Activities of muscadine grape skin and polyphenolic constituents against Helicobacter pylori

Authors


Correspondence

Xiuping Jiang, Department of Food, Nutrition, and Packaging Sciences, 217 P&A Bldg., Clemson, SC 29634, USA. E-mail: xiuping@clemson.edu

Abstract

Aims

To identify active phenolic constituents in muscadine grape skin (MGS) extracts and determine interactions among compounds while further exploring their anti-Helicobacter pylori potential in vitro.

Methods and Results

The inhibitory effects of quercetin and resveratrol, active polyphenols identified in MGS extracts, against H. pylori were investigated. Quercetin and resveratrol significantly (< 0·05) reduced H. pylori counts regardless of pH with minimal bactericidal concentrations of 256 and 128 μg ml−1, respectively. MGS extracts displayed the highest efficacy, suggesting additional unidentified compounds not determined in this study. Time-course viability experiments showed a dose-dependent anti-H. pylori response to quercetin and resveratrol. Interestingly, neither quercetin nor resveratrol affected H. pylori outer membrane (OM) integrity as determined by 1-N-phenylnaphthylamine (NPN) uptake assays. However, treatment with MGS extract did increase NPN uptake, indicating OM destabilization possibly by additional unknown components. Furthermore, quercetin was found to enter H. pylori as measured by HPLC supporting intracellular drug accumulation.

Conclusions

Quercetin and resveratrol possess strong anti-H. pylori activity in vitro and are independent of pH. Our results also suggest that these compounds do not affect H. pylori OM integrity as previously hypothesized and that the primary antimicrobial activity of quercetin may be linked to interactions with intracellular components.

Significance and Impact of the Study

The anti-H. pylori effects of quercetin and resveratrol suggest that these compounds may be useful in the dietary prevention and/or treatment of H. pylori infection.

Introduction

Helicobacter pylori is a well-known human pathogen and the aetiology of various gastric diseases including peptic ulcers, gastritis and stomach cancer (Cover and Blaser 1995; Dunn et al. 1997). While infection is common, disease outcome is likely multifactorial with virulence of infecting strain, host age, genetic constitution, and environment all contributing to disease progression (Bergonzelli et al. 2003). Because of the complexity of H. pylori infection, all possible host/pathogen interactions should be considered. One important factor frequently underplayed is host dietary habits which may serve a more important role during and immediately following infection with H. pylori than previously thought (Testerman et al. 2001). Studies have established a direct link between H. pylori infection and host diet/nutrition with high-salt diets associated with increased risk and/or extent of disease (Tsugane et al. 1994; Fox et al. 1999; Willis et al. 1999; Gancz et al. 2008). Contrarily, consumption of fruits and vegetables rich in certain vitamins, antioxidants and constitutive bioactive compounds (e.g. phytochemicals) has been shown to significantly reduce the incidence of H. pylori infection and/or ameliorate associated symptoms (Buiatti et al. 1990; Zhang et al. 1997; Yamada et al. 1998; Bennedsen et al. 1999; Fukai et al. 2002; Yanaka et al. 2009).

Although many strains of H. pylori are susceptible to most currently used antibiotics (e.g. clarithromycin, metronidazole, amoxicillin) in vitro, treatment is increasingly challenging due to antibiotic resistance and reinfection in certain groups. Therefore, novel, diet-based therapeutics for use where conventional antibiotic therapies have failed, are unavailable, and/or expensive have received considerable attention.

Numerous studies have investigated naturally occurring plant-derived substances as potential alternatives for H. pylori prophylaxis or treatment (O'Gara et al. 2000; Bergonzelli et al. 2003; Lin et al. 2005; Paraschos et al. 2007; Yang et al. 2008; De et al. 2009; Pastene et al. 2010). In particular, studies have shown that grape polyphenols have strong anti-H. pylori activity, inhibiting growth (Mahady and Penland 2000; Mahady et al. 2003) while reducing H. pylori- and vacuolating cytotoxin-induced gastritis in animal models (Tombola et al. 2003; Yahiro et al. 2005; Ruggiero et al. 2006, 2007). We have previously reported that muscadine grapes (Vitis rotundifolia) are a valuable source of anti-H. pylori compounds with activity against multiple strains in vitro and in vivo with effects most likely due to major phenolic compounds (i.e. ellagic acid, myricetin, quercetin, trans-resveratrol, gallic acid) acting alone or in synergy (Brown et al. 2009, 2010). Polyphenols are naturally found in fruits and vegetables, especially grapes; however, the combination of ellagic acid, quercetin and resveratrol is unique to muscadine species, suggesting that these compounds may be largely responsible for its reported biological activities (Mertens-Talcott and Percival 2005).

Our earlier studies demonstrating the effectiveness of muscadine grapes and the pure flavonoid quercetin against H. pylori prompted us to further explore these products. The objective of this study was to identify active phenolic constituents in muscadine grape skin (MGS) extracts and further explore their anti-H. pylori mechanisms.

Materials and methods

Chemicals

Pure standards of ellagic acid, gallic acid, myricetin, quercetin, trans-resveratrol and amoxicillin were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dimethyl sulfoxide (DMSO), ethanol, acetonitrile, methanol, acetic acid, ethylenediaminetetraacetic acid (EDTA) and water (HPLC grade) were purchased from Fisher Scientific (Norcross, GA, USA).

MGS extraction

Four muscadine grape cultivars, namely two purple (Carlos and Woodruff) and two bronze (Noble and Cowart) provided by Mr. Jody Martin at the Clemson University Pee Dee Research and Education Center (Florence, SC, USA) were used, and the dried MGSs were prepared as described previously (Brown et al. 2009). A modified method of Jayaprakasha et al. (2003) was used for extraction of total phenolics. Briefly, 5 ml of acetone/water/ acetic acid (90 : 9·5 : 0·5, v/v/v) was added to 1 g of dried MGS each in a 10-ml screw-capped glass tube, vortexed for 30 s and incubated with shaking in a 60°C water bath for 8 h. Following centrifugation to remove debris, the supernatant was collected, and the solvent was removed using a SpeedVac (Thermo Fisher Scientific, Waltham, MA, USA) at 43°C. Dried extracts were stored at −80°C under nitrogen or resuspended in DMSO and used immediately.

Total and major phenolics determination

The total phenolic content of each extract was determined by the Folin-Singleton colorimetric method (Singleton and Rossi 1965) and expressed as mg gallic acid equivalents (GAE) per g extract dry weight (DW).

Major phenolics were determined using conditions described by Pastrana-Bonilla et al. (2003). Briefly, extracted samples were filtered through a 0·2-μm nylon syringe filter (VWR International, West Chester, PA, USA) and injected into a Hewlett–Packard (Avondale, PA, USA) HP 1090 high-performance liquid chromatography (HPLC) system with diode array and fluorescence detectors. The mobile phases were solvent A, methanol/acetic acid/water (10 : 2 : 88, v/v/v); solvent B, acetonitrile and solvent C, water. The linear gradient used for phenolic separation was as follows: at 0 min, 100% solvent A; at 5 min, 90% solvent A and 10% solvent B; at 25 min, 30% solvent A and 70% solvent B, and with 5 min postrun with 100% solvent C. The flow rate was 1 ml min−1. The Regis REXCHROM C18 ODS 4·6 × 250 mm column was used with a column temperature of 45°C. The sample injection volume was 20 μl. All extractions were performed under reduced lighting to protect phenolic compounds from degradation.

Helicobacter pylori culture conditions

Two cagA- and vacA-positive H. pylori strains, ATCC 26695 and D5178 (a clinical isolate), were used in this study. Both strains were routinely grown on brain heart infusion (BHI) agar (Difco Laboratories, Detroit, MI, USA; pH 7·2 ± 0·2) supplemented with 7% horse serum (HS; Sigma, St. Louis, MO, USA) at 37°C for 72 h under microaerobic conditions (5% O2, 10% CO2 and 85% N2) in a GasPak jar (BBL Microbiology Systems, Cockeysville, MD, USA; Jiang and Doyle 2000). Plate-grown bacteria were subcultured in BHI/HS broth (pH 7·2) and grown overnight at 37°C in conical flasks, with shaking (120 rpm), under microaerobic conditions.

Active phenolics determination from MGS extracts

Helicobacter pylori 26695 was grown as described and inoculated onto BHI-HS agar plates (c. 5 × 107 CFU per plate) using an Autoplate 4000 spiral plater (Spiral Biotech, Inc., Norwood, MA, USA). Twenty-five microlitres of each MGS extract, adjusted to 2 mg ml−1 of total phenolics, was added to sterile cotton discs (6 mm in diameter; Becton, Dickenson and Company, Franklin Lakes, NJ, USA), placed onto freshly inoculated plates and incubated under microaerobic conditions for 72 h or until visible growth occurred. Each test was performed in triplicate. Following incubation, zones of inhibition were measured to the nearest millimetre and visually compared to determine any differences among MGS extract effectiveness. After determining phenolic concentrations by HPLC for each extract, single phenolics and all possible combinations were tested against the same H. pylori strain under identical conditions to deduce which phenolic(s) were most likely responsible for anti-H. pylori activity.

In vitro analysis of active compounds in combination against Helicobacter pylori

Combination effects of quercetin and resveratrol were tested using a two-dimensional checkerboard assay (McLaren 1997) in 96-well cluster trays (Corning Costar Corp., Cambridge, MA, USA). Briefly, quercetin and resveratrol were suspended in DMSO, filter-sterilized and added to BHI/HS broth yielding final concentrations ranging from 8 to 512 and 2 to 512 μg ml−1, respectively. Helicobacter pylori was grown as previously described and added to wells containing compounds or broth plus DMSO (control) for a final concentration of c. 1 × 106 CFU per well. Plates were incubated under microaerobic conditions, without shaking, for 24 and 48 h. At each time point, wells were sampled using a replica plater (Sigma) and aseptically transferred (c. 2 μl pin−1) to BHI/HS agar plates supplemented with vancomycin (100 μg ml−1), polymyxin B (3·3 μg ml−1), bacitracin (200 μg ml−1), amphotericin B (50 μg ml−1) and nalidixic acid (10·7 μg ml−1; Sigma) and incubated microaerobically for up to 1 week. Plates were scored by assessing the amount of growth on each spot relative to no-treatment controls. All tests were performed in triplicate.

Rate-of-kill analysis of bioactive compounds against Helicobacter pylori

The bactericidal activity of quercetin, resveratrol and amoxicillin were determined using a modification of the Miles-Misra method (McLaren 1997). Briefly, H. pylori strains 26695 and D5178 were grown as previously described and resuspended in BHI/HS broth to a final concentration of c. 5 × 106 CFU ml−1. Time-course killing effects were determined at both neutral (pH 7·4 ± 0·2) and acidified (pH 5·8 ± 0·2) conditions (the later reflecting pH conditions in the human gastric juxtamucosal environment; Fahey et al. 2002). Quercetin, resveratrol and amoxicillin were tested at 0·5×, 1×, 2× and 4× the MIC determined previously (Brown et al. 2010). After 0, 1, 3, 6 and 24 h of incubation, samples were plated onto BHI/HS supplemented with the antibiotic mixture previously described. Untreated H. pylori was plated as a control. Tests were performed in triplicate and expressed as mean log10 CFU ml−1.

Uptake of 1-N-phenylnaphthylamine (NPN) by Helicobacter pylori

NPN uptake assays were performed to determine MGS extract, quercetin and resveratrol's ability to permeabilize H. pylori membranes. Briefly, an overnight liquid culture of H. pylori 26695 was harvested by centrifugation (3500 g, 10 min), washed twice in 5 mmol l−1 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer (pH 7·4) containing 10 μmol l−1 MgCl2 and resuspended to an OD at 600 nm of 0·5 in 5 mmol l−1 HEPES-5 mmol l−1 glucose (Bina et al. 2000). NPN fluorescence was measured using black 96-well plates (Greiner Bio-One Inc., Monroe, NC, USA). Aliquots (100 μl) of the cell suspension were added to each well containing 50 μl of quercetin, resveratrol, EDTA or HEPES buffer (with and without MgCl2) as a control. Fluorescence measurements were taken using a SPECTRAmax GEMINI microplate spectrofluorometer (Molecular Devices, Inc., Sunnyvale, CA, USA) with an excitation wavelength of 350 nm and an emission wavelength of 420 nm. After addition of NPN (10 μmol l−1 final concentration), measurements were taken following 30-min incubation at 37°C, with shaking before each read. All NPN assays were performed in triplicate.

Uptake of quercetin by Helicobacter pylori

To determine quercetin uptake, H. pylori 26695 was grown as described previously to mid-logarithmic phase. Cells were pelleted by centrifugation (3500 g, 10 min), washed once in PBS and resuspended in BHI broth to a final concentration of c. 1 × 108 CFU ml−1. Heat-inactivated controls were prepared by boiling cells at 100°C for 15 min. After cooling to room temperature, quercetin was added to live and heat-inactivated cell suspensions at a final concentration of 256 μg ml−1. In addition, quercetin (in DMSO) was added to broth as a cell-free control. Cells were incubated microaerobically at 37°C for 0, 1, 2 and 3 h. At each time point, H. pylori counts were determined by plating onto BHI/HS. Additional samples were also collected at each time point and analysed by HPLC under the described conditions. Five millilitre aliquots were collected and centrifuged at 12 000 g for 10 min at 4°C. Supernatants were removed and stored at −20°C until analysis. Bacterial pellets were washed twice in PBS and resuspended into 1 ml in the same buffer. Cells were then subjected to ultrasound for 8 min at 30-s intervals with a 1-min cooling period between pulses. All samples were kept on ice during cell lysis. Sonicates from each collection time were plated onto selective media described previously to verify total cell lysis. Cell material was pelleted by centrifugation at 4000 g for 10 min at 4°C, and the supernatant was collected for immediate HPLC analysis using the same conditions described previously.

Statistical analysis

One- and two-way anova tests were used to determine differences between data. All values were considered statistically significant at < 0·05 using SAS 9.1 software (SAS Institute Inc., Cary, NC, USA).

Results

Major and active phenolics determination from MGS extracts

Total phenolic content was determined from four MGS cultivars with values ranging from 67·9 to 109·8 mg GAE per g DW of starting material. Standardized extracts (2 mg total phenolics ml−1) were subjected to HPLC analysis to reveal any measurable difference in major phenolic profiles and determine selected phenolic concentrations (ellagic acid, myricetin, quercetin, trans-resveratrol, gallic acid). These compounds were chosen because they constitute much of the total phenolics present in MGS as determined previously, have been chemically characterized, are commercially available and have been reported to possess antimicrobial properties against a wide variety of micro-organisms, including H. pylori.

Major phenolics were identified by retention times and characteristic spectra from HPLC chromatograms. Quantification was made by comparing peak areas with those obtained from phenolic standards at known concentrations. Selected compounds were also verified by UV–Vis spectrometry. Table 1 shows the major phenolics identified and quantified in selected muscadine cultivars.

Table 1. Major phenolics in muscadine grape skin (MGS) extracts as analysed by HPLC
CultivarMajor phenolic (μg ml−1 extract)a
Ellagic acidMyricetinQuercetinResveratrolGallic acid
  1. a

    Approximate concentration calculation based on 200 μg ml−1 standards. All MGS samples contained 2 mg ml−1 of total soluble phenolics.

Carlos728·7230·096·024·4119·8
Woodruff764·170·6125·869·9102·3
Noble411·0133·021·483·3125·7
Cowart739·8113·3210·4181·134·5

Following chemical analysis, pure compounds at equivalent concentrations to those in Table 1 were tested alone and in combination against H. pylori 26695 by disc diffusion to identify the most active phenolic(s). Inhibition zones were measured and compared with those using MGS extracts to determine which compound(s) was most effective against H. pylori and whether any additive/synergistic effect was evident between compounds (Table 2).

Table 2. Helicobacter pylori 26695 inhibition by major phenolic combinations by disc diffusion
Phenolic alone or in combinationTotal phenolics(μg ml−1)Zone diameter (mm)a
  1. a

    Average of three independent experiments; MGS, Cowart muscadine grape skin; DMSO, solvent control.

Ellagic acid739·89
Ellagic acid, myricetin853·19
Ellagic acid, quercetin950·211
Ellagic acid, resveratrol920·910
Ellagic acid, gallic acid774·39
Myricetin113·310
Myricetin, quercetin323·713
Myricetin, resveratrol294·412
Myricetin, gallic acid147·811
Quercetin210·413
Quercetin, resveratrol391·516
Quercetin, gallic acid244·913
Resveratrol181·112
Resveratrol, gallic acid215·611
Gallic acid34·58
Ellagic acid, myricetin, quercetin, resveratrol, gallic acid1279·117
MGS extract2000·020
DMSO00

Based on disc diffusion results, MGS extracts were the most effective against H. pylori growth, suggesting possible additive/synergistic action among compounds present in MGS preparations. To further support this, all major phenolic compounds were combined at concentrations equivalent to those determined previously by chemical analyses (Table 2). However, the same result (i.e. inhibition diameter) could not be reproduced, suggesting the presence of additional unidentified antimicrobial components in MGS not determined in this study. Although we were unable to determine the definitive compound(s) responsible for activity using MGS by our method, quercetin repeatedly showed the highest anti-H. pylori activity with large zones of inhibition, clearly defined edges and no observable tailing. Additionally, when combined with resveratrol, quercetin showed a marked increase in activity. Therefore, quercetin and resveratrol were selected for further testing.

Bactericidal activity of quercetin and resveratrol alone and in combination

Having established the bacteriostatic activity of quercetin previously (MIC, 32–64 μg ml−1) and resveratrol (MIC, 16–32 μg ml−1) against multiple H. pylori strains (Brown et al. 2009, 2010), we next evaluated the compounds' bactericidal potency by a time-to-kill assay with the reference H. pylori strain (26695) and a clinical isolate (D5178). All experiments were run in parallel at pH 7·4 (neutral) and 5·8 (acidic) as most H. pylori colonize the mucous layer and gastric pits of the antrum with a pH c. 5·5 (Fahey et al. 2002). Bactericidal activity was defined as a reduction in plate counts of ≥1000 CFU ml−1. Amoxicillin was used as a reference drug.

In time-course studies with quercetin and resveratrol, similar dose-dependent patterns upon H. pylori 26695 and D5178 were observed at all concentrations with MICs and minimal bactericidal concentrations (MBCs) corresponding appropriately with those obtained in additional 24-h MIC and MBC experiments. Anti-H. pylori activity was greater (twofold-lower MIC and MBC) for resveratrol than quercetin. As shown in Fig. 1, the MBC for quercetin and resveratrol was 128 and 64, respectively. There was not much effect of changes in pH over the range examined with either strain or any of the compound concentrations tested. At 4× the MIC, H. pylori populations were reduced almost three logs within 6 h by quercetin and resveratrol at pH 7·4 and within 24 h at pH 5·8. At all subinhibitory concentrations (1× and 0·5× the MIC) of quercetin and amoxicillin tested against strains 26695 and D5178, significant reductions (> 0·05) in viable cell counts were not observed at any time during the assay. Interestingly, 2× concentrations of resveratrol did not result in significant reductions (> 0·05) over the 24-h treatment period (similar to lower doses); suggesting that a concentration threshold must be achieved before resveratrol activity is lethal to H. pylori.

Figure 1.

Time-course bactericidal activity of quercetin (A), resveratrol (B) and amoxicillin (C) on Helicobacter pylori 26695 after exposure to 0·5×, 1×, 2× and 4× the MIC at pH 7·4 (a) and pH 5·8 (b). MICs were 32, 16 and 0·25 μg ml−1 for quercetin, resveratrol and amoxicillin, respectively. (image_n/jam12129-gra-0001.png) 4×; (image_n/jam12129-gra-0002.png) 2×; (image_n/jam12129-gra-0003.png) 1×; (image_n/jam12129-gra-0004.png) 0·5× and (image_n/jam12129-gra-0005.png) 0.

As both quercetin and resveratrol showed a concentration- and time-dependent effect on H. pylori (< 0·05), and preliminary tests suggested quercetin and resveratrol may act synergistically or additively against H. pylori, both compounds were tested using a checkerboard assay for 24 and 48 h with results shown in Fig. 2.

Figure 2.

Combination effects of quercetin and resveratrol against Helicobacter pylori 26695 at 24 (A) and 48 (B) h.

Minimal bactericidal concentrations of quercetin and resveratrol required to completely kill H. pylori were lower with extended treatment time. The lowest combined concentrations of quercetin and resveratrol required to completely kill H. pylori were 16 μg and 128 μg ml−1 (24 h) and 8 μg and 64 μg ml−1 (48 h), respectively.

Effect of MGS extract, quercetin and resveratrol on Helicobacter pylori membrane permeability

Figure 3 shows the results of NPN uptake with MGS extract, quercetin, resveratrol or EDTA (a permeabilizer acting by chelation) with or without MgCl2. EDTA weakened the outer membrane (OM) of H. pylori as indicated by strong increases in NPN uptake. Neither quercetin nor resveratrol enhanced (> 0·05) NPN uptake, suggesting that their effects on H. pylori are likely not associated with reducing OM integrity. In contrast, MGS extract-treated samples showed elevated fluorescence levels. The addition of MgCl2 had no significant (> 0·05) impact on fluorescence levels with any treatment other than EDTA.

Figure 3.

Uptake of NPN by Helicobacter pylori 26695 in the presence of EDTA, MGS, quercetin (40 μmol l−1, 13·3 μg ml−1) and resveratrol (40 μmol l−1, 9·1 μg ml−1) with or without MgCl2. MGS, muscadine grape skin extract (1 mg total phenolics ml−1); RFU, relative fluorescence units. (image_n/jam12129-gra-0006.png) Sample and (image_n/jam12129-gra-0008.png) sample + MgCl2.

Quercetin uptake by Helicobacter pylori

Quercetin and resveratrol did not (> 0·05) affect H. pylori OM integrity as revealed by NPN assays. Therefore, we next set out to determine the degree of quercetin uptake by H. pylori as reported in other cell types. To do this, H. pylori 26695 was treated with quercetin for 1, 2 and 3 h under conditions described previously. HPLC was then used to measure quercetin uptake in live and metabolically active H. pylori cells while measuring overall quercetin losses in culture supernatant, inactive cell biomass (i.e. cell membranes/protein) and cell-free controls (to account for possible oxidation/degradation losses) to confirm active drug uptake and validate our approach.

Quercetin levels in live H. pylori cells were highest following 1-h incubation with 20·2 μg ml−1 detected in cell lysates. At 2 and 3 h, quercetin levels decreased to 14·7 and 11·5 μg ml−1, respectively, while at the same time reducing H. pylori counts c. 0·2 logs. Of the initial quercetin amount added (256 μg ml−1), inactive cell biomass accounted for 3·5% of quercetin losses, whereas culture supernatants and cell-free controls accounted for losses of 24·1 and 18·6%, respectively (Fig. 4).

Figure 4.

Compound distribution following 3-h incubation of Helicobacter pylori 26695 with 256 μg ml−1 of quercetin. (image_n/jam12129-gra-0006.png) Pellet; (image_n/jam12129-gra-0007.png) pellet control; (image_n/jam12129-gra-0008.png) supernatant and (image_n/jam12129-gra-0009.png) supernatant control.

Discussion

Muscadine grapes possess significant anti-H. pylori activity as shown previously (Brown et al. 2009, 2010). However, the primary bioactive component(s) in MGS extracts against H. pylori has yet to be definitively identified but believed to be a major phenolic based on earlier reports and established chemical profiles (ellagic acid > myricetin > quercetin > resveratrol > gallic acid). Therefore, in this study, further work was carried out to identify the most active chemical constituent(s) in MGS responsible for anti-H. pylori activity and explore some of the proposed interactions between these compounds and H. pylori. Following extraction and determination of phenolic concentrations among four muscadine grape varieties, and subsequent in vitro screening, it was found that quercetin, while in association with resveratrol, yielded the greatest anti-H. pylori activity.

Following time-course viability experiments, quercetin and resveratrol both demonstrated bactericidal modes of action against H. pylori regardless of pH (Fig. 1). To evaluate drug interactions, a checkerboard assay was used. The MBC of quercetin was reduced from 128–256 to 8–16 μg ml−1, a fourfold reduction, when used in combination with resveratrol. However, the anti-H. pylori activity of resveratrol remained the same or was reduced, suggesting that quercetin might temporarily interfere with resveratrol entry and/or target binding. However, this has yet to be fully understood.

To determine whether quercetin, resveratrol or MGS extract affects H. pylori via disruption of its OM, NPN uptake in H. pylori 26695 was measured following exposure to these compounds or EDTA in the presence or absence of MgCl2. NPN is a small fluorescent probe used to determine drug translocation across the OMs of Gram-negative bacteria (Bina et al. 2000). Fluorescence results when this hydrophobic probe associates with a glycerophospholipid environment such as the lipid bilayers of the OM interior. Therefore, increased fluorescence indicates weakening of the biological membrane. NPN is normally excluded from the OM and fluoresces only weakly in the extracellular aqueous environment. Additionally, although many bacteria require destabilization of the OM before uptake of hydrophobic antibiotics, (e.g. polymyxin B), H. pylori has been reported to demonstrate relatively high intrinsic uptake of NPN compared to other bacteria (Bina et al. 2000).

Neither quercetin nor resveratrol increased NPN fluorescence levels in H. pylori compared to no-treatment controls, regardless of time or concentration, suggesting a lack of damage to the OM. However, the possibility of active NPN efflux could not be ruled out. Observations of H. pylori by phase-contrast microscopy revealed that cells in the presence of quercetin, with glucose (5 mmol l−1), appeared de-energized over time via reduced motility possibly, suggesting the absence of active NPN efflux (data not shown). This may also suggest that the plasma membrane (chemiosmosis) is affected; however, this was not determined in this study.

As neither quercetin nor resveratrol was shown to affect OM integrity, we next examined quercetin uptake by H. pylori. Quercetin has been reported to rapidly accumulate in the cytosol and mitochondria of eukaryotes (Fiorani et al. 2010; Walgren et al. 2000). Although quercetin was previously found to enter both aerobic and anaerobic, Gram-positive bacteria and Saccharomyces cerevisiae (Braune et al. 2001; Schoefer et al. 2003; LoCascio et al. 2006), to date, no reports exist regarding quercetin uptake by H. pylori.

To determine quercetin uptake, H. pylori cells and culture supernatants were measured to determine intracellular compound accumulation. Because quercetin may oxidize and degrade during extended incubation, cell-free negative controls were included for each trial. In addition, because quercetin may passively bind to various components in the cytoplasmic and/or OM, basal levels of quercetin absorption onto metabolically inactive cell biomass were determined using heat-inactivated H. pylori as control. Therefore, any difference between quercetin detected in extracts of live cells and heat-inactivated cells, cell-free controls or culture supernatants would indicate active uptake.

Because bacterial numbers declined during treatment, accumulation rates and drug efflux could not be determined. However, if quercetin was pumped out of the cell, levels should have increased over time in cell supernatants, minus oxidation and losses due to washing and/or extraction (lysis efficiency was 100% as there was no detectable growth following sonication). Differences in quercetin levels gained or lost may be attributed to its intracellular degradation by metabolically active H. pylori. This may further explain why quercetin levels decreased over time in cell pellets but were not detected in corresponding supernatants; however, no metabolites were revealed by our chromatographic method.

Following treatment, relatively high levels of quercetin were detected in live H. pylori extracts suggesting four possibilities. Quercetin may have intercalated in the hydrophobic region of the cell envelope lipid bilayers, passively diffused through the cell membrane into the cell cytosol, interacted with cell membrane proteins or been actively imported into the cytosol. Also, it should be noted that a combination of these factors may have occurred. However, if quercetin did passively intercalate within the cell envelope or associate with cell membrane proteins, similar amounts should have been detected in heat-inactivated controls as live, metabolically active cells. Because quercetin levels were low in controls compared to live cell extracts, we believe that quercetin actively entered the cytosol (Fig. 4).

We acknowledge that inhibitory concentrations of quercetin and resveratrol are relatively high compared to conventional antimicrobials. This is not uncommon with plant-based compounds. However, the average human daily diet contains between 500 and 1000 mg of mixed flavonoids, supporting their overall safety (Ross and Kasum 2002). Additionally, although concentrations similar to those used in this study may not be realistic in the normal human diet, supplementation may be an option. Furthermore, subinhibitory concentrations of these compounds may be useful against H. pylori infection while not necessarily eliminating the organism as shown in our earlier study (Brown et al. 2010).

In this study, we examined the anti-H. pylori properties of different muscadine grape cultivars and determined that of the major phenolics identified, quercetin and resveratrol have the strongest anti-H. pylori activity in vitro and are independent of pH. Neither quercetin nor resveratrol affected H. pylori OM integrity as previously hypothesized. However, treatment with MGS extract did increase NPN uptake, indicating OM destabilization possibly by additional unknown components not revealed in this study. Furthermore, quercetin was transported into the cell cytosol, suggesting its primary anti-H. pylori activity may be linked with interactions with intracellular components; however, this has yet to be fully determined. Our results indicate that MGS extracts and constitutive phenolics quercetin and resveratrol exert strong anti-H. pylori activity and may have the potential to be incorporated into an effective, diet-based approach for prevention and/or treatment of H. pylori infection.

Acknowledgements

This research was partially supported by a grant from the Institute for Nutraceutical Research at Clemson University and the South Carolina Research Authority. We would like to thank Dr. Melissa Riley and Mrs. Frances Harper for their technical assistance with HPLC analyses.

Ancillary