Construction and Characterization of Outbreak Escherichia coli O157:H7 Surrogate Strains for Use in Field Studies

Abstract

Escherichia coli O157:H7 has been the causative agent of many outbreaks associated with leafy green produce consumption. Elucidating the mechanism by which contamination occurs requires monitoring interactions between the pathogen and the plant under typical production conditions. Intentional introduction of virulent strains into fields is not an acceptable practice. As an alternative, attenuated strains of natural isolates have been used as surrogates of the virulent strains; however, the attachment properties and environmental stabilities of these attenuated isolates may differ from the unattenuated outbreak strains. In this study, the Shiga toxin (stx ₁, stx ₂, and/or stx _2c) genes as well as the eae gene encoding intimin of two E. coli O157:H7 outbreak isolates, F4546 (1997 alfalfa sprout) and K4492 (2006 lettuce), were deleted. Individual gene deletions were confirmed by polymerase chain reaction (PCR) and DNA sequencing. The mutant strains did not produce Shiga toxin. The growth kinetics of these mutant strains under nutrient-rich and minimal conditions were identical to those of their wild-type strains. Attachment to the surface of lettuce leaves was comparable between wild-type/mutant pairs F4546/MD46 and K4492/MD47. Adherence to soil particles was also comparable between the virulent and surrogate pairs, although the F4546/MD46 pair exhibited statistically greater attachment than the K4492/MD47 pair (p≤0.05). Wild-type and mutant pairs F4546/MD46 and K4492/MD47 inoculated into wet or dry soils had statistically similar survival rates over the 7-day storage period at 20°C. A plasmid, pGFPuv, containing green fluorescent protein was transformed into each of the mutant strains, allowing for ease of identification and detection of surrogate strains on plant material or soil. These pGFPuv-containing surrogate strains will enable the investigation of pathogen interaction with plants and soil in the farm production environment where the virulent pathogen cannot be used.

Introduction

Shiga toxin–producing Escherichia coli O157:H7 (STEC) strains have been identified as a cause of the life-threatening hemolytic–uremic syndrome (Banatvala et al., 2001). Although E. coli O157:H7 (O157) is primarily associated with animals, contamination of fresh produce and outbreaks attributed to O157 have been a recurring issue (Erickson and Doyle, 2012).

O157 strains contain a 4.1-Mb gene backbone that is conserved in all E. coli stains, with an additional 1.4 Mb obtained from horizontal transfer of foreign DNA from prophage or prophage -like elements (Perna et al., 2001; Wick et al., 2005). The production of Shiga toxin and the ability to attach to and efface intestinal epithelial cells are two of the major virulence factors of the pathogen. Similar in structure and function to toxin produced by Shigella dysenteriae type 1 (O'Brien et al., 1983), Shiga toxin genes were horizontally transferred to E. coli by bacteriophage (O'Brien et al., 1984). Two major types of Shiga toxin are produced by STEC strains, Shiga toxin 1 (Stx1) and Shiga toxin 2 (Stx2) (Strockbine et al., 1986) with Stx2c being the most common variant of Stx2 found in O157 strains (Schmitt et al., 1991; Beutin et al., 2002). In addition to Shiga toxin, the locus of enterocyte effacement (LEE) encodes a number of virulence factors, including intimin and its receptor, Tir (Jerse et al., 1990; Deibel et al., 1998). Intimin is a product of the eae gene, which is responsible for the adherence to Tir that has been translocated into enterocytes in the membrane of the small intestine. Intimin-producing strains of enterohemorrhagic E. coli (EHEC) that were Shiga-toxin negative have been associated with causing bloody diarrhea in patients (Bielaszewska et al., 2008).

There are two primary approaches to studying the interaction of O157 with fresh produce. In one approach, defined environmental conditions may be applied in laboratory growth chambers, greenhouses, or hydroponic systems to gain an understanding of how O157 interacts with plants or soil. In the second approach, the interaction of O157 with plants or soils occurs in field studies where variable environmental conditions are encountered. A limitation of field studies is that virulent O157 cannot be intentionally introduced into the environment to contaminate fields. Hence, environmental E. coli O157:H7 Shiga toxin–deficient isolates were used in a field study (Erickson et al., 2010); however, the physiological response of such strains may be different from outbreak-causing virulent strains.

The purpose of this study was to delete the Shiga toxin genes and the eae gene producing intimin in F4546 (1997 alfalfa sprout) and K4492 (2006 lettuce) O157 outbreak strains to create O157 strains that could be used as surrogates for field studies. The mutant strains were labeled with a plasmid-containing green fluorescent protein (pGFPuv) for ease of detection during field- and growth-chamber studies. Finally, the mutant strains were compared to their wild-type, virulent strains for growth, Shiga toxin production, plasmid stability, attachment to leaf surfaces and soil particles, and survival in soil.

Materials and Methods

Bacterial strains

Bacterial strains and plasmids used in this study are listed in Table 1. Strains were streaked from frozen stock onto tryptic soy agar (TSA) (Neogen, Lansing, MI) and incubated at 37°C for 18 to 21 h. Single colonies from TSA plates were used to inoculate Luria Bertani broth for mutant construction or tryptic soy broth (TSB) for use in experimental studies to compare growth, attachment, and survival of wild-type and mutant strains. Growth media for strains containing pGFPuv were supplemented with 100 μg/mL ampicillin (Amp) (Roche Diagnostics, Indianapolis, IN).

Table 1.

Bacterial Strains and Plasmids Used in This Study

Strain or plasmid	Description	Reference or source
pGFPuv	pUC19 derivative containing gfpuv gene Amp^R	Clonetech
F4546	Clinical isolate associated with alfalfa sprout outbreak, 1997	UGA CFS
F4546 pGFPuv	Clinical isolate associated with alfalfa sprout outbreak, 1997, pGFPuv	(Ma et al., 2011b)
K4492	Taco Bell lettuce outbreak, 2006	UGA CFS
K4492 pGFPuv	Taco Bell lettuce outbreak, 2006, pGFPuv	(Ma et al., 2011b)
MD42	F4546 Δstx ₁ Δstx ₂ Δ eae	This study
MD46	F4546 Δstx ₁ Δstx ₂ Δ eae pGFPuv	This study
MD36	K4492 Δstx ₂ Δ eae	This study
MD41	K4492 Δstx ₂ Δstx _2c Δ eae	This study
MD47	K4492 Δstx ₂ Δstx _2c Δ eae pGFPuv	This study

pGFPuv, plasmid-containing green fluorescent protein; UGA CFS, Center for Food Safety, The University of Georgia.

Construction of mutant strains

The genes deleted in strain F4546 were stx ₁, stx ₂, and eae. The genes deleted in strain K4492 were stx ₂, stx _2c, and eae (strain K4492 did not contain the stx ₁ gene). Nucleotide deletions in the stx ₁, stx ₂, stx _2c, and eae genes were constructed using the Quick and Easy Gene Deletion kit (Gene Bridges, Heidelberg, Germany) by identifying 50 nucleotides adjacent to the upstream and downstream insertion sites of each gene to be deleted as well as nucleotides to amplify a neomycin-resistance cassette flanked by a flipase recognition target, FRT (Table 2). The PCR products generated were integrated into the strains by Red/ET recombination directed from plasmid pRedET(tet). Neomycin-resistance cassettes were resolved from the chromosome via flipase recombinase to generate nonpolar mutations (Buchholz et al., 1996; Zhang et al., 1998). EDL933 sequence, Genbank accession number AE005174, was used to determine the stx ₁, stx ₂, and eae regions for deletion in strain F4546 (Perna et al., 2001). The sequence of stx _1A, stx _1B, and the intergenic sequence between nucleotide 2996866 and 2995836 (complement) were deleted. The sequence between nucleotide 1352451 and 1353380 was deleted for stx _2A and stx _2B. Finally, the eae gene was deleted between nucleotide 4667711 and 4665776 (complement) to create strain MD42. TW14359 sequence, Genbank accession number CP001368, was used to determine the stx ₂, stx _2c, and eae regions for deletion in strain K4492 (Kulasekara et al., 2009). The stx ₂ A and B subunits, and the intergenic sequence between nucleotides 3226028 and 3226957 (complement), were deleted. The sequence between nucleotides 2695335 and 2696262 (complement) was deleted for stx _2c A and B subunits, and the intergenic space between. The eae gene was deleted between nucleotide 4657873 and 4655938 (complement) to construct strain MD41. A CaCl₂ heat-shock method (Sambrook et al., 1989) was used to transform plasmid pGFPuv (Clonetech, Palo Alto, CA) containing genes to produce green fluorescent protein (GFPuv) and ampicillin resistance to each recombinant strain. The resulting strains MD46 (F4546 Δstx ₁, Δstx ₂, Δeae pGFPuv) and MD47 (K4492 Δstx ₂, Δstx _2c, Δeae pGFPuv) produced bright green fluorescence under ultraviolet light (365 nm).

Table 2.

Oligonucleotides Used in This Study

Gene	Sequence (5′→3′)
stx ₁F	GCGAAGGAATTTACCTTAGACTTCTCGACTGCAAAGACGTATGTAGATTCGGATCCAATTAACCCTCACTAAAGGGCG ^a
stx ₁R	AAGCTTTTATGACAGGCATTAGTTTTAATGGTTACAGTCATCCCCGTAATTTGCGCTAATACGACTCACTATAGGGCTC
eaeF	CATTAAATTATGCGGCACAACAGGCGGCGAGTCTCGGTAGCCAGCTTCAGAATTAACCCTCACTAAAGGGCG
eaeR	GCACGGTGCTTTTATAGTGTAACTTACTGTTTGCTTATCACCAGATGTGGTAATACGACTCACTATAGGGCTC
stx₂/stx_2cF	TCGTTAAATAGTATACGGACAGAGATATCGACCCCTCTTGAACATATATCAATTAACCCTCACTAAAGGGCG
stx₂/stx_2cR	AACGGTTGCAGATTCGACCGACTGGTCCAGTATTCTTTCCCGTCAACCTTTAATACGACTCACTATAGGGCTC

Underlined sequence is homologous to neomycin cassette template.

Detection of Shiga toxin

Wild-type and mutant pairs, F4546/MD42 and K4492/MD41, were cultured in Gram-negative broth (Thermo Fisher Scientific, Lenexa, KS) at 37°C for 18 to 24 h. In three replicate trials, diluted cultures of these isolates along with positive/negative controls were applied to The ImmunoCard STAT!^® EHEC (Meridian Biosciences, Inc., Cincinnati, OH) to detect Shiga toxin 1 and 2.

Stability of pGFPuv in O157 strains

Mutant O157 strains MD46 and MD47 containing pGFPuv were grown in TSB with and without ampicillin (9 mL) at 37°C for 25 (trial two) or 30 days (trial one). After overnight culture, 10-μL portions were removed and either inoculated into fresh culture medium or applied to TSA-Amp plates for overnight incubation (37°C) and observation of fluorescent colonies under ultraviolet light (365 nm).

Comparison of growth curves for wild-type and mutant O157

Single colonies of wild-type/mutant pairs F4546/MD46 and K4492/MD47 (each strain contained pGFPuv) were inoculated individually into 3-mL TSB-Amp or M9 minimal salts broth (M9B) (Theofel et al., 2009) supplemented with 100 μg/mL ampicillin and incubated overnight at 37°C. Overnight cultures were diluted and approximately 100 colony-forming units (CFU) were added to 50 mL TSB-Amp or M9B-Amp. Cultures were incubated at 37°C with agitation (150 rpm), and sampled at 0, 1, 2, 4, 6, 8, 10, 12, and 24 h (TSB-Amp) or 0, 4, 6, 8, 12, 15, 24, 27, and 30 h (M9B-Amp). Samples were diluted accordingly and 100 μL was applied to TSA-Amp plates for enumeration. Three (TSB-Amp) and two (M9B-Amp) replicate trials were performed at separate times using strains reconstituted from the frozen state.

Culture of wild-type and mutant O157 for attachment and survival studies

Individual inocula were prepared for each isolate (F4546 pGFPuv, MD46, K4492 pGFPuv, and MD47) by thawing the isolates, streaking onto TSA-Amp, and incubating at 37°C for 21–24 h. Individual colonies from these plates were streaked onto a second TSA-Amp plate, incubated overnight, and single colonies from this second plate were used to inoculate 50 mL TSB-Amp that was incubated at 37°C with agitation (150 rpm) for 21–24 h. Cultures were sedimented by centrifugation (4193×g, 15 min, 4°C), washed twice in 0.1% peptone water, and resuspended in 0.1% peptone water to give≈9 log CFU/mL.

Comparison of wild-type and mutant O157 for leaf attachment

Green leaf lettuce leaves were aseptically removed from live plants (≈14-days post-transplantation) and held in 2 L of each isolate (6 log CFU/mL in sterile deionized water [SDW]) for 24 h at 4°C (Xicohtencatl-Cortes et al., 2009) with the stalk-end above the inoculum surface. Leaf samples were dried (30 min/side) prior to either analyzing immediately for total O157 on nonrinsed leaves or placing in a Whirl-Pak bag, swirling with 100 mL SDW for 1 min, and then enumerating for firmly attached O157 on rinsed leaves. Both nonrinsed and rinsed leaves (n=10 each) were macerated in a mortar and pestle with 0.1% peptone water (1:9, wt:vol). O157 were enumerated on TSA-Amp after 24-h incubation at 37°C. Three replicates were performed at separate times for this experiment using freshly prepared cultures.

Comparison of wild-type and mutant O157 for soil attachment

A Tifton Sandy Loam soil was obtained from produce fields in Tifton, GA. Moisture content of the soil was determined following drying at 100°C for 24 h and then water was added to give moisture contents ranging from 8.6% to 10.6%. Soil (5 g) was added to a glass funnel lined with a sterile #5 Whatman filter (GE Healthcare, Piscataway, NJ). Working inocula of each isolate (1 mL of 5 log CFU/mL in SDW) was applied to the center of the soil sample. After 30 min, SDW (25 mL) was applied four times to the contaminated soil to remove unattached organisms. Enumeration of attached O157 was determined by shaking filter and soil from each inoculated sample with 0.1% peptone water (1:9 wt:vol) for 30 s. Tenfold dilutions of this soil dispersion were made and aliquots applied to TSA-Amp plates for subsequent enumeration of fluorescent colonies. Four replicates were performed at separate times using freshly prepared cultures.

Comparison of wild-type and mutant O157 for survival in soil

Tifton soil (380 g) was spread over the bottom of four 3-L Gladware containers (Clorox Company, Oakland, CA) and sprayed with a mixture of either the virulent (F4546, K4492) or mutant (MD46, MD47) isolates (3.8 mL of 7 log CFU/mL in 0.1% peptone water). After mixing, the inoculated soil was acclimated for 24 h (20°C, 70% humidity, dark). Acclimated soil (17 g) was mixed with 153 g of noninoculated soil prior to storage in 532-mL plastic cups (Pactiv Corporation, Lake Forest, IL). Five cups of inoculated soil did not receive water (dry soil), whereas 80 mL of SDW was added to another five inoculated cups to maintain 5 mm of water on the surface of the soil (wet soil) prior to storing dry and wet soil samples at 20°C in the dark. Samples (10 g) were removed from each of the cups on days 0, 1, 2, 3, 4, and 7 and dispersed in 90 mL of 0.1% peptone water. After removing aliquots for enumeration of O157, 90 mL of 2x-TSB-Amp was added to the homogenized soil sample for overnight enrichment at 37°C. Aliquots of these enriched samples were streaked on TSA-Amp plates for overnight incubation at 37°C, after which the presence of fluorescent colonies revealed a positive result for the presence of O157 in the soil sample. Only one replicate trial was performed.

Statistical analysis

Enumeration data were log transformed prior to conducting statistical analyses (StatGraphics Centurion XVI, StatPoint, Inc., Herndon, VA). Growth rates from each of the wild-type/mutant pair trials were calculated from the linear portion of the logarithmic scale-curve and then subjected to the t-test. In addition, both enumeration and percent attachment data for each isogenic mutant pair were subjected to a t-test, whereas the survival in soil of wild-type and mutant isolates was subjected to analysis of variance (p≤0.05).

Results and Discussion

Construction of strains and detection of Shiga toxin

Surrogate strains of O157 associated with produce outbreaks and that had been traced to a production source were constructed in this study. The Gene Bridges system developed by Muyrers et al. (2001) was used to delete Shiga toxin and intimin genes from outbreak strains, F4546 and K4492. Two rounds of mutagenesis using the same oligonucleotides (Table 2) were performed on strain K4492 to delete stx ₂ and stx _2c genes, since the sequence for stx ₂ and stx _2c were 98% identical based on Blast alignment (Altschul et al., 1990). Individual deletions as specified above were confirmed by PCR and DNA sequencing of each strain (data not shown). Though intermediate steps during construction of MD42 and MD41 employed the use of antibiotic-resistant markers, antibiotic resistance was removed (based on sequencing data), resulting in triple mutant (stx1 and /or stx2, stx2c, eae) strains that did not retain these antibiotic resistance markers as evidenced by their inability to grow on agar containing the antibiotic (data not shown).

The wild-type/mutant pairs F4546/MD42 and K4492/MD41 were assayed for Shiga toxin 1 and 2 production using the Immunocard STAT!^® EHEC. The wild-type strain, F4546, produced both Shiga toxin 1 and 2, whereas the mutant, MD42, produced neither (Fig. 1). The lettuce outbreak–associated stain, K4492, which lacks stx₁ , did not produce a band corresponding to Shiga toxin 1 but did produce one corresponding to Shiga toxin 2. The mutant MD41, however, did not produce a band corresponding to Shiga toxin 2 (Fig. 1). Strain MD36, in which stx ₂ was deleted from K4492 but stx _2c was not, did produce a band corresponding to Shiga toxin 2 according to the Immunocard STAT!^® EHEC test, indicating that the test system detected both Shiga toxin 2 and 2c (data not shown).

FIG. 1.

Detection of Shiga toxin in wild-type and mutant O157. (A) Diluent negative control, supplied by kit. (B) F4546. (C) MD42 (F4546 Δstx₁ Δstx₂ Δeae). (D) K4492. (E) MD41 (K4492 Δstx₂ Δstx_2c Δ _eae ). (F) Shiga toxin 1 and 2 positive control, supplied by kit.

Evaluation of plasmid stability in mutant O157 strains

Recombinant strains MD42 and MD41 were transformed with pGFPuv to create strains MD46 and MD47 to enhance the detection of these strains when inoculated into the rhizosphere and/or phyllosphere of fresh produce. As a measure of plasmid stability, pGFPuv in strain MD46 was maintained for 12 days in the absence of antibiotic, whereas pGFPuv in strain MD47 was maintained for 30 days. In the presence of ampicillin, pGFPuv was maintained in MD46 and MD47 for the duration of the trial (30 days). The difference in plasmid stability of these two isolates may be due to inherent physiological differences between the two outbreak isolates used in this study, similar to the variable stability observed for a broad host range plasmid, IncP-1, which had been incorporated into different microbial strains from the same genus or species (De Gelder et al., 2007) and to the wide range of stability shown by pGFPuv incorporated into different isolates of E. coli O157:H7 (Ma et al., 2011b). Confirmation that the plasmid remained in these recipient strains was validated in two field studies wherein strains MD46 and MD47 continued to be detected in and on lettuce plants 14 days after application to field-grown cutback lettuce shoots (Erickson et al., 2013) and in soil 39 days after application of contaminated compost (Erickson et al., 2014).

Growth kinetics of wild-type and mutant O157 strains

In this study, the isogenic pairs (K4492 pGFPuv/MD47 and F4546 pGFPuv/MD46) were subjected to growth in nutrient-rich (TSB-Amp) and minimal media (M9B-Amp). Cultures grown in TSB-Amp exhibited a lag phase as that culture was started with cells from an overnight culture that had entered the stationary phase. In contrast, no lag phase occurred for cultures grown in M9B-Amp as the cells used to start that culture were from an overnight culture that was still in the logarithmic phase. With both isogenic pairs, however, statistically similar logarithmic growth rates were observed in nutrient-rich and minimal media for the wild-type and isogenic mutant (Fig. 2, p>0.05). Hence, similar to other studies involving the loss of virulence factors or the insertion of the GFP-plasmid (Allison et al., 2007; Ma et al., 2011a), the surrogates generated in this study had fitness similar to that of the virulent isolates under nutrient-rich and nutrient-minimal growth conditions.

FIG. 2.

Growth curves of wild-type and mutant O157 and TSB-amp (A, B) or M9B-Amp (C, D) at 37°C with agitation. A and C: F4546 pGFPuv, MD46 (F4546 Δstx ₁ Δstx ₂ Δeae pGFPuv). B and D: K4492 pGFPuv, MD47 (K4492 Δstx ₂ Δstx _2c Δeae pGFPuv). Point values are the means of three (TSB-Amp) or two (M9B-Amp) replicate trials conducted at separate times using freshly prepared cultures grown from the frozen stock. Error bars represent±SD. CFU, colony-forming units.

Comparison of wild-type and mutant O157 for leaf and soil attachment

To ensure that genetic manipulation of the mutant strains did not disrupt innate processes that would affect their fate in environmental systems, studies were conducted to compare the attachment of wild-type and mutant strains to either plant tissue or soil. In the case of plant tissue, leaves were held in each inoculum type for an extended period (24 h) to ensure uniform exposure and sufficient time for attachment at a temperature (4°C) where growth would not occur. No significant differences in the total (nonrinsed leaves) or firmly attached (rinsed leaves) populations or in the percentage of total bacteria that were firmly attached to leaves were observed among the wild-type and mutant O157 strains (Table 3, p>0.05). While no differences in leaf attachment were observed between wild-type and mutant strains, it is acknowledged that intimin is important for attachment to human and animal cells (Lai et al., 2013) and therefore might potentially play a role in pathogen attachment to leaves. In any event, in the case of soil, both wild-type and mutant isolates had comparable degrees of attachment. Increased attachment by either F4596 or MD46 compared to K4492 or MD47 pair (Table 4), however, highlight that the degree of soil attachment can vary between different isolates.

Table 3.

Comparison of Wild-Type and Mutant O157 on Bacterial Attachment to the Surface of Lettuce Leaves

	Mean±SD
Parameter	F4546	MD46	K4492	MD47
Inoculum concentration (log)	5.80±0.50	5.14±0.30	5.04±0.49	5.28±1.01
Total bacteria (log) attached to leaf tissue^a	5.24±0.60	4.65±0.41	4.70±0.34	4.85±0.87
Surface bacteria (log) firmly attached to leaf tissue^b	3.75±0.95	3.10±0.75	3.02±0.79	3.23±0.70
Percent bacteria firmly attached to leaf tissue^c	5.3±3.8	4.5±1.5	4.1±3.5	4.7±3.6

Bacteria attached to leaf following soaking in inoculum for 24 h at 4°C, draining, and then drying.

Bacteria remaining on leaf tissue following soaking, draining, drying, and then swirling in a 10-fold volume of 0.1% peptone water.

(Number surface bacteria/number total bacteria)×100.

Table 4.

Comparison of Wild-Type and Mutant O157 on Bacterial Attachment to Soil

	Mean±SD
Parameter	F4546	MD46	K4492	MD47
Inoculum concentration^a	5.56±0.10	5.03±0.45	5.23±0.49	5.79±0.40
Attached bacteria recovered from soil^b	4.64±0.08	4.11±0.48	4.15±0.53	4.54±0.70
Percent bacteria attached to soil^c	12.4±3.2	12.9±4.9	8.9±3.6	7.5±5.2

Mean log colony-forming units (CFU)/mL.

Mean log CFU/g after rinsing soil.

(Number attached bacteria/number bacteria applied)×100.

Comparison of survival in soil of wild-type and mutant strains

Wild-type and mutant O157 isolates had statistically similar survival rates in wet and dry soil systems over the course of a 7-day sampling period (Table 5, p>0.05). Survival of the wild-type and mutant pairs, however, was much greater in the wet soil system compared to the dry soil (Table 5). Previously, O157 was determined to alter gene expression when exposed to soil (Duffitt et al., 2011) or plant lysates (Kyle et al., 2010), implying metabolic adjustments to their environment to facilitate survival. The similar survival rates of the pairs of wild-type and mutant isolates suggests that the Shiga toxin, intimin, or GFP plasmid genes did not affect mechanisms associated with survival. This conclusion is supported by a previous study in which it was determined that a mutant of O157 EDL933, lacking stx and eae, and a pGFP harboring Pseudomonas aeruginosa GP41B had survival rates in soil systems comparable to their wild-type counterparts (Dandie et al., 2001).

Table 5.

Short-Term Survival of Wild-Type (K4546/K4992) and Mutant O157 (MD46, MD47) Isolates in Wet and Dry Soils

	Escherichia coli O157 populations ^a
	Dry soil		Wet soil
Day	F4546/K4992	MD46/MD47	F4546/K4992	MD46/MD47
0	3.75±0.38	3.63±0.29	3.56±0.16	3.64±0.18
1	3/5	2/5	3.43±0.16	3.62±0.18
2	2/5	0/5	3.45±0.15	3.75±0.09
3	1/5	1/5	3.40±0.22	3.36±0.20
4	0/5	1/5	3.22±0.23	3.22±0.16
7	1/5	0/5	2.22±0.20	2.14±0.32

(Log colony-forming units/g, mean±SD, or number positive by enrichment culture/total number samples analyzed).

The need for safe surrogates of O157 is vital to expand our understanding of how the organism survives in the environment. In this study, virulence genes were removed from two O157 outbreak isolates under the premise that such isolates may possess other attributes that facilitate their survival in fields. Given that growth and attachment to soil and plant surfaces and survival in soil was not affected by the genetic alterations, it is envisioned that these recombinant isolates would respond similarly to the virulent, wild-type strains and hence would be comparable surrogates to apply in field studies.

Footnotes

Acknowledgments

This work was supported by the United States Department of Agriculture NIFSI grant 2007-51110-03812. The authors thank Gaylen Uhlich for helpful technical discussions.

Disclosure Statement

No competing financial interests exist.

References

Allison

, Sattenstall

. The influence of green fluorescent protein incorporation on bacterial physiology: A note of caution. J Appl Microbiol, 2007; 103:318–324.

Altschul

, Gish

, Miller

, Myers

, Lipman

. Basic local alignment search tool. J Mol Biol, 1990; 215:403–410.

Banatvala

, Griffin

, Greene

, Barrett

, Bibb

, Green

, Wells

. The United States national prospective hemolytic uremic syndrome study: Microbiologic, serologic, clinical, and epidemiologic findings. J Infect Dis, 2001; 183:1063–1070.

Beutin

, Kaulfuss

, Cheasty

, Brandenburg

, Zimmermann

, Gleier

, Willshaw

, Smith

. Characteristics and association with disease of two major subclones of Shiga toxin (Verocytotoxin)-producing strains of Escherichia coli (STEC) O157 that are present among isolates from patients in Germany. Diagn Microbiol Infect Dis, 2002; 44:337–346.

Bielaszewska

, Middendorf

, Kock

, Friedrich

, Fruth

, Karch

, Schmidt

, Mellmann

. Shiga toxin-negative attaching and effacing Escherichia coli: Distinct clinical associations with bacterial phylogeny and virulence traits and inferred in-host pathogen evolution. Clin Infect Dis, 2008; 47:208–217.

Buchholz

, Angrand

P-O

, Stewart

. A simple assay to determine the functionality of Cre or FLP recombination targets in genomic manipulation constructs. Nucleic Acids Res, 1996; 24:3118–3119.

Dandie

, Thomas

, McClure

. Comparison of a range of green fluorescent protein-tagging vectors for monitoring a microbial inoculant in soil. Lett Appl Microbiol, 2001; 32:26–30.

De Gelder

, Ponciano

, Joyce

, Top

. Stability of a promiscuous plasmid in different hosts: No guarantee for a long-term relationship. Microbiology, 2007; 153:452–463.

Deibel

, Krämer

, Chakraborty

, Ebel

. EspE, a novel secreted protein of attaching and effacing bacteria, is directly translocated into infected host cells, where it appears as a tyrosine-phosphorylated 90 kDa protein. Mol Microbiol, 1998; 28:463–474.

10.

Duffitt

, Reber

, Whipple

, Chauret

. Gene expression during survival of Escherichia coli O157:H7 in soil and water. Int J Microbiol, 2011; 2011:1–12.

11.

Erickson

, Doyle

. Plant food safety issues: Linking production agriculture with One Health. In: Choffnes ER, Relman DA, Olsen L, Hutton R, Mack A (eds.). Improving Food Safety Through a One Health Approach. Washington, DC: Institute of Medicine, National Academies Press, 2012, pp. 140–175.

12.

Erickson

, Webb

, Diaz-Perez

, Phatak

, Silvoy

, Davey

, Payton

, Liao

, Ma

, Doyle

. Surface and internalized Escherichia coli O157:H7 on field-grown spinach and lettuce treated with spray-contaminated irrigation water. J Food Prot, 2010; 73:1023–1029.

13.

Erickson

, Webb

, Diaz-Perez

, Davey

, Payton

, Flitcroft

, Phatak

, Doyle

. Internalization of Escherichia coli O157:H7 following spraying of cut shoots when leafy greens are regrown for a second crop. J Food Prot, 2013; 76:2052–2056.

14.

Erickson

, Webb

, Diaz-Perez

, Davey

, Payton

, Flitcroft

, Phatak

, Doyle

. Absence of internalization of Escherichia coli O157:H7 into germinating tissue of field-grown leafy greens. J Food Prot, 2014; 77:189–196.

15.

Jerse

, Yu

, Tall

, Kaper

. A genetic locus of enteropathogenic Escherichia coli necessary for the production of attaching and effacing lesions on tissue culture cells. Proc Natl Acad Sci USA, 1990; 87:7839–7843.

16.

Kulasekara

, Jacobs

, Zhou

, Wu

, Sims

, Saenphimmachak

, Rohmer

, Ritchie

, Radey

, McKevitt

, Freeman

, Hayden

, Haugen

, Gillett

, Fong

, Chang

, Beskhlebnaya

, Waldor

, Samadpour

, Whittam

, Kaul

, Brittnacher

, Miller

. Analysis of the genome of the Escherichia coli O157:H7 2006 spinach-associated outbreak isolate indicates candidate genes that may enhance virulence. Infect Immun, 2009; 77:3713–3721.

17.

Kyle

, Parker

, Goudeau

, Brandl

. Transcriptome analysis of Escherichia coli O157:H7 exposed to lysates of lettuce leaves. Appl Environ Microbiol, 2010; 76:1375–1387.

18.

Lai

, Rosenshine

, Leong

, Frankel

. Intimate host attachment: Enteropathogenic and enterohaermorrhagic Escherichia coli . Cell Microbiol, 2013; 15:1796–1808.

19.

, Ibekwe

, Yi

, Wang

, Yamazaki

, Crowley

, Yang

C-H

^a. Persistence of Escherichia coli O157:H7 and its mutants in soils. PLoS ONE, 2011a;6:e23191.

20.

, Zhang

, Doyle

^b. Green fluorescent protein labeling of Listeria, Salmonella, and Escherichia coli O157:H7 for safety-related studies. PLoS ONE, 2011b;6:e18083.

21.

Muyrers

JPP

, Zhang

, Stewart

. Techniques: Recombinogenic engineering—New options for cloning and manipulating DNA. Trends Biochem Sci, 2001; 26:325–331.

22.

O'Brien

, Lively

, Chen

, Rothman

, Formal

. Escherichia coli O157:H7 strains associated with hemorrhagic colitis in the United States produce a Shigella dysenteriae-1 (Shiga) like cytotoxin. Lancet, 1983; 1:702.

23.

O'Brien

, Newland

, Miller

, Holmes

, Smith

, Formal

. Shiga-like toxin-converting phages from Escherichia coli strains that cause hemorrhagic colitis or infantile diarrhea. Science, 1984; 226:694–696.

24.

Perna

, Plunkett

, Burland

, Mau

, Glasner

, Rose

, Mayhew

, Evans

, Gregor

, Kirkpatrick

, Posfai

, Hackett

, Klink

, Boutin

, Shao

, Miller

, Grotbeck

, Davis

, Limk

, Dimalanta

, Potamousis

, Apodaca

, Anantharaman

, Lin

, Yen

, Schwartz

, Welch

, Blattner

. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature, 2001; 409:529–533.

25.

Sambrook

, Fritsch

, Maniatis

. Molecular Cloning: A Laboratory Manual, 2 ^nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989.

26.

Schmitt

, McKee

, O'Brien

. Two copies of Shiga-like toxin II-related genes common in enterohemorrhagic Escherichia coli strains are responsible for the antigenic heterogeneity of the O157:H^- strain E32511. Infect Immun, 1991; 59:1065–1073.

27.

Strockbine

, Marques

LRM

, Newland

, Smith

, Holmes

, O'Brien

. Two toxin-converting phages from Escherichia coli O157:H7 strain 933 encode antigenically distinct toxins with similar biologic activities. Infect Immun, 1986; 53:135–140.

28.

Theofel

, Harris

. Impact of preinoculation culture conditions on the behavior of Escherichia coli O157:H7 inoculated onto romaine lettuce (Lactuca sativa) plants and cut leaf surfaces. J Food Prot, 2009; 72:1553–1559.

29.

Wick

, Qi

, Lacher

, Whittam

. Evolution of genomic content in the stepwise emergence of Escherichia coli O157:H7. J Bacteriol, 2005; 187:1783–1791.

30.

Xicohtencatl-Cortes

, Sánchez Chacón

, Saldaña

, Freer

, Girón

. Interaction of Escherichia coli O157:H7 with leafy green produce. J Food Prot, 2009; 72:1531–1537.

31.

Zhang

, Buchholz

, Muyrers

JPP

, Stewart

. A new logic for DNA engineering using recombination in Escherichia coli . Nat Genet, 1998; 20:123–128.