Zinc deficiency alters host response and pathogen virulence in a mouse model of enteroaggregative escherichia coli-induced diarrhea

Abstract

Enteroaggregative Escherichia coli (EAEC) is increasingly recognized as a major cause of diarrheal disease globally. In the current study, we investigated the impact of zinc deficiency on the host and pathogenesis of EAEC. Several outcomes of EAEC infection were investigated including weight loss, EAEC shedding and tissue burden, leukocyte recruitment, intestinal cytokine expression, and virulence expression of the pathogen in vivo. Mice fed a protein source defined zinc deficient diet (dZD) had an 80% reduction of serum zinc and a 50% reduction of zinc in luminal contents of the bowel compared to mice fed a protein source defined control diet (dC). When challenged with EAEC, dZD mice had significantly greater weight loss, stool shedding, mucus production, and, most notably, diarrhea compared to dC mice. Zinc deficient mice had reduced infiltration of leukocytes into the ileum in response to infection suggesting an impaired immune response. Interestingly, expression of several EAEC virulence factors were increased in luminal contents of dZD mice. These data show a dual effect of dietary zinc in benefitting the host while impairing virulence of the pathogen. The study demonstrates the critical importance of zinc and may help elucidate the benefits of zinc supplementation in cases of childhood diarrhea and malnutrition.

Abbreviations

EAEC

Enteroaggregative Escherichia coli

dZD

defined zinc deficient diet

dC

defined control diet

Cftr

cystic fibrosis transmembrane regulator

Introduction

Enteroaggregative Escherichia coli (EAEC) is a major cause of diarrheal disease worldwide in both children and adults,¹ and commonly affects travelers,^2,3 immuno-compromised individuals,^4,5 and children from developing countries.⁶ EAEC is acquired through the consumption of contaminated food or water. Diarrheal disease caused by EAEC can present either acutely or persist, with mucoid and/ or watery stool.

Malnutrition has persisting effects on child growth and development,⁷ and is associated with recurrent diarrhea with EAEC infection,⁸ and may itself contribute to a malnourished state.⁹ This ‘vicious cycle’ of diarrhea and malnutrition has been shown to cause growth stunting and diminished IQ scores in children.^7,10,11 We previously demonstrated that malnutrition increased severity of the symptoms of EAEC infection in a murine model.^12,13 Deficiency of micronutrients, including zinc, has been associated with decreased immunity.^14,15 Zinc deficiency can increase the risk of a child to develop a more severe diarrheal illness following the ingestion of pathogens.¹⁶ The benefits of zinc on host immunity are widely recognized,¹⁷ and zinc supplementation has been shown to improve intestinal barrier function and reduce diarrhea in children.^18-20

Zinc also exerts effects directly on the EAEC pathogen. We have previously shown that sub-MIC zinc concentrations from 0.01–0.05mM decreases adherence to intestinal epithelial cells in vitro.²¹ The EAEC gene aggR is a major regulator of EAEC virulence factor expression, including factors aap, aatK, and virK.^22,23 In vitro expression of these factors were significantly higher when grown in media without zinc. Supplementation of the media with as little as 0.01 mM zinc greatly decreased virulence factor expression and biofilm formation without inhibiting growth.²¹ Zinc concentrations ≥0.1 mM inhibited growth of the bacterium.

The objectives of this study were to develop an improved model of EAEC infection in a murine system and to use this model to determine the separate and combined effects of zinc on host outcome, host immune response, and EAEC virulence.

Results

Antibiotic pretreatment accelerates and enhances EAEC infection

Antibiotics are commonly used in mouse models of enteric infection to deplete the resident intestinal microbiota prior to infection.^24-26 In order to test the effect of antibiotic pretreatment in our model of EAEC infection, mice fed a defined control diet were given either normal water or a cocktail of metronidazole, vancomycin, colistin, and gentamicin in water for 72 h to deplete the intestinal microbiota. Mice were then switched to normal water 24 h prior to infection to clear the antibiotics. Antibiotic (ab) pretreatment significantly increased the amount of weight loss in mice after a single oral challenge of 10⁹ EAEC strain 042 (Fig. 1A; * P < 0.001 control vs. ab+ inf days 3–5, P < 0.05 inf vs. ab+ inf days 3–5). These weight differences after infection occurred several days earlier than our previous studies without use of antibiotics where weight differences were not observed until at least a week after challenge.¹² In addition, mice treated with antibiotics shed significantly more EAEC than animals that did not receive antibiotics (Fig. 1B; * P < 0.05). These data demonstrate the effectiveness of antibiotic pretreatment enhancing susceptibility in our mouse model of EAEC infection.

Figure 1.

Effect of antibiotic pretreatment on EAEC-induced weight loss and stool shedding. Panel A; Antibiotic pretreatment enhances EAEC infection. Mice were fed a control diet for 14 days, and then given an antibiotic cocktail of metronidazole, vancomycin, colistin, and gentamicin in drinking water for 48 hours. Mice were then switched to normal water 24 hours prior to infection to clear the antibiotics. Antibiotic pretreatment significantly increased the amount of weight loss in after a single oral challenge of 10⁹ EAEC strain 042, * P < 0.001 no ab vs ab+inf days 3–5, P < 0.05 inf vs ab+inf days 3–5. Panel B; Feces were collected from infected mice, DNA extracted, and quantitative PCR assayed for the EAEC gene aap to determine the amount of organism shed. Antibiotic pretreated infected mice (abs+inf) shed significantly greater amounts of EAEC than non-antibiotic treated infected mice (inf), * P < 0.05.

Mice fed a zinc deficient diet have lower serum and intestinal zinc levels

Zinc deficiency is a major contributor to diarrhea, and child mortality¹⁹ and oral zinc supplementation has been shown to reduce the duration of diarrhea in children.²⁷ Normal serum zinc levels range from 0.70–1.10 μg/mL (0.011–0.017 mM) in both humans and mice.^28,29 To determine whether we could induce zinc deficiency in mice, we fed mice either a control diet (dC) or a diet without any added zinc (dZD) for 2 weeks and then assayed serum for zinc levels. As expected, the dZD fed mice had significantly lower serum zinc levels (Fig. 2A; *, P < 0.0001 vs. dC). The control diet contains approximately 30ppm Zn while dZD contains approximately 0.085 ppm Zn. More importantly, we saw a significantly lower level of zinc in colon and ileal contents of mice fed the dZD diet compared to dC (Fig. 2B; P < 0.05).

Figure 2.

Zinc levels in serum and tissue. Panel A; Mice were fed either a control diet (dC) or a diet without any added zinc (dZD) for 2 weeks and then assayed for serum zinc levels. dZD fed mice had the lowest serum zinc levels (*, P < 0.0001 compared to dC). Panel B; Mice were fed for 14 d as described above and euthanized to collect ileum and colon tissue as well as luminal contents. A colorimetric assay for zinc was performed as described in methods and the values normalized to protein content. Zinc levels were significantly lower in the colon and ileum contents of mice fed the dZD diet compared to control (*, P < 0.05).

Zinc deficient mice have diarrhea and greater weight loss after EAEC 042 challenge

Mice maintained on dC or dZD for 2 weeks were pretreated with antibiotic cocktail as described above prior to infection. As shown in Figure 3A, after a single oral challenge of 10⁹ EAEC strain 042, dZD fed mice had pronounced diarrhea (days 1–6) and significant weight loss on days 2–6 (P < 0.01 compared to all other groups). Conversely, control mice (dC) infected with EAEC 042 had an insignificant amount of weight loss and no diarrhea. The presence of diarrhea was scored as positive or negative and was observed as mucoid, watery soft stools, often with a wet area around the anus. There was no blood observed in the stool of any mice in this study. EAEC stool shedding was significantly lower in zinc deficient infected mice on days 5–7, potentially due to the persistent diarrhea in these mice. Stool shedding increased in the zinc deficient mice as the diarrhea subsided around day 7.

Figure 3.

Weight loss, shedding, and tissue burden during EAEC infection in control and zinc deficient infected mice. Panel A; Mice were fed either a control diet (dC) or diet without added zinc (dZD) for 2 weeks. All mice were pretreated with antibiotic cocktail as described in methods prior to infection. After a single oral challenge of 10⁹ EAEC strain 042, dZD fed mice had pronounced diarrhea (days 1–6 post infection; indicated by red “d”) and significant weight loss on days 2–6 (*, P < 0.01 compared to all other groups). Nourished control mice infected with EAEC 042 had an insignificant amount of weight loss and no diarrhea. Panel B; Stool shedding of EAEC was significantly lower in dZD fed mice on days 5 and 7 post infection, potentially due to the persistent diarrhea in these mice. Stool shedding increased in dZD mice after day 7 as diarrhea subsided.

Intestinal cytokines

In order to further investigate leukocyte recruitment in dC and dZD mice infected with EAEC strain 042, we performed quantitative RT-PCR for MCP-1 and KC (mouse homolog of the human cytokine IL-8) on ileum tissue. Both MCP-1 and KC were significantly elevated in dZD EAEC infected mice (Fig. 4A&B). To confirm these results, we performed ELISA assays for MCP-1 and KC on ileal and colon tissue as well as serum. As shown in Figure 4C, MCP-1 levels in ileum and serum were significantly increased in dZD infected mice compared to all other groups (P < 0.001 and P < 0.01 respectively). Similarly, KC levels were significantly elevated in the ileum, colon, and serum of dZD +0 42 mice compared to all other groups (Fig. 4D *, P < 0.05). Additionally, we performed quantitative RT-PCR for several cytokines involved in macrophage and T-cell recruitment, maturation and response; transcripts for TNFα, IL-1β, IL-23α, and Csf3 were significantly lower in dZD + 042 mice compared to dC + 042 mice (Fig. 4E). These data suggest that while markers of epithelial inflammation (MCP-1 and KC) are elevated in dZD + 042 mice, macrophage associated cytokines are significantly lower than in dC + 042 mice. We therefore performed staining of intestinal sections for F4/80 and performed flow cytometry analysis on intestinal sections for leukocytic cell populations.

Figure 4.

For figure legend, see page 591. Figure 4 (See previous page). Local and Systemic Cytokine Expression. Mice were fed either a control diet (dC) or diet without added zinc (dZD) for 2 weeks as described in methods. Three days after a single oral challenge of 10⁹ EAEC strain 042, serum from whole blood, and intestinal tissue were collected at time of euthanasia. Tissue were either lysed in RIPA buffer for ELISA or mRNA isolated and qPCR performed as described in methods. (A) Infected-zinc deficient mice (dZD+042) had significantly higher levels of MCP-1 in ileal tissue than all other groups, P < 0.001. (B) dZD+042 also had significantly more MCP-1 protein present in serum than all other groups, P < 0.01. (C and D) Zinc deficient infected mice had significantly higher MCP-1 and KC protein levels in both ileum and sera as measured by ELISA (*, P < 0.05). Panel E. Ileal expression of several leukocyte produced cytokines *, P < 0.05; **, P < 0.01.

Macrophage recruitment and T-cell populations are altered in zinc deficient mice infected with EAEC

The intraepithelial layer from ileal tissue was isolated on day 3 post-infection and cells were labeled and analyzed by flow cytometry. As shown in Figure 5A, EAEC 042 infection caused a significant increase in CD45⁺CD11b⁺MHCII⁺ cells in dC + 042 mice compared to dC (*, P < 0.05). However, dZD mice had a slightly lower level of leukocytes in the uninfected groups (vs. dC) and had significantly less infiltrating CD45⁺CD11b⁺MHCII⁺ cells in response to EAEC 042 infection (**, P < 0.05). These data suggest an impaired myeloid cell response in to EAEC infection in the milieu of zinc deficiency. CD4/CD8 T-cell ratios were also altered in zinc deficient mice (Fig. 5B); specifically the ratio was significantly higher in dC mice compared to dZD mice (*, P < 0.05). The CD4/CD8 ratio dropped significantly in dC+042 mice (**, P < 0.01) while remaining unchanged in the dZD+042 group. We had immunohistochemistry performed on ileum sections for F4/80 to assess macrophage recruitment. The arrows in Figure 5C point to F4/80 stained macrophages and confirms the flow data shown in Figure 5A using a different myeloid cell marker.

Figure 5.

Leukocyte infiltration in the ileum. Mice were fed either a control diet (dC) or diet without added zinc (dZD) for 2 weeks. All mice were pretreated with antibiotic cocktail as described in methods prior to infection. Three days after a single oral challenge of 10⁹ EAEC strain 042, the intraepithelial layer from ileal tissue was isolated at time of euthanasia as described in methods. Cells were labeled and analyzed by flow cytometry. (A) EAEC 042 infection caused a significant myeloid cell infiltration in nourished infected mice compared to control (*, P < 0.05), while infiltration of this CD45⁺CD11b⁺MHCII⁺ population of cells was significantly less in dZD+042 (**, P < 0.05). (B) CD4⁺/CD8⁺ T-cell ratio was significantly higher in uninfected dC vs. dZD mice (*, P < 0.05), and was significantly diminished after infection (dC+042; **, P < 0.01) while remaining unchanged in the dZD+042 group.

Intestinal architecture is altered and mucus secretion is increased in zinc deficient mice infected with EAEC

It has been previously published by our group and others that EAEC infection causes increased goblet cells in the small intestine.^13,30 Additionally, upregulation of the cystic fibrosis transmembrane regulator (Cftr) has been shown to be associated with increased diarrhea.³¹ We isolated ileum sections from dC and dZD mice infected with EAEC 042 to assess histopathology (Fig. 6A) and adjacent sections for mRNA extraction and qPCR. Villus to crypt ratio was significantly altered in zinc deficient mice infected with EAEC 042 compared to dC or dZD uninfected controls (Fig. 6B; *, P < 0.05). As shown in Figure 6C there was increased mucus secretion (as noted by increased fuchsia staining using the PAS stain) and mRNA expression of Cftr and the mucus gene, Muc2 (Panel D).

Figure 6.

Histopathological findings on intestinal tissues. Ileum sections were isolated and fixed using paraformaldehyde and stained with H&E or periodic acid Schiff (PAS) staining. (A) The intestinal architecture was disrupted in zinc deficient mice challenged with EAEC 042. (B) Additionally, villus to crypt ratio was significantly altered in these mice compared to either uninfected control (*, P < 0.05). (C) Ileum and colon sections were stained with PAS to detect glycoprotein (indicative of mucus) secretion. Zinc deficient animals had increased mucus staining in both ileum and colon. (D) This observation of was confirmed by qPCR for Muc2 and the cystic fibrosis transmembrane conductor regulator, Cftr (*, dC 042 vs dZD 042, P < 0.001).

Effect of zinc deficiency on EAEC virulence factors in vivo

We recently published that culturing EAEC in the presence of sub-MIC levels of zinc significantly reduced aggR and aggR-dependent virulence factor expression in vitro.²¹ To determine if the lower levels of zinc in the intestine (see Fig. 2) altered the virulence of EAEC in vivo, we isolated colon contents from dC and dZD mice 3 d post-infection and isolated mRNA for qPCR analysis of EAEC virulence factor expression. The putative virulence factors aap, shf, virK, aaiC, and aggR were all increased in the dZD colon contents, while the expression of the chaperone protein degP remained unaltered (Fig. 7). These data show that reduced levels of zinc in the intestine allow for increased EAEC virulence factor expression in vivo, likely contributing to increased disease severity.

Figure 7.

EAEC virulence in vivo. Intestinal contents from the ileum were collected 3 d post infection and mRNA isolated to measure EAEC virulence factor expression. Several virulence factors (aap, virK, aaiC, aggR) were significantly upregulated in the absence of zinc (*P < 0.05, **P<0 .01, ***P < 0.0001). Expression levels of the chaperone protein degP were unchanged in the zinc deficient luminal contents.

Discussion

Enteroaggregative Escherichia coli (EAEC) is a major cause worldwide of diarrheal disease, primarily affecting travelers, immunocompromised individuals, and malnourished people in resource-limited countries. Repeated episodes of diarrhea in children, or sub-clinical persistent presentation of diarrheal diseases, including EAEC, may lead to growth or cognitive shortfalls later in life.⁷ Zinc deficiency affects about one-third of the world's population,¹⁴ and intestinal disease and diarrhea can lead to loss of dietary zinc through malabsorption.³²

Our aim in this study was to examine the individual and combined effects of zinc deficiency and EAEC infection in a murine system that may help model zinc deficiency and EAEC infections in humans. We found that zinc deficiency and EAEC infection had synergistic effects altering host outcome negatively (diarrhea, increased weight loss, immunological response), while promoting increased EAEC virulence in vivo.

Zinc deficiency has been associated with adverse effects on immune system function, including thymic atrophy and differentiation and function of T cells.³³ In addition, experimental zinc deficiency in humans was shown to alter T cell populations (lower CD4/CD8 ratio) and decrease cellular levels of zinc in leukocytes.^34,35 Quantitatively, fewer leukocytes has the potential to impact the immune response to infection; moreover, low levels of Zn have been shown to affect cellular responses.³⁶ Zinc deficiency is associated with increased systemic inflammation and increased nuclear factor-κB (NF-κB) activity ³⁷. Similarly, zinc repletion has been shown to decrease activation of the NF-κB pathway.^38,39 IL-2 signaling via NF-κB stimulates growth and differentiation of CD4+ and CD8+ T cells,^40,41 which are key components of the adaptive immune system. In the current study, we observed altered numbers of CD4+ and CD8+ cells in the ileum of zinc deficient mice. Future studies are currently underway to elucidate the role of T-cells in zinc deficiency and EAEC infection.

We have recently demonstrated EAEC infection in a weaned mouse model.^12,42 While mice infected with EAEC strain 042 showed significant growth deficits and stool shedding of organism, this model lacked any significant histopathological differences or typical clinical outcomes such as overt diarrhea. In our other mouse models of intestinal infections (Clostridium difficile, Giardia lambia), we utilize an antibiotic cocktail to reduce the resident microbiota prior to challenge in order to obtain a more robust colonization and host response. In the current study, we enhanced EAEC infection by the use of the cocktail previously described with the modification of no clindamycin injection.^43,44

Antibiotic pretreatment of mice prior to EAEC challenge caused increased weight loss and diarrhea suggesting a role for the resident microbiota (abundance and diversity) offering protection against EAEC infection. The resident microbiota in other enteric infections, such as C. difficile, plays an important protective role in disease outcome. Future studies comparing the intestinal microbiota in normal and antibiotic treated mice can help determine whether specific phylotypes or probiotic interventions are protective in the treatment of EAEC.

In a recent study, we have shown that culturing EAEC in the presence of low levels of Zn greatly reduce virulence factor expression and biofilm formation.²¹ Additionally, in our previous study, mice fed a diet with reduced Zn (∼3.2 ppm) had greater growth deficits than mice supplemented with Zn daily. We therefore chose to test EAEC infection in mice fed a virtually zinc-free protein source defined diet (dZD) or same diet with 30 ppm added Zn (dC). Using this diet (dZD), we demonstrate significantly lower levels of zinc in both blood serum and intestinal tissue (P < 0 .0001 dC vs. dZD). Additionally, for the first time, we observe that dZD mice consistently developed diarrhea when infected with EAEC (compared to no diarrhea in the dC+042 group) and increased weight loss over the first 3 d of infection compared with all other treatment groups (P < 0.01).

Examining the immune response to EAEC infection, our results show that fewer CD11b⁺MHCII⁺ myeloid cells are recruited to the intestine during EAEC infection despite high levels of the chemokines, MCP-1 and KC both locally in the intestine and systemically. Taken together, these results are suggestive of a defect in monocyte/macrophage trafficking and/or maturation rather than recruitment through chemokines. At least one study has looked at zinc-dependent leukocyte adhesion.⁴⁵

Zinc has been shown to inhibit the growth of other enteric pathogens such as salmonellae, Shigella, and Vibrio in vitro.⁴⁶ Additionally, we have previously shown an inhibitory effect of zinc on EAEC in vitro.²¹ In the current study, we were able to demonstrate in vivo, increased expression of several aggR-regulated EAEC virulence factors in the absence of dietary zinc. Dietary supplementation of zinc not only empowers host immunity,³³ but also has effects on pathogen virulence factors and their upstream regulators which may decrease disease severity.²¹

In addition to being malnourished, roughly one-third of the world population is zinc deficient.⁴⁷ Dietary zinc has many important benefits. The WHO supports the use of zinc supplementation in reducing childhood diarrhea in low and middle income countries.⁴⁸ Additionally, a Cochrane Review has recommended the use of zinc in enhancing immune response to the common cold.¹⁷ In the current study, we show the combined effects of dietary zinc on both increasing the host immunity as well as decreasing the pathogenicity of EAEC. In summary, zinc is a critical micronutrient in the prevention and treatment of EAEC infection.

Material and Methods

Reagents

DNeasy, RNeasy, and Stool DNA extraction kits were all purchased from Qiagen (Valencia, CA). iScript cDNA synthesis kits and SYBR green supermix was from BioRad (Hercules, CA) and primers for PCR were purchased from Operon (Huntsville, AL). MCP-1 and KC ELISA kits were purchased from R&D Systems and used according to manufacturer's instructions. Zinc assay kits were from Biovision (Milpitas, California) and used according to manufacturer's instructions. Serum zinc levels were measured at the University of Virginia Healthsystem Core Labs.

Animal studies

This study involved the use of mice. All animals were treated and cared for according to protocols approved by the Institutional Care and Use Committee at the University of Virginia, as per Public Health Service policy and the Animal Welfare Act requirements. Mice used in this study were male, 28 d old, C67BL/6 strain, and ordered from Charles River Laboratories (Wilmington, MA;stock #027). Mice weighed approximately 12 grams on arrival and were co-housed in groups up to 5 animals per cage. The vivarium was kept at a temperature of between 68–74°F with a 14 hour light and 10 hour dark cycle.

To control for zinc levels in diet, mice were maintained on either a defined 20% egg white protein control (dC) or defined 20% egg white protein without added zinc (dZD) from Research Diets (New Brunswick, NJ) Table S1; 2 weeks prior to infection and for the duration of the experiment; 3–14 d post infection.^12,49 All diets were isocaloric. Mice were randomly assigned diet and infection groups. Selected groups of mice received antibiotics in their water for 72 h prior to challenge (vancomycin [45 μg/mL], metronidazole [215 μg/mL], gentamicin [35 μg/mL], and colistin [35.5 μg/mL]), followed by 24 hours off antibiotics. Following challenge with EAEC, as described below, mice were monitored and weighed daily. Stools were collected every other day starting 24 hours post-challenge. Studies ranged from 3–14 d in duration. Following completion of the study, animals were humanely euthanized by interperitoneal administration of a Ketamine/Xylazine cocktail followed by manual cervical dislocation. Samples collected from these animals include blood (for serum), ileum and colon sections, as well as luminal contents. Weight at the time of euthanasia was also recorded. Tissues were flash frozen in liquid nitrogen and stored at -80°C for further analysis.

Preparation of enteroaggregative escherichia coli for mouse challenge

EAEC strain 042 was obtained from Dr. Nataro (University of Virginia). All bacteria were grown from glycerol stocks maintained at −80°C and prepared as previously described.^12,13 The bacteria were maintained on ice until administered to mice via oral gavage using 22 gauge feeding needles at a concentration of 10⁹ in 100 μL DMEM high glucose. Uninfected control animals were gavaged with 100 μL DMEM high glucose for vehicle/gavage control.

Stool shedding and tissue burden

DNA was extracted from mouse feces using the DNA Stool Mini kit (Qiagen) for E. coli stool shedding analysis. For tissue burden analysis, adjacent sections to the cecum (1cm distal ileum and proximal colon) were isolated. Genomic DNA was extracted from mouse tissues using the DNeasy kit (Qiagen) for determining E. coli tissue burden. Quantitative realtime PCR for AAP, a protein specific to enteroaggregative Escherichia coli, was performed on the extracted DNA. An 8-point standard curve containing extracted DNA from 10¹–10⁸ cfu/mL was included in each PCR run. The aap gene,⁵⁰ was used as a specific target to quantify presence of EAEC in stool and tissue as previously described.²¹

Flow cytometry

Intraepithelial layer cells from the ileum were isolated as previously described.⁵¹ D’auria, et. al showed that ∼87% of cells elicited in this manner (EDTA/DTT) are epithelial cells with the remaining cells consisting of CD45+ cells including CD11b, CD3, and B220 positive cells. Cells were stained with antibody for 20 minutes at 4°C in the dark for CD8 (53–6.7), MHCII (I-A/I-E), CD4 (GK1.5), CD3 (145–2C11), CD45.2 (104), CD11b (M1/70). Samples were analyzed at the University of Virginia Flow Cytometry Core using a CyAn™ ADP Analyzer (Beckman Coulter). The leukocyte population was gated based on forward/side scatter and equal events within the gate collected (i.e., 100,000). Cell analysis was performed using FlowJo software.

Histology

Tissues were processed and embedded in paraffin at the University of Virginia Histology Core and stained with H&E or periodic acid Schiff (PAS) to stain glycoproteins.

Virulence factor expression analysis

Total cellular RNA was obtained from cecal contents using the RNeasy kits (Qiagen), and cDNA was synthesized from 1 μg

RNA using iScript (Biorad). For quantitative PCR analyses of virulence factor mRNA abundance, cDNA was diluted 1:8; 4 μL of this dilution were used for each

PCR reaction. Reagents from the BioRad real-time PCR kit containing Sybr

Green were used for quantitative PCR reactions. Primer sequences used are listed in Supplemental Data Table 2. The PCR conditions were: 95°C 5 min, followed by 40 cycles of 95°C 30 secs, 60°C 30 secs, and 72°C 30 secs, followed by a melt curve analysis. Data were analyzed and are presented based upon the relative expression method.⁵² The formula used for calculation was: Relative expression = 2^{−(SΔCT- CΔCT)}, where ΔCT is the difference in threshold cycle between the gene of interest (i.e.,aap) and the housekeeping gene (cat). In this equation, S = E. coli challenged mice and C = uninfected mice.

Data management and statistical analyses

Data for all experiments were recorded and analyzed using the GraphPad Prism software program and stored in the Labkey data management system. Comparisons between groups were performed using one-way and 2-way analysis of variance (ANOVA) methods. Data are graphically represented as mean + SEM, in which each mean consists of a minimum of 5 replicates per group with each experiment repeated at least 3 times. Comparisons between groups and tests of interactions were made assuming a 2-factor analysis with the interaction term testing each main effect with the residual error testing the interaction.

Disclosure of Potential Conflicts of Interest

D Bolick, G Kolling, J Moore, L Oliveira, K Tung, C Philipson, M Viladomiu, R Hontecillas, J Bassaganya-Riera, and R Guerrant have no conflicts of interest.

Acknowledgments

The authors would like to thank Dr James Nataro and Fernando Ruiz for the EAEC strain 042 and for virulence factor primer sequences and valuable advice.

Funding

This work was supported by NIAID Contract No. HHSN272201000056C from the National Institutes of Health to JBR at the Virginia Bioinformatics Institute, Virginia Tech. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. L Oliveira was supported by the Fogarty GIDRT Training grant of the National Institutes of Health under award number D43TW006578.

Author Contributions

DTB, GLK, RLG, RH, and JB designed research; DTB, GLK, JHM, LAO conducted research; DTB, GLK, and KT analyzed data; DTB, GLK, JHM, and RLG wrote the paper; DTB had primary responsibility for final content; CP and MV provided helpful revisions; all authors have read and approved the final manuscript.

Ethics Statement

This study included the use of mice. This study was carried out 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.

Supplemental Material

Supplemental data for this article can be accessed on the publisher's website.