3, before inoculation into G. mellonella larvae. A 10 μL inoculum was injected into the last left proleg using a 10 μL Hamilton syringe. After 2 h, compounds were administered at into the last right proleg and the wax moths were incubate at 37 °C. Three control groups -(1) injected with PBS only, (2) inoculated with H. pylori but treated with sham injections, (3) no manipulation -were included. G. mellonella survival was evaluated up to 120 h and considered dead if unresponsive to touch. Killing curves and differences in survival were analyzed by the Kaplan-Meier method using GraphPad Prism version 6.04 (GraphPad Software, La Jolla, CA, USA). Statistical analysis (Kruskal-Wallis test) was carried out using the same program.
Statistical analysis. Statistical analysis (Two-way ANOVA followed by Bonfererroni post-test) was carried out using GraphPad Prism version 6.04 (GraphPad Software, La Jolla CA, USA) and p values of < 0.05 were considered significant. Data availability statement. No datasets were generated or analysed during the current study.