The micronutrient zinc inhibits EAEC strain 042 adherence, biofilm formation, virulence gene expression, and epithelial cytokine responses benefiting the infected host

Abstract

Enteroaggregative Escherichia coli (EAEC) is a major pathogen worldwide, associated with diarrheal disease in both children and adults, suggesting the need for new preventive and therapeutic treatments. We investigated the role of the micronutrient zinc in the pathogenesis of an E. coli strain associated with human disease. A variety of bacterial characteristics—growth in vitro, biofilm formation, adherence to IEC-6 epithelial cells, gene expression of putative EAEC virulence factors as well as EAEC-induced cytokine expression by HCT-8 cells—were quantified. At concentrations (≤ 0.05 mM) that did not alter EAEC growth (strain 042) but that are physiologic in serum, zinc markedly decreased the organism’s ability to form biofilm (P < 0.001), adhere to IEC-6 epithelial cells (P < 0.01), and express putative EAEC virulence factors (aggR, aap, aatA, virK) (P < 0.03). After exposure of the organism to zinc, the effect on virulence factor generation was prolonged (>3 h). Further, EAEC-induced IL-8 mRNA and protein secretion by HCT-8 epithelial cells were significantly reduced by 0.05 mM zinc (P < 0.03). Using an in vivo murine model of diet-induced zinc-deficiency, oral zinc supplementation (0.4 µg/mouse daily) administered after EAEC challenge (10¹⁰ CFU/mouse) significantly abrogated growth shortfalls (by >90%; P < 0.01); furthermore, stool shedding was reduced (days 9–11) but tissue burden of organisms in the intestine was unchanged. These findings suggest several potential mechanisms whereby physiological levels of zinc alter pathogenetic events in the bacterium (reducing biofilm formation, adherence to epithelium, virulence factor expression) as well as the bacterium’s effect on the epithelium (cytokine response to exposure to EAEC) to alter EAEC pathogenesis in vitro and in vivo. These effects may help explain and extend the benefits of zinc in childhood diarrhea and malnutrition.

Introduction

Enteroaggregative Escherichia coli (EAEC) is a major pathogen responsible for diarrheal disease in the world, affecting travelers, HIV patients, and children from developing countries, leading to acute or even persistent diarrhea.¹ In some cases, malabsorption and chronic inflammation are predominant in infected patients.^2,3 These distinct clinical presentations are thought to be due to genetic heterogeneity among strains of EAEC, reflecting the carriage of differing profiles of chromosomal and plasmid genes responsible for virulence in the infected host, and resulting in specific disease manifestations.⁴

Malnutrition enhances the severity of EAEC infection.^5,6 In particular, deficiency of micronutrients, especially zinc, has been associated with decreased immunity.^7-9 Supplementation of zinc in children has been shown to improve intestinal barrier function and reduce diarrhea,^10-15 even in individuals without overt zinc deficiency.^16,17 However, the mechanisms by which zinc exerts its effects are not fully elucidated.

A first-line of host defense against these enteric pathogens is the intestinal epithelium through its barrier function, mucus secretion, and rapid cell turnover. Further, the epithelium can trigger a variety of innate and adaptive immune responses.¹⁸ For EAEC pathogenesis, the contact between bacteria and epithelium is a key determinant that initiates immune response,¹⁹ which is responsible for the symptoms and disease manifestations.²⁰ Therefore, interventions that disrupt bacterial adherence to epithelium and subsequent downstream events hold promise to be an effective non-antibiotic therapy against EAEC.

Although anti-microbial effects of zinc have been reported for several pathogenic species, including EPEC, ETEC, and STEC,^21-25 none have characterized its effects on EAEC. Therefore, we sought to investigate whether zinc has a potential role in EAEC-host cell interaction at concentrations similar to physiological serum and intestinal lumen levels (0.01–0.05 mM).^26,27 The objectives of this study were to elucidate the effects of zinc on important properties of a strain of EAEC (042) associated with human disease,²⁸ as it interacts with intestinal epithelium, and to begin to identify specific mechanisms in vitro responsible for these effects. Mechanisms of particular interest were the potential for zinc to alter important traits of EAEC strain 042 such as (1) adherence to the epithelium; (2) gene expression of putative virulence factors; and (3) epithelial cell-generated pro-inflammatory cytokine secretion in response to infection. Further, we examined the ability of zinc deficiency or supplementation to alter the clinical outcomes of EAEC-induced disease in vivo.

Results

Zinc ≤ 0.05 mM does not inhibit EAEC strain 042 growth

Although zinc is an essential micronutrient for bacteria at low concentrations, it is also anti-bacterial at higher concentrations.²⁵ In order to perform studies at concentrations of zinc that are not inhibitory for growth of EAEC strain 042, the bacterium was cultured in liquid medium (DMEM) containing zinc, shaking at 37 °C, at concentrations of 0.01–1.0 mM, serially monitoring bacterial density spectrophotometrically over 16 h. At ≤0.05 mM zinc oxide, growth of EAEC strain 042 was not inhibited. At 0.1 mM zinc oxide, growth was partially inhibited, and it was completely inhibited at 1 mM and 10 mM (Fig. 1). These findings suggested that, to study the effects of zinc that do not involve growth inhibition of the study bacterium, physiological zinc concentrations (≤0.05 mM) could be used. Similar results were seen when using zinc acetate or sulfate, while the negative control, manganese sulfate, had no effect on bacterial growth (data not shown).

Figure 1. Growth of EAEC strain 042 in DMEM in the presence of zinc. Bacteria (1 × 10⁸ /well) were incubated with selected concentration of zinc and monitored spectrophotometrically (A₆₀₀). *P < 0.05, days 5–16, comparing cell growth in the absence of zinc with that at 0.1 mM, 1 mM, and 10 mM zinc (n = 8 wells/condition). Data are representative of 2 independent studies.

Sub-inhibitory zinc concentrations reduce biofilm formation of EAEC strain 042

Generation of biofilm, a known property of EAEC strain 042, is involved in bacterial pathogenesis.^29,30 Because the ability of zinc at physiological concentrations to perturb this property is unknown, biofilm formation was assessed in the presence of selected concentrations of zinc oxide, as quantified by the level of crystal violet staining recorded spectrophotometrically. Biofilm formation was significantly reduced when the bacterium was incubated for three hours with a 0.05 mM (but not 0.01 mM) concentration of zinc oxide (P < 0.001) (Fig. 2A). The pattern of adherence of EAEC strain 042 was visually altered in the presence of 0.05 mM zinc when compared with that in the absence of zinc; bacterial colonies were thinned and more scattered (Fig. 2B). These findings suggested that zinc at concentrations that are sub-inhibitory for growth (≤0.05 mM, Fig. 1) can significantly reduce biofilm formation by the EAEC strain 042.

Figure 2. Biofilm formation by EAEC strain 042 incubated in DMEM with zinc. (A) Bacteria were incubated with selected concentrations of zinc for 3 h in DMEM, stained with crystal violet, and biofilm formation quantified spectrophotometrically by absorbance at 570 nm. *P < 0.001, comparing biofilm formation when zinc was absent with that at a 0.05 mM concentration of zinc. n = 8 wells/condition. Data are representative of 2 independent studies. (B) Qualitative assessment by giemsa-staining shows altered morphology of bacterial colonies at 0.05 mM zinc (see text).

Epithelial cell proliferation is not altered by zinc at ≤ 0.01 mM

Zinc is toxic for eukaryotic cells at higher concentrations, resulting in alterations of cell morphology as well as apoptosis.³¹ Because planned in vitro studies (below) involved intestinal epithelial cells, we next explored what concentrations of zinc were non-toxic for cells in our model system. IEC-6 epithelial cells were incubated for four hours with selected concentrations of zinc, and cell viability was assessed by their metabolic activity. While zinc oxide concentrations of 0.2 mM and 1 mM significantly reduced viability of IEC-6 epithelial cells to 38% and 44%, respectively, of that seen in non-zinc treated controls (P < 0.05), lower concentrations of zinc, up to 0.1 mM, showed no effect on viability (Fig. 3). These findings suggested that, for in vitro studies, zinc at concentrations of ≤0.1 mM are non-toxic for IEC-6 epithelial cells, and were used to guide the experiments below. Similar results were seen when using zinc acetate or sulfate, while the negative control, manganese sulfate, had no effect on epithelial viability (data not shown).

Figure 3. Effect of zinc on epithelial cell viability. 4 × 10⁴ IEC-6 cells/well were incubated with selected concentrations of zinc for 3 h at 37 °C, 5% CO₂, and proliferation quantified spectrophometrically (A₄₅₀) after addition of the tetrazolium salt WST-1. Tetrazolium salt is cleaved to formazan by cellular enzymes, which correlates with cell viability. *P < 0.005, comparing proliferation when zinc was absent with that at 0.2 mM and 0.1 mM zinc. n = 9 wells/ zinc concentration. Data are representative of 2 independent studies.

EAEC 042 adherence to IEC-6 epithelial cells is inhibited by zinc

An early stage of EAEC pathogenesis involves adherence to intestinal epithelium by the bacterium’s aggregative adherence fimbriae.²⁹ In order to evaluate the effect of zinc on cell adherence by EAEC, IEC-6 epithelial cells were incubated with EAEC strain 042, washed thoroughly to remove non-adherent bacteria, and the number of remaining bacteria in each well quantified by PCR. In the presence of ≥ 0.05 mM zinc oxide, adherence of EAEC strain 042 to epithelial cells was significantly reduced (P < 0.01), but not at 0.01 mM (Fig. 4). Similar results were observed with zinc acetate and zinc sulfate. These data demonstrate that zinc concentrations that do not affect bacterial growth i.e., ≤ 0.05 mM (Fig. 1) or proliferation of IEC-6 epithelial cells, i.e., ≤ 0.05 mM (Fig. 3), nevertheless inhibit the adherence of EAEC strain 042 to intestinal epithelial cells.

Figure 4. Bacterial adhesion to epithelial cells in the presence of zinc. Epithelial cells (1–2 × 10⁶ /well) were incubated with EAEC strain 042, 10⁷/well, for 3 h in FBS-free-medium at 37 °C, 5% CO₂. After thorough washing, cells were collected, DNA extracted, and the number of organisms quantified by qPCR. *P < 0.05, comparing findings when zinc was absent with those at 0.01 and 0.05 mM zinc. n = 5 wells/condition. Data are representative of 2 independent studies.

EAEC virulence factor expression reduced by zinc

To explore potential mechanisms by which zinc might alter EAEC virulence, biofilm formation, adherence of EAEC strain 042 to epithelium, and the expression of four putative virulence factors of EAEC were quantified. EAEC strain 042 was incubated with IEC-6 epithelial cells for 3 h in the presence of selected concentrations of zinc, cells harvested, RNA extracted, and RT-PCR performed. We observed that aggR expression, a major regulator of EAEC virulence factors,⁴ significantly decreased by 10-fold when bacteria were cultured with 0.01 mM or 0.05 mM zinc (P < 0.002) (Fig. 5A). At the same time, expression of aap, aatA, and virK decreased significantly at 0.01 mM and 0.05 mM zinc oxide (P < 0.003) (Fig. 5B–D). Nevertheless, the expression of the chaperone protein, degP, with no known association with bacterial virulence of EAEC, increased under identical conditions (Fig. 5E). Immunoblot analysis of Aap showed similar reductions (P < 0.01), while aafA, an adherence protein not associated with virulence, was increased (Fig. 5F and G). These results suggest that zinc at sub-inhibitory concentrations for bacterial growth markedly suppresses transcription of EAEC strain 042 virulence factor as well as subsequent protein expression in vitro.

Figure 5. Expression of EAEC strain 042 virulence factor and control mRNA and protein in the presence of zinc. EAEC strain 042 (10⁷ cfu/well) was incubated in selected concentrations of zinc in FBS-free-medium. Bacterial mRNA was extracted for quantification of expression of four putative EAEC virulence factors (aggR, aap, virK, and aatA) as well as a control gene (degP). (A–E) *P < 0.05, comparing mRNA expression of virulence genes when zinc was absent with that for two non-toxic concentrations of zinc. n = 5 wells/zinc concentration. 16S rRNA was used as the housekeeping gene. (F and G) From additional wells treated identically, adherent bacteria with cells were collected in RIPA buffer, and protein quantified by PAGE/Western blotting using factor-specific antibodies. *P < 0.001 comparing findings with zinc exposure with those from wells not cultured with zinc. The columns represent densitometry readings of triplicate bands shown at the top of each figure. The data are representative of 3 independent studies.

Prolonged effect of zinc on virulence factor expression

While the effects of zinc were evident and significant after three hours in the studies above, the time to onset as well as the duration of zinc’s impact on EAEC virulence factor expression were unknown. Therefore, fresh cultures of EAEC strain 042 (10⁷ cfu/ml) were exposed to either 0.05 mM zinc sulfate in DMEM or DMEM with no zinc for 3 h in tissue culture dishes, then to fresh media (without zinc) for an additional 3 h. Expression of aap, aatA, and virK in the absence of zinc was detected within 60–180 min of initiation of culture (Fig. 6). The largest increase in relative expression was in aap (30-fold) (Fig. 6A). The protein chaperone gene (degP) demonstrated no increase in expression during the same time period (Fig. 6D). In contrast, 0.05 mM zinc attenuated expression by ≥70% reduction at 180 min for aap and aatA, and by 40% for virK (P < 0.05).

Figure 6. Duration of zinc effect on EAEC virulence factors. A series of wells containing EAEC 042 (10⁷/well) were incubated with DMEM medium containing zinc sulfate (0.05 mM) or DMEM with no added zinc for 3 h, after which the media was removed and fresh medium without zinc added (indicated by the dashed line) and incubation continued. Triplicate wells were harvested at each of the times shown, mRNA extracted, and virulence factor expression as well as that of a control gene quantified. (A) aap, (B) aatA, (C) virK, and (D) degP. *P < 0.05, comparing, at each time point, wells containing zinc with those in which zinc was absent.

The duration of zinc’s effect on virulence factor expression when this anion was removed from the culture after three hours was next explored. Expression of virulence factors continued to be reduced to ≤30%, compared with that in control cultures that had not been exposed to zinc. On the other hand, expression of degP, remained 4-fold over that in non-zinc controls (P < 0.05; Fig. 6D). This is of particular importance as MacRitchie et al. and others have demonstrated that degP expression downregulates expression of several virulence factors in enteropathogenic E. coli.^32,33 Only a minimal amount of factor-specific mRNA was detected in the same cultures when they were incubated overnight. Together, these results suggest zinc sulfate has a broad and prolonged impact on generation of EAEC virulence factors as well as a modest effect on the protein chaperone gene (degP).

EAEC strain 042-induced epithelial cell cytokine expression is decreased by sub-inhibitory concentrations of zinc

Intestinal inflammation is reported in persons infected with EAEC^20,34 and, correspondingly, EAEC strain 042 induces cytokine production by intestinal epithelial cells (see below). It is unknown, however, whether zinc might alter epithelial generation of EAEC-induced cytokines in our in vitro model. Therefore, HCT-8 epithelial cells were grown to 80% confluency and challenged with EAEC strain 042 (10⁷ cfu/well) that had been pre-incubated in selected concentrations of zinc sulfate. After 1, 2, and 3 h, cells were harvested for RNA extraction, and analyzed by RT-PCR, using cytokine-specific primers. After 3 h EAEC strain 042 induced IL-8 mRNA ~150-fold (P < 0.0001) in the absence of zinc, and this was reduced ~75% by 0.05 mM zinc (P < 0.005; Fig. 7). TNFα showed similar results with an increase (~800-fold) by EAEC, P < 0.0001, and reduction (~75%) by 0.05 mM zinc, P < 0.005. IL-6 induction was decreased by 0.05 mM zinc (~75%, P < 0.05) and the changes in TGFβ were insignificant. These findings suggest that zinc has an anti-inflammatory role in inhibiting epithelial cell generation of mRNA of potentially proinflammatory cytokines, when exposed to EAEC strain 042.

Figure 7. Cytokine expression in HCT-8 cells. HCT-8 cells were cultured to 80–90% confluency and challenged with EAEC for 30, 60, and 180 min with and without ZnSO₄ (0.01 mM or 0.05 mM). (Aand B) IL-8 and TNFα mRNA were significantly increased after 180 min (P < 0.001). ZnSO₄ (0.05mM) significantly reduced expression of both IL-8 and TNFα at the same time point (P < 0.05). (C) IL-6 mRNA was not significantly altered by EAEC infection or ZnSO₄. (D) TGFβ was significantly reduced by 180 min of EAEC infection (P < 0.001) and ZnSO₄ had no effect.

EAEC-induced growth decrements are abrogated by zinc supplementation

Given the in vitro findings above of important effects of sub-inhibitory concentrations of zinc on bacterial adherence to cells (Fig. 4) and virulence factor generation (Fig. 5), it remained unclear whether these properties would alter EAEC strain 042-induced disease in a mouse model of EAEC.⁵ All mice were made zinc-deficient using a moderately malnourished diet with no added zinc (7% protein, <3.2 ppm zinc). Serum zinc levels in plasma decreased by 35% (0.550 ± 0.061 µg/ml) compared with the level in normally fed mice (0.845 ± 0.0353 µg/ml). After challenge with EAEC strain 042 (1 × 10¹⁰ cfu/mouse), mice were daily supplemented with zinc sulfate (4 µg/mouse daily) by oral gavage beginning 1 h after challenge, with water used as a gavage control.

Mice given a zinc-deficient diet prior to EAEC challenge had a marked growth shortfall (days 4–7, P < 0.05, following challenge compared with infected mice with zinc supplementation. In contrast, mice given zinc supplementation were largely free of growth short-falls and closely paralleled the growth of non-infected controls fed a zinc-deficient diet. Late in the course (days 9 and 11 post-challenge), stool shedding of organisms was less in mice receiving zinc supplementation (P = 0.027, compared with that of infected control mice), while changes in tissue burden in the colon and duodenum at day 14 were not significant (Fig. 8C and D).

Figure 8. Effect of zinc deprivation and replacement in vivo. C57BL/6 mice were fed a zinc-deficient diet (<3.2 ppm) for 14 d, then challenged oro-gastrically with 10¹⁰ cfu/mouse of EAEC strain 042 with or without zinc supplementation. n = 5/group. On day 8, 3 mice from each group were euthanized for tissue, and the reminder followed for a total of 14 d. (A) Growth rate. *P < 0.05 for growth rate, days 4–7, comparing EAEC strain 042-challenged mice that remained zinc deprived vs. those that received daily zinc supplement. (B) Stool shedding in EAEC strain 042-challenged mice described in (A). (CandD) Tissue burden of EAEC strain 042 in the colon (C) and duodenum (D) of mice described in (A). n = 3 or 2/condition at 7 and 14 d respectively. Differences in (B–D) were not statistically significant.

Discussion

Enteroaggregative E. coli (EAEC), a major cause of diarrhea worldwide, is particularly severe in children with malnutrition.² For this population, certain micronutrients are reported to be therapeutic¹³ but the cellular and molecular basis for the benefit is unclear. In the current study, using an in vitro model with intestinal epithelial cells (IEC-6 and HCT-8 cells) co-cultured with a human pathogenic strain of EAEC (042), we sought to elucidate the means by which zinc might ameliorate disease in the EAEC-infected host. Using zinc concentrations that are both physiological in serum (0.01–0.05 mM) and sub-inhibitory for EAEC growth in vitro (Fig. 1), we observed important effects, on both the bacterium (reduced biofilm formation, adherence to epithelium, and generation of virulence factors) and intestinal epithelial cells (diminished EAEC-induced production of inflammatory cytokines).

Zinc has been shown to inhibit enterotoxigenic E. coli’s (ETEC’s) ability to adhere to intestinal epithelial cells in the absence of bacteriocidal effects, through binding to structures with which the bacterium attaches to the epithelial cell surface.³⁵ Further, EAEC is well known for its brick-like structures that enable adherence to the epithelium through fimbrial adhesion and outer membrane proteins. Indeed, epithelial adherence characterized by a distinctive aggregative pattern^28,36 and biofilm formation are regarded as traits likely requisite for the pathogenesis of EAEC.²⁹ In the current study, both adherence and biofilm formation of EAEC strain 042 were significantly reduced in the presence of 0.05 mM zinc (P < 0.001), a concentration not inhibitory for growth of this bacterium.

As a mechanism to account for these findings, our studies suggest that, underlying the anti-biofilm and anti-adherence effects of zinc against EAEC, is downregulation of several genes associated with putative virulence factors for EAEC, including aggR, whose expression decreased by 10-fold when bacteria grow in the presence of 0.05 mM zinc. Corroborating the importance of aggR for disease, children with diarrhea infected with EAEC in Tanzania showed that generation of biofilm is associated with expression of aggR in the infecting bacterium.³⁷ Additionally, in our studies, transcription of three other putative virulence factor genes under aggR control (aap, aatA, and virK) was decreased in the presence of sub-inhibitory concentrations of zinc, while that of a chaperone protein (degP) was increased (Fig. 5A–E). Thus, although studies have shown benefit of zinc at higher concentrations because of its direct bactericidal effect,^24,25 our studies suggest that aggR repression represents an additional and novel property of zinc that may in part account for its benefit against this enteric pathogen.

Infection with EAEC is associated with intestinal inflammation, as suggested in children and adults infected with EAEC who have elevated IL-8 in their stools.^29,34 Jiang correlated this finding to a single nucleotide polymorphism (SNP) in the IL-8 gene of the host since its presence associated with the occurrence of EAEC-associated diarrhea and increased levels of fecal IL-8.²⁰ Modeling epithelial cell cytokine secretion in vitro showed increased IL8 expression by intestinal epithelial cells increased by 150-fold, when cells were co-incubated with EAEC strain 042 for 3 h. This increase was reduced by 75% in the presence of sub-inhibitory concentrations of zinc. (Fig. 7A). EAEC-induced epithelial TNFα and IL-6 expression was decreased by ~75% when co-culture was performed in the presence of 0.05 mM zinc (Fig. 7B and D). A possible mechanism for this effect is that zinc reduced EAEC-induced cytokine responses indirectly by decreasing the time of contact between bacteria and epithelial cells. This notion is supported by the finding that EAEC‘s adherence to epithelia was reduced in the presence of 0.05 mM zinc (Fig. 4), decreasing the time available for activation by EAEC of the innate immune response which is dependent on bacterial contact with the epithelium.¹⁹

In the current study, physiologic concentrations of zinc (0.05 mM), decreased virulence gene expression. These results were independent of the form of zinc (oxide, sulfate, or acetate) and were not seen with a control anion (manganese). Additionally, these studies addressed the duration of exposure to zinc necessary to effect a change in gene expression, as well as the persistence of the effect once exposure to zinc stopped. A full 180 min of zinc exposure was required before aap expression increased in a time-course study (Fig. 6), but, once induced, remained elevated for 180 min or more, even in the absence of zinc. To determine whether intracellular zinc levels persist following zinc exposure to account for this finding, bacteria were lysed over time and zinc quantified. After three hours of zinc exposure, cell-associated zinc rose more than 3-fold (0.07 mM to 2.5 mM), but returned to baseline (0.08 mM) after 3 h in the absence of exogenous zinc. These data suggest that exposure of the bacterium to zinc has lasting effects on virulence, even after the zinc is no longer present.

Zinc-induced reductions in bacterial adherence, biofilm formations and virulence factor expression (Figs. 2–6), though significant and consistent, were modest, raising the question whether these effects could be important in vivo. Therefore, mice were made zinc-deficient by dietary means, challenged with EAEC with/without zinc supplementation (4 μg/mouse) daily and followed. We observed that growth short-falls in diet-induced zinc-deficient mice were markedly reduced when mice were orally supplemented daily with zinc beginning one hour after challenge with EAEC strain 042, although stool shedding (days 9–13) and tissue burden of organisms in the duodenum and colon of these mice was not significantly changed, compared with non-supplemented mice (Fig. 7). One hypothesis is that zinc renders the EAEC less virulent without enhancing the ability of the host to expel the organism. Interestingly, zinc supplementation was also shown to improve performance in piglets challenged with enterotoxigenic Escherichia coli (ETEC), preventing colonization of bacteria and reducing severity of diarrhea.³⁸ Other animal studies have shown diminished weight loss and improved immune response due to zinc supplementation.^21,39 This is consistent with the therapeutic efficacy of zinc in human populations, and recent clinical trials appear to be consistent with this notion as well.^10-12,14,15 Together with already published data on EPEC, STEC, and ETEC, our studies on EAEC provide rationale for the use of zinc against 4 established pathotypes of diarrheagenic E. coli.^33,35

In conclusion, zinc alters important properties of EAEC strain 042 by reducing biofilm formation, adherence to epithelium, virulence factor expression as well as its ability to attenuate EAEC-induced cytokine mRNA by epithelial cells. The broad multi-faceted effects of zinc on EAEC 042 would suggest that zinc supplementation could be beneficial in populations exposed to E. coli and perhaps other enteric pathogens.

Material and Methods

Materials

All tissue culture reagents were obtained from Invitrogen. Cell proliferation reagent WST-1 was purchased from Roche. Kits for DNA and RNA extraction were purchased from Qiagen. iScript cDNA synthesis kits and SYBR green were from Biorad, and primers for pcr from Operon. All zinc salts (-oxide, -sulfate, and -acetate) as well as manganese sulfate were purchased from Sigma. IEC-6 and HCT-8 cells were obtained from ATCC (CRL-1592; CCL-244) and EAEC strain 042 was a kind gift from Dr James Nataro. Twenty-one-day-old male C57BL/6 mice were purchased from Charles River Laboratories, Inc. (027). Upon arrival, mice were acclimated, weighed, and distributed in groups. On day 24 of life, zinc-deprived groups were given isocaloric chow containing 7% protein and no added zinc (<3.2 ppm) (Research Diets, D09081702). All animals continued their diets for a total of 2 weeks prior to challenge with EAEC strain 042. The protocol was approved and is in accordance with the Institutional Animal Care and Use Committee policies of the University of Virginia.

Cell culture and preparation of bacteria

The rat intestinal epithelial cell line IEC-6 (passages 15–25) was grown in Dulbecco’s modified Eagle’s medium (DMEM) with high glucose (4.5 g/L) supplemented with 10% fetal bovine serum (FBS), penicillin/streptomycin (50 U/ml), sodium pyruvate (1 mM) and bovine insulin (10 µg/ml), with incubation at 37 °C in an atmosphere of 5% CO₂. Cells were seeded at 1–2 × 10⁶ cells per well in 6-well-tissue culture plates.

For preparation of bacteria, one colony was picked from LB agar plates and shaken in 1 ml of antibiotic-free DMEM for 3 h. One hundred microliters of a bacterial suspension at 10⁸ per mL (OD₆₀₀ of 1.4) were added to each well, achieving a MOI (multiplicity of infection) of 10:1. Individual wells contained selected concentrations of zinc oxide or sulfate (0.01–2.0 mM, see below). To determine precisely the density of the challenge EAEC strain 042, the bacterial suspension was serially diluted, streaked on MacConkey plates, and enumerated after 24 h of incubation at 37 °C.

Bacterial growth and biofilm formation

For measurement of biofilm formation, bacteria was grown overnight in LB, diluted in DMEM up to an OD₆₀₀ = 0.1 and combined to DMEM with selected concentrations of zinc in a total volume of 200 μL/well in 96-well plates. After a 3 h culture, plates were washed three times with water, fixed with 75% ethanol for 10 min (RT), and stained with 0.5% crystal violet (5 min). The plate was then washed and read spectrophotometrically (A₅₇₀). For a qualitative assessment of biofilm, bacteria were grown in DMEM as for infection of IEC-6 cells protocol, added to chamber-slides, then incubated with selected concentrations of zinc (0.01–1.0 mM). After 3 h, the slides were stained with 10% Giemsa for 20 min, washed, and assessed by light microscopy by two independent examiners.

Preparation of zinc solution

Zinc oxide was dissolved in 5% acetic acid in deionized water; zinc sulfate, zinc acetate, and manganese sulfate were all dissolved in deionized water, for a stock solution of 400 mM; for addition to cultures of cells and/or bacteria, dilutions of the stock solution were prepared in DMEM.³⁵ The pH was maintained at 7.

Proliferation assay (WST)

Proliferation was determined using the tetrazolium salt WST-1 to quantify the rate of epithelial cell growth, following the manufacturer recommendations (Roche). 4 × 10⁴ IEC-6 cells were seeded into each well of a 96-well plate and allowed to attach for 48 h. After wells were washed with PBS, selected concentrations of zinc oxide (0.01, 0.05, 0.1, 0.2, and 1 mM), prepared by dilution in DMEM were added. After 3 h of incubation, 10 µL of tetrazolium salt was added per well, plates incubated for 4 h, and colorimetric changes recorded by an ELISA reader (A₄₅₀).

Adherence assay

IEC-6 cells (1–2 × 10⁶ per well) were seeded into 6-well plates and allowed to attach until 100% confluent. Subsequently, cells were challenged with 10⁷ CFU EAEC/well, and incubated for 3 h in FBS-free-medium containing zinc oxide at specific concentrations. After washing twice with PBS, 0.05% trypsin was used to collect bacterial cells for DNA extraction and quantification of organisms using qPCR (specific primers listed in Table 1).

Table 1. Primer sequences.

Primers	Sequences
aggR F	5′-CCTAAAGGAT GCCCTGATGA-3′
R	5′-GAATCGTCAG CATCAGCTAC A-3′
aap F	5′-TGGAACGCAG ATAATGTGGA3′
R	5′-TACCCCAGAG ACAGACACCC-3′
aatA F	5′-GGGAGGTGCA TTGGGTAATA3′
R	5′-CTGGCGAATC TCTTTTTCCA-3′
virK F	5′-GCCGTGCATT GAGGATATTC-3′
R	5′-GCACAAAGCG ATAATGCTCA A-3′
degP F	5′-CTGACCCTGG GCTTACTGCG-3′
R	5′-CCACGCCCTG ATCTTTGCCT-3′
16s F	5′-GTGCCAGCAG CCGCGGTAA-3′
R	5′-GCCTCAAGGGC ACAACCTCCA AG-3′

Analysis of epithelial cytokine and EAEC virulence factor expression

HCT-8 cells (0.8–1 × 10⁶ per well) were seeded into 6-well plates and allowed to grow until 80% confluent. Cells were challenged with 10⁷ CFU EAEC per well and incubated for 3 h in FBS-free-medium. After washing, bacterial and mammalian cells were harvested with 0.05% trypsin, RNA extracted, cDNA generated, and expression measured using RT-PCR as previously described.⁵

Western blot

Cells were harvested and lysed in buffer containing 50 mM TRIS-HCl (pH 8.0), 150 mM NaCl, 1% NP40, 10 mM NaF, 2 mM Na3VO4, and protease inhibitors. After centrifugation, supernatant was collected and protein concentration quantified. Ten micrograms of protein was analyzed by 4–12% SDS-PAGE in MOPS running buffer and transferred to nitrocellulose. 5% milk in TBS was used as a blocking agent. Membranes were probed with a 1:1000 dilution of aap or aafA rabbit polyclonal antibody and a 1:4000 dilution of anti-rabbit IgG-Cy3 (Li-Cor). Blots were imaged and densitometry measured using a Typhoon (GE Healthcare).

Animal studies

Twenty-one-day-old-C57BL/6 mice, purchased from Charles River Inc. were acclimated, then maintained on a zinc-deficient diet for two weeks. At 38 d of life, individual groups (n = 5) were challenged oro-gastrically with 1 × 10¹⁰ cfu/mouse of EAEC strain 042 or DMEM, then followed daily for their rate of growth as well as for collection of stools. A sub-set of mice (n = 3/group) were euthanized at days 7 or 14 for duodenal and colonic tissue. The intensity of shedding of EAEC in the stool as well as the burden of EAEC organisms associated with stool-free intestine were determined by quantitative pcr, as previously described.⁵

Statistical analysis

Data were analyzed using Graph PadPrism software. Statistical comparison between groups was performed using two-way ANOVA or Mann–Whitney non-parametric test; P values < 0.05 were designated as statistically significant. All experimental conditions were performed in triplicate or greater and each experiment was performed at least twice.

10.4161/viru.26120

Ethics Statement

This study included the use of mice. This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of the University of Virginia (Protocol Number: 3315). All efforts were made to minimize suffering. This protocol was approved and is in accordance with the Institutional Animal Care and Use Committee policies of the University of Virginia. The University of Virginia is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care, International (AAALAC). AAALAC is an independent accreditation body which uses the standards outlined in the Guide for the Care and Use of Laboratory Animals (ILAR, NAS, 1996) and the Animal Welfare Act (CFR 9) as amended (P. L. 94-279) as minimum criteria in evaluating research programs which use laboratory animals. AAALAC accreditation is recognized by research funding agencies as identifying national research institutions which have an acceptable level of standards in the operating procedures and practices of their laboratory animal program.