Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
. 2014 Apr;80(8):2516–2525. doi: 10.1128/AEM.04203-13

Evaluation of a Loop-Mediated Isothermal Amplification Suite for the Rapid, Reliable, and Robust Detection of Shiga Toxin-Producing Escherichia coli in Produce

Fei Wang a,b, Qianru Yang a,c, Yinzhi Qu b, Jianghong Meng b, Beilei Ge a,✉
Editor: D W Schaffner
PMCID: PMC3993192  PMID: 24509927

Abstract

Shiga toxin-producing Escherichia coli (STEC) strains are a leading cause of produce-associated outbreaks in the United States. Rapid, reliable, and robust detection methods are needed to better ensure produce safety. We recently developed a loop-mediated isothermal amplification (LAMP) suite for STEC detection. In this study, the STEC LAMP suite was comprehensively evaluated against real-time quantitative PCR (qPCR) using a large panel of bacterial strains (n = 156) and various produce items (several varieties of lettuce, spinach, and sprouts). To simulate real-world contamination events, produce samples were surface inoculated with a low level (1.2 to 1.8 CFU/25 g) of individual STEC strains belonging to seven serogroups (O26, O45, O103, O111, O121, O145, and O157) and held at 4°C for 48 h before testing. Six DNA extraction methods were also compared using produce enrichment broths. All STEC targets and their subtypes were accurately detected by the LAMP suite. The detection limits were 1 to 20 cells per reaction in pure culture and 105 to 106 CFU per 25 g (i.e., 103 to 104 CFU per g) in produce, except for strains harboring the stx2c, eae-β, and eae-θ subtypes. After 6 to 8 h of enrichment, the LAMP suite achieved accurate detection of low levels of STEC strains of various stx2 and eae subtypes in lettuce and spinach varieties but not in sprouts. A similar trend of detection was observed for qPCR. The PrepMan Ultra sample preparation reagent yielded the best results among the six DNA extraction methods. This research provided a rapid, reliable, and robust method for detecting STEC in produce during routine sampling and testing. The challenge with sprouts detection by both LAMP and qPCR calls for special attention to further analysis.

INTRODUCTION

Shiga toxin-producing Escherichia coli (STEC) strains are zoonotic agents of significant public health concern (1). In the United States, STEC ranked among the top three causes of food-borne disease outbreaks as well as outbreak-associated illnesses, hospitalizations, and deaths from 1998 to 2010 (2, 3). Although E. coli O157:H7 remains the single most common STEC strain (4), the clinical importance of certain non-O157 STEC strains is growing worldwide (5). Recent years of FoodNet data in the United States have consistently shown more laboratory-confirmed infections caused by non-O157 STEC strains than O157 STEC strains (6–8). Currently, the U.S. regulation designates seven STEC serogroups (O26, O45, O103, O111, O121, O145, and O157) as adulterants in raw, nonintact beef products (9).

Produce ranked second only to beef in causing the largest percentage of STEC outbreaks, many of which are large-scale, multistate outbreaks (3). In September 2006, tainted prepackaged spinach triggered an E. coli O157:H7 outbreak, resulting in 205 confirmed illnesses and 3 deaths in 26 states (10). In 2010, a multistate outbreak of STEC O145 infections linked to shredded romaine lettuce from a single processing facility led to 26 confirmed cases in five states (11). The May 2011 massive outbreak of hemolytic-uremic syndrome (HUS; 852 cases) in Germany and several other countries was attributed to a rare STEC serotype, serotype O104:H4, in sprouts (12). Additionally, several recent multistate outbreaks have been caused by E. coli O157:H7 in romaine lettuce, organic spinach/spring mix blend, and ready-to-eat salads and by STEC O26 in clover sprouts (13).

To reduce the incidence of produce-associated outbreaks, a multifaceted approach from farm to table is required. In particular, the industry has drastically increased raw and finished product testing as a tool to better identify contamination risks (14). Nonetheless, STEC detection in produce remains a challenging task (15, 16). Due to the highly perishable nature of produce, a rapid test is critical. Produce items are also diverse and complex, with many harboring assay inhibitors and therefore requiring effective sample preparation and commodity-specific method validation (16). Additionally, pathogens in produce are usually injured cells present at low levels, whereas the normal flora is present at high levels, resulting in the requirement for a highly sensitive and specific assay (15). The need to identify STEC as a group and certain STEC serogroups specifically adds yet another layer of complexity (17).

Owing to their rapidity, specificity, and sensitivity, molecular-based methods, such as PCR and real-time quantitative PCR (qPCR), have gained widespread applications in produce testing (14, 16). Enrichment is commonly used to increase target cell numbers and simultaneously dilute assay inhibitors and the normal flora in produce (15). However, false-positive and false-negative results are observed, and few PCR assays for STEC have been validated on a commodity-specific basis (16). Besides, a sophisticated thermal cycling instrument is indispensable to carry out these nucleic acid amplification tests (NAATs), limiting their wider applications.

Recently, a novel NAAT termed loop-mediated isothermal amplification (LAMP) has emerged as a promising alternative to PCR for pathogen detection (18, 19). LAMP uses four to six specially designed primers and a strand-displacing Bst DNA polymerase to amplify up to 109 copies of target DNA under isothermal conditions (∼65°C) within an hour (19). Since it is isothermal, LAMP can be performed in much simpler instruments, such as a heater or water bath. To date, multiple LAMP assays targeting STEC Shiga toxin genes (stx1 and stx2) have been developed (20–24) and evaluated in food, primarily beef (25–28), as have several others targeting the E. coli O157 rfbE gene (encoding perosamine synthetase) (23, 24, 29–31). Very recently, we developed a suite of LAMP assays for STEC (targeting common virulence genes stx1, stx2, and eae) and the seven adulterant STEC serogroups (targeting the wzx or wzy gene on the respective O-antigen gene clusters) (32, 33). Although it has been reported to be rapid, specific, and sensitive, the LAMP suite has not been evaluated using a large number of STEC strains harboring various stx and eae subtypes or tested with a variety of produce items using conditions mimicking real-world contamination events (e.g., low-level surface inoculation, cold storage). In addition, despite the critical role that sample preparation plays, there is a scarcity of data regarding the effectiveness of DNA extraction methods for STEC detection in produce.

The aims of this study were to comprehensively evaluate the STEC LAMP suite against qPCR using a large panel of bacterial strains and various produce items (varieties of lettuce, spinach, and sprouts) under conditions mimicking real-world contamination events and to examine the effect of DNA extraction methods on assay performance. The qPCR assays tested were recently developed by scientists from the U.S. Department of Agriculture (USDA) (34, 35).

MATERIALS AND METHODS

Bacterial strains and target gene characterization.

A total of 123 E. coli strains representing 41 serogroups and 33 non-E. coli strains (Table 1) were used for specificity testing. Among them, seven STEC strains (underlined in Table 1), one from each of the seven adulterant serogroups, were used for sensitivity testing and spiked-produce experiments. All strains were cultured as described previously (33).

TABLE 1.

Bacterial strainsc used in this study to evaluate the specificity and sensitivity of the LAMP suite and qPCR assays

Strain group Serotype or species Straina stx1 subtype stx2 subtype Intimin subtype Origin Sourceb
STEC strains of target serogroups (n = 83)
    O26 (n = 20) O26:H11 CVM 9935 stx1a β Animal (antelope) FDA, CVM
O26:H11 CVM 9942 stx1a β Animal (cow) FDA, CVM
O26:H11 CVM 9952 stx1a β Animal (pig) FDA, CVM
O26:H11 CVM 9953 stx1a β Animal (pig) FDA, CVM
O26:H11 CVM 9965 stx1a β Animal (cow) FDA, CVM
O26:H11 CVM 9966 stx1a β Animal (cow) FDA, CVM
O26:H11 CVM 9967 stx1a β Animal (cow) FDA, CVM
O26:H11 CVM 9988 stx1a β Animal (cow) FDA, CVM
O26:H11 CVM 9995 stx1a β Human FDA, CVM
O26:H11 CVM 9997 stx1a β Human FDA, CVM
O26:H11 CVM 9998 stx1a β Human FDA, CVM
O26:H11 CVM 9999 stx1a β Human FDA, CVM
O26:H11 CVM 10000 stx1a β Human FDA, CVM
O26:H11 CVM 10001 stx1a β Human FDA, CVM
O26:H11 CVM 10007 stx1a β Human FDA, CVM
O26:H11 CVM 10008 stx1a β Human FDA, CVM
O26:H11 CVM 10112 stx1a β Animal (cow) FDA, CVM
O26:H11 CVM 10128 stx1a β Human FDA, CVM
O26:H11 CVM 10224 stx1a β Human FDA, CVM
O26:H11 SJ3 stx2a β Unknown CDC
    O45 (n = 6) O45:H2 05-6545 stx1a ε Unknown Lab collection
O45:H2 A9619-C2 stx1a ε Unknown Lab collection
O45:H2 EC1467 stx1a ε Human FDA, CFSAN
O45:H2 EC1674 stx1a ε Human FDA, CFSAN
O45:H2 MI4 stx1a ε Unknown Lab collection
O45:H2 SJ9 stx1a stx2a β Unknown CDC
    O103 (n = 20) O103:H2 7828/95 stx1a stx2a ε Human Lab collection
O103:H2 CVM 9260 stx1a ε Animal (deer) FDA, CVM
O103:H2 CVM 9301 stx1a ε Animal (goat) FDA, CVM
O103:H2 CVM 9305 stx1a ε Animal (sheep) FDA, CVM
O103:H2 CVM 9318 stx1a ε Animal (cow) FDA, CVM
O103:H2 CVM 9322 stx1a stx2d ε Animal (cow) FDA, CVM
O103:H2 CVM 9328 stx1a ε Human FDA, CVM
O103:H2 CVM 9380 stx1a ε Human FDA, CVM
O103:H2 CVM 9385 stx1a ε Human FDA, CVM
O103:H2 CVM 9439 stx1a ε Human FDA, CVM
O103:H2 CVM 9440 stx1a ε Human FDA, CVM
O103:H2 CVM 9446 stx1a ε Human FDA, CVM
O103:H2 CVM 9451 stx1a ε Human FDA, CVM
O103:H2 CVM 9453 stx1a ε Human FDA, CVM
O103:H2 PMK-5 stx1a ε Human Lab collection
O103:H11 CVM 9320 stx1a β Animal (cow) FDA, CVM
O103:H11 SJ12 stx1a β Unknown CDC
O103:H25 CVM 9340 stx1a θ Human FDA, CVM
O103:H25 CVM 9353 stx1a θ Animal (cow) FDA, CVM
O103:H25 CVM 9354 stx1a θ Animal (cow) FDA, CVM
    O111 (n = 19) O111:H8 CVM 9467 stx1a θ Animal (cow) FDA, CVM
O111:H8 CVM 9557 stx1a stx2d θ Animal (cow) FDA, CVM
O111:H8 CVM 9574 stx1a stx2a θ Human FDA, CVM
O111:H8 CVM 9603 stx1a θ Human FDA, CVM
O111:H8 CVM 9610 stx1a θ Animal (cow) FDA, CVM
O111:H8 CVM 9614 stx1a θ Animal (cow) FDA, CVM
O111:H8 CVM 9617 stx1a θ Animal (cow) FDA, CVM
O111:H8 CVM 9619 stx1a θ Human FDA, CVM
O111:H11 CVM 9505 stx1a β Animal (cow) FDA, CVM
O111:H11 CVM 9529 stx1a β Animal (cow) FDA, CVM
O111:H11 CVM 9530 stx1a β Animal (pig) FDA, CVM
O111:H11 CVM 9534 stx1a β Animal (cow) FDA, CVM
O111:H11 CVM 9535 stx1a β Animal (cow) FDA, CVM
O111:H11 CVM 9548 stx1a β Animal (cow) FDA, CVM
O111:H11 CVM 9553 stx1a β Animal (cow) FDA, CVM
O111:H11 CVM 9591 stx1a β Animal (cow) FDA, CVM
O111:NM 78/92 stx1a θ Human Lab collection
O111:NM P1340 stx1a stx2a θ Animal (calf) Lab collection
O111:NM SJ13 stx1a stx2a θ Unknown CDC
    O121 (n = 3) O121:H19 ESC0601 stx2a ε Food (lettuce) FDA, CFSAN
O121:NM EC1406 stx2a ε Human FDA, CFSAN
O121:NM EC1670 stx2a ε Unknown FDA, CFSAN
    O145 (n = 5) O145:H28 CVM 9785 stx1a γ1 Animal (Cow) FDA, CVM
O145:H28 CVM 9790 stx2c γ1 Human FDA, CVM
O145:H28 EC1792 stx1a stx2a γ1 Unknown FDA, CFSAN
O145:NM EC1789 stx2a γ1 Human FDA, CFSAN
O145:NM SJ23 stx1a stx2a γ1 Unknown CDC
    O157 (n = 10) O157:H7 85-1 stx1a stx2a γ1 Human Lab collection
O157:H7 B6903 stx1a stx2a γ1 Human Lab collection
O157:H7 C8 stx1a γ1 Animal (sheep) Lab collection
O157:H7 E-0019 stx2a γ1 Animal (calf) Lab collection
O157:H7 E-0122 stx1a stx2a γ1 Animal (calf) Lab collection
O157:H7 E-0342 stx2c γ1 Animal (calf) Lab collection
O157:H7 EC600V stx1a stx2a γ1 Food (steak) Lab collection
O157:H7 MDL 3562 stx2a γ1 Food (produce) Lab collection
O157:H7 OH-495-189 stx2a γ1 Food (beef) Lab collection
O157:H7 W2-2 stx1a stx2a γ1 Food (poultry) Lab collection
Other STEC strains (n = 25) O2:H27 SJ5 stx2a Unknown CDC
O5:NM 3143-85 stx1a β Unknown Lab collection
O5:NM 3812-3 stx1c Animal (sheep) Lab collection
O8:H28 ESC0604 stx2a Food (lettuce) FDA, CFSAN
O15:H16 N5789 stx2g Food (beef) Lab collection
O22:H8 P1330 stx2d Food (beef) Lab collection
O36:H14 ESC0603 stx2g Food (sprouts) FDA, CFSAN
O46:H38 P1332 stx1a stx2d Food (beef) Lab collection
O50:H7 3056-85 stx2a ε Unknown Lab collection
O55:H7 5906 stx2d γ1 Unknown Lab collection
O73:H18 ESC0608 stx2a Food (spinach) FDA, CFSAN
O83:H8 N11682 stx2d Food (beef) Lab collection
O88:H49 P1333 stx2a Food (meat) Lab collection
O91:H21 P1334 stx1a stx2d Animal (cow) Lab collection
O104:H21 94-3024 stx2a Unknown Lab collection
O113:H21 ESC0615 stx2a Food (spinach) FDA, CFSAN
O116:H21 ESC0609 stx2d Food (spinach) FDA, CFSAN
O125:NM 3153-86 stx1a β Unknown Lab collection
O126:H8 78-4084 stx1a Unknown Lab collection
O128:H16 CVM 9652 stx1a Animal (okapi) FDA, CVM
O146:H21 90–3158 stx1a Unknown Lab collection
O168:H8 ESC0613 stx2a Food (spinach) FDA, CFSAN
O174:H36 ESC0602 stx2d Food (lettuce) FDA, CFSAN
ONT:H7 N5545 stx2d Food (beef) Lab collection
OX25 ESC0606 stx2d Food (spinach) FDA, CFSAN
Other E. coli strains (n = 15) O1:K1:H7 U5-41 Unknown Lab collection
O3:K2ab:H2 U414-41 Unknown Lab collection
O9:K103 987 Unknown Lab collection
O18:K1:H7 88-766 Unknown Lab collection
O44:K74:H18 H702c Unknown Lab collection
O77:K96:NM E10 Unknown Lab collection
O78:H11 EC463 Unknown Lab collection
O86:H25 H35A Unknown Lab collection
O111:H11 CVM 9515 β Animal (cow) FDA, CVM
O118:H16 P1341 Animal (calf) Lab collection
O124:NM EC230 Unknown Lab collection
O145 CVM 9818 γ1 Animal (cow) FDA, CVM
O145:NM EC1790 Unknown FDA, CFSAN
O145:NM EC1793 Unknown FDA, CFSAN
O157:H7 G-13 γ1 Animal (sheep) Lab collection
Other bacteria (n = 33)
    Campylobacter coli ATCC 33559 Animal (pig) Lab collection
    Campylobacter jejuni ATCC 43430 Animal (calf) Lab collection
    Campylobacter lari ATCC 35222 Animal (dog) Lab collection
    Citrobacter freundii 10053 Unknown Lab collection
    Enterobacter cloacae 95MV2 Human Lab collection
    Hafnia alvei ATCC 23280 Unknown Lab collection
    Listeria grayi ATCC 19120 Animal Lab collection
    Listeria innocua ATCC 33090 Animal (cow) Lab collection
    Listeria ivanovii ATCC 19119 Animal (sheep) Lab collection
    Listeria monocytogenes ATCC 15313 Animal (rabbit) Lab collection
    Listeria seeligeri UMD 489 Unknown Lab collection
    Listeria welshimeri ATCC 35897 Plant Lab collection
    Salmonella enterica serovar Braenderup FDA71 Unknown FDA, CFSAN
    S. enterica serovar Enteritidis SE 9 Unknown FDA, CFSAN
    S. enterica serovar Heidelberg 63B2 Unknown FDA, CFSAN
    S. enterica serovar Javiana 7 N Unknown FDA, CFSAN
    S. enterica serovar Mbandaka 37 N Food (candy) FDA, CFSAN
    S. enterica serovar Montevideo 1 H Food (whole eggs) FDA, CFSAN
    S. enterica serovar Muenchen 1501 H Feed (feather meal) FDA, CFSAN
    S. enterica serovar Newport FDA197 Unknown FDA, CFSAN
    S. enterica serovar Oranienburg 1410 H Feed (feather meal) FDA, CFSAN
    S. enterica serovar Poona 2861 H Animal (turtle) FDA, CFSAN
    S. enterica serovar Stanley 1243 H Feed (bonemeal) FDA, CFSAN
    S. enterica serovar Typhimurium ATCC 43971 Unknown FDA, CFSAN
    Shigella boydii RH12-23-11 Unknown Lab collection
    Shigella dysenteriae RH12-25-9 Unknown Lab collection
    Shigella flexneri G45 Unknown Lab collection
    Vibrio aestuarianus ATCC 35048 Animal (oyster) Lab collection
    Vibrio cholerae 15937-E6 Unknown Lab collection
    Vibrio harveyi ATCC 35084 Animal (shark) Lab collection
    Vibrio mimicus ATCC 33655 Human Lab collection
    Vibrio parahaemolyticus ATCC 27969 Animal (crab) Lab collection
    Vibrio vulnificus 515-4C2 Animal (oyster) Lab collection
Open in a new tab
a

The seven underlined strains were used for the evaluation of specificity and sensitivity and spiked-produce experiments, while the others were used only for the specificity testing.

b

CDC, Centers for Disease Control and Prevention; FDA, CFSAN, U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition; FDA, CVM, U.S. Food and Drug Administration, Center for Veterinary Medicine; Lab collection, the strain collection maintained at the University of Maryland, College Park, MD.

c

A total of 156 strains were tested.

E. coli strains were examined for the presence of STEC virulence genes (stx1, stx2, and eae) by PCR, followed by restriction fragment length polymorphism (RFLP) analysis to determine their respective gene subtypes (36, 37).

LAMP.

A LAMP suite of 10 assays recently developed by our research group (32, 33) was evaluated. The first three assays targeted common STEC virulence genes (stx1, stx2, and eae), while the remaining seven each targeted a gene (wzx or wzy) on the O-antigen gene clusters of the seven adulterant STEC serogroups. Each LAMP assay employed five to six specially designed LAMP primers (see Table S1 in the supplemental material), two outer primers, two inner primers, and one or two loop primers that recognized specific regions of the target DNA sequences.

The assays were performed as described previously (32, 33). Briefly, the LAMP reagent mix (25 μl) contained 1× ThermoPol reaction buffer (New England BioLabs, Ipswich, MA), 6 mM MgSO4, 1.2 mM each deoxynucleoside triphosphate (dNTP), 0.1 μM each outer primer (Integrated DNA Technologies, Coralville, IA), 1.8 μM each inner primer, 1 μM each loop primer, 10 U of Bst DNA polymerase (New England BioLabs), and 2 μl of template DNA. All LAMP reactions were carried out at 65°C (63°C for the O157 LAMP) for 1 h and terminated at 80°C for 5 min in an LA-320C real-time turbidimeter (Eiken Chemical Co., Ltd., Tokyo, Japan). Turbidity readings at 650 nm were obtained every 6 s, and time threshold (Tt) values (in minutes) were determined when the turbidity increase measurements (differential values of the moving average of the turbidity) exceeded 0.1.

qPCR.

As a comparison, qPCR assays developed by USDA scientists (34, 35) were performed. Similarly, 10 sets of primers/probes were used; 3 targeted common STEC virulence genes (stx1, stx2, and eae) and 7 targeted the wzx or wzy gene on the O-antigen gene clusters of the seven adulterant STEC serogroups.

The qPCR assays were carried out as described previously (32, 33). The reagent mix (25 μl) consisted of 1× PCR buffer, 4 mM MgCl2, 0.2 mM each dNTP, 0.25 μM each primer (see Table S1 in the supplemental material), 0.1875 μM probe, 1.5 U of GoTaq Hot Start polymerase (Promega, Madison, WI), and 2 μl of template DNA. The qPCRs were conducted using 40 cycles of denaturation at 94°C for 20 s, annealing at 60°C for 30 s, and extension at 72°C for 50 s in a SmartCycler II system (Cepheid, Sunnyvale, CA). Fluorescence readings were acquired using the 6-carboxyfluorescein (FAM) channel, and cycle threshold (CT) values (in number of cycles, with approximately 2 min per cycle) were recorded when the fluorescence readings reached 30 units.

Specificity and sensitivity tests.

For assay specificity, DNA templates of the 156 bacterial strains (Table 1) were prepared by heating at 95°C for 10 min. Aliquots (2 μl) were subjected to LAMP and qPCR, and assays were repeated twice.

Assay sensitivity (limit of detection) was determined by using 10-fold serial dilutions of the seven STEC strains of the adulterant serogroups. DNA templates were prepared from stationary-phase cultures as described previously (32). Aliquots (2 μl) were tested by LAMP and qPCR, and assays were repeated three times.

Assay evaluation with produce with high-level inoculation (assay sensitivity for detection of STEC in produce).

Eight produce items, including varieties of lettuce (iceberg, romaine), spinach (baby, savoy, semisavoy), and sprouts (alfalfa, clover, mung bean), were obtained from a local grocery store and analyzed within 2 h of collection. Briefly, lettuce and spinach leaves were cut into 4-cm2 pieces using sterile scissors, and 25-g samples were weighed out. Sprouts were also divided into 25-g analysis portions.

To determine assay sensitivity for detection of STEC in produce, for each produce item, 35 samples (one sample per strain [n = 7] per inoculation level [n = 5]) were inoculated and 3 samples were included as uninoculated controls. Spot inoculation on the produce surface was performed as described previously (38). Briefly, 1.5 ml of overnight STEC cultures 10-fold serially diluted from 105 to 109 CFU/ml was added to each 25-g test sample. For lettuce and spinach, the inoculum was equally divided among the number of leaf pieces. Sprout samples were grouped into three equal portions, and 500 μl of the inoculum was added onto the surface of each portion. Aerobic plate counts were determined for the uninoculated controls (n = 3) by the standard pour plate method. All samples were air dried in a laminar flow biosafety cabinet for 2 h, followed by storage at 4°C for 48 h.

After cold storage, each sample was homogenized with 225 ml of buffered peptone water (BD Diagnostic Systems, Sparks, MD) for 1 min in a food stomacher (model 400; Seward, Cincinnati, OH). Aliquots (1 ml) of each homogenate were centrifuged at 16,000 × g for 3 min, and pellets were suspended in 100 μl of PrepMan Ultra sample preparation reagent (Applied Biosystems, Foster City, CA). The mixtures were heated at 95°C for 10 min and centrifuged again at 12,000 × g for 2 min. The supernatants (2 μl) were tested by both LAMP and qPCR, and assays were repeated three times.

Assay evaluation with produce with low-level inoculation.

For assay evaluation with produce with low-level inoculation, the same eight produce items described above were used and similar inoculation, processing, and testing procedures were followed, with three major exceptions: inoculation level, enrichment, and replication scheme. The inoculation level applied was 100 CFU/25 g, with two additional ones (102 and 103 CFU/25 g) being used for sprouts only. After surface inoculation and cold storage, the samples were first incubated at 42°C for up to 24 h, and aliquots (1 ml) were removed at 6, 8, 10, and 24 h for processing by use of the PrepMan Ultra reagent and testing by LAMP and qPCR as described above. The experiments were independently repeated three times with different produce samples. In total, there were 21 inoculated samples (one sample per strain [n = 7] in three repeats [n = 3]) and 3 uninoculated controls tested per produce item. The number of inoculated samples was tripled for the sprouts varieties, i.e., 63 samples were tested, due to the two additional inoculation levels tested.

Comparison of DNA extraction methods.

Six DNA extraction methods were evaluated using the same eight produce items described above spiked with 1.2 to 1.8 CFU/25 g (or 1.2 × 102 to 1.8 × 102 CFU/25 g for sprouts varieties) of E. coli O157:H7 strain MDL 3562 and enriched for various periods (6, 8, 10, 24 h). For each produce item, experiments were independently repeated three times with different samples. These methods were (i) testing of raw enrichment broth, i.e., direct testing without any DNA preparation step; (ii) boiling preparation at 95°C for 10 min; (iii) two-step centrifugation by first centrifuging at 500 × g for 1 min to remove produce tissues and then centrifuging again at 16,000 × g for 5 min to pellet bacterial cells and resuspending the pellet in 100 μl of TE (Tris-EDTA) buffer; (iv) two-step centrifugation followed by boiling, i.e., heating the bacterial suspension in TE buffer prepared by the two-step centrifugation method at 95°C for 10 min; (v) the method with the PrepMan Ultra reagent, as describe above; and (vi) the FTA Elute method, which, briefly, consisted of adding 65 μl of enrichment broths onto an FTA card (Whatman Inc., Florham Park, NJ), punching out two 2-mm disks after absorption, allowing the disks to dry, washing the disks with sterile water, and then heating with water at 95°C for 30 min. Aliquots (2 μl) of DNA templates prepared by all six methods were tested by eae LAMP and eae qPCR.

Data analysis.

Means and standard deviations of Tt values for LAMP and CT values for qPCR were calculated by use of the Microsoft Excel program (Seattle, WA). The Tt and CT values sorted by assay target, gene subtype, produce type, enrichment time, and DNA extraction method were compared by using analysis of variance (SAS for Windows, version 9.2; SAS Institute, Cary, NC). Differences between the mean values were considered significant when P was <0.05.

RESULTS

Assay specificity.

Among 156 bacterial strains (Table 1) tested by the STEC LAMP suite, false-positive or false-negative results were not observed; i.e., LAMP results matched 100% with known strain characteristics for all of the target genes. The overall mean Tt values ranged from 16.1 min for eae LAMP to 24.3 min for stx2 LAMP (data not shown). Among the three assays targeting STEC virulence genes, stx1 and eae LAMPs yielded significantly lower Tt values than stx2 LAMP, while O26 and O121 LAMPs had the lowest Tt values among the seven serogroup-specific assays (P < 0.05) (data not shown). The Tt values also differed among strains carrying various stx2 and eae subtypes. Notably, the stx2 LAMP proceeded the slowest in strains harboring stx2g (mean Tt, 30.6 min), followed by LAMPs with stx2c (24.7 min), stx2d (24.4 min), and stx2a (23.7 min), while by the eae LAMP, strains containing eae-γ1 had Tt values (11.6 min) significantly shorter than those for strains containing eae-β (16.9 min), eae-ε (17 min), or eae-θ (18.4 min) (P < 0.05).

For qPCR, the vast majority of specificity tests accurately detected the 156 strains, with the overall mean CT values ranging from 12.8 cycles for the eae qPCR to 20 cycles for the O103 qPCR (data not shown). However, false-negative results were consistently generated by the stx2 qPCR for two strains (ESC0603 and N5789) carrying the stx2g gene. In contrast, the CT values did not vary significantly among strains possessing different stx2 or eae subtypes (P > 0.05; data not shown).

Assay sensitivity.

Table 2 summarizes LAMP and qPCR sensitivity when testing 10-fold serial dilutions of STEC strains belonging to the seven adulterant serogroups. In pure culture testing, the LAMP suite consistently detected 10 to 20 CFU/reaction of target STEC strains. In one or two out of three repeats, stx2 and O45 LAMPs were capable of detecting the respective STEC strains at an even lower concentration, i.e., 1 CFU per reaction. However, the stx2 and eae LAMP assays were less sensitive (up to 100-fold) for strains carrying certain target gene subtypes, e.g., stx2c, eae-β, and eae-θ. The detection limits for qPCR assays consistently fell between 1 and 20 CFU per reaction for all assay targets and their subtypes (Table 2).

TABLE 2.

Sensitivity of the LAMP suite and qPCR assays in pure culture and with spiked producek

Assay target Target gene and subtype Strain Detection limit (CFU/reaction in pure culture or CFU/25 g in spiked produce)
Pure culture
 Lettuce
 Spinach
 Sprouts
LAMP qPCR LAMP qPCR LAMP qPCR LAMP qPCR
Stx1 stx1a SJ9 10 10 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105-106g 1.4 × 105
Stx2 stx2a SJ9 1–10b 10 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105
Stx2 stx2c CVM 9790 1 × 103 1–10b 1.7 × 107 1.7 × 105 1.7 × 107 1.7 × 105 1.7 × 107 1.7 × 105
Stx2 stx2d CVM 9322 14 14 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105-106h 1.5 × 105-106g
Intimin eae-β SJ9 1 × 102 10 1.4 × 106-107c 1.4 × 105 1.4 × 107 1.4 × 105-106e 1.4 × 107 1.4 × 105-106i
Intimin eae-γ1 CVM 9790 10 1–10b 1.7 × 105 1.7 × 105 1.7 × 105 1.7 × 105 1.7 × 105 1.7 × 105
Intimin eae-ε CVM 9322 14 14 1.5 × 105 1.5× 105 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105
Intimin eae-θ SJ13 2 × 102 2–20a 1.4 × 108 1.4 × 105 1.4 × 108 1.4 × 105-106f 1.4 × 108 1.4 × 105-106j
O26 wzy (O26) SJ3 12 1.2–12a 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105
O45 wzy (O45) SJ9 1–10a 1–10b 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105
O103 wzx (O103) CVM 9322 14 14 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105 1.5 × 105
O111 wzy (O111) SJ13 20 2 1.4 × 105 1.4 × 105 1.4 × 105-106d 1.4 × 105 1.4 × 105-106g 1.4 × 105
O121 wzy (O121) ESC0601 20 20 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105
O145 wzx (O145) CVM 9790 10 10 1.7 × 105 1.7 × 105 1.7 × 105 1.7 × 105 1.7 × 105-106g 1.7 × 105
O157 wzy (O157) MDL 3562 12 1.2–12a 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105 1.4 × 105
Open in a new tab
a

One out of three repeats was positive for the lower detection limit.

b

Two out of three repeats were positive for the lower detection limit.

c

Iceberg lettuce had the lower detection limit.

d

Baby and savoy spinach had the lower detection limit.

e

Baby spinach had the lower detection limit.

f

Savoy and semisavoy spinach had the lower detection limit.

g

Clover and mung bean sprouts had the lower detection limit.

h

Clover sprouts had the lower detection limit.

i

Alfalfa and clover sprouts had the lower detection limit.

j

Alfalfa and mung bean sprouts had the lower detection limit.

k

In produce testing, 1.5 ml of 10-fold serially diluted overnight STEC cultures (prepared at different times from those used in the pure culture testing) was added to each 25-g produce sample.

Assay sensitivity in spiked produce is also summarized in Table 2. For the uninoculated controls, aerobic plate counts averaged 104 to 105 CFU/g in lettuce and spinach varieties and 106 to 108 CFU/g in sprouts varieties, and all target genes tested negative by LAMP and qPCR (data not shown). In lettuce and spinach varieties, the LAMP suite detected down to 105 CFU per 25 g produce (approximately 103 CFU/g, equivalent to 2 CFU per reaction). In alfalfa (but not clover and mung bean) sprouts, 10-fold higher cell concentrations (106 CFU/25 g) were needed for accurate detection by the stx1, O111, and O145 LAMPs. Regardless of produce type or variety, the stx2 and eae LAMPs were less sensitive (up to 1,000-fold) for strains containing stx2c, eae-β, and eae-θ subtypes. In comparison, all of the qPCR assays detected down to 105 CFU/25 g in lettuce, spinach, and sprouts varieties. Small variations in detection limit (10-fold) among varieties of the same produce type were observed by either LAMP or qPCR (Table 2).

Effect of DNA extraction methods.

Table 3 compares the effects of six DNA extraction methods on detection of E. coli O157:H7 by eae LAMP and eae qPCR in baby spinach samples spiked with 1.2 to 1.8 CFU/25 g of E. coli O157:H7 strain MDL 3562 (eae-γ1). Positive LAMP results were obtained by all six methods but at different enrichment times. For samples enriched for 6 h, the method with the PrepMan Ultra sample preparation reagent was the only one that consistently gave positive LAMP results. The methods with FTA Elute and raw enrichment broth required 8 and 10 h of enrichment, respectively, to generate positive LAMP results. For qPCR, 8 h of enrichment was needed for the majority of methods and 10 h was needed for the two-step centrifugation method. Boiling facilitated LAMP and qPCR detection, as evidenced by lower Tt and CT values in boiled samples and/or the shorter enrichment time needed for detection. Among all six methods, the method with FTA Elute tended to generate the highest Tt or CT values at each enrichment period in baby spinach (Table 3).

TABLE 3.

Effect of six DNA extraction methods on eae LAMP and eae qPCR detection of STEC in baby spinach samples spiked with 1.2 to 1.8 CFU/25 g of Escherichia coli O157:H7 strain MDL 3562 (eae-γ1) and tested after cold storage and various enrichment periodsc

Method Avg LAMP Tt (min) after enrichment for:
 Avg qPCR CT (no. of cycles) after enrichment for:
6 h 8 h 10 h 24 h 6 h 8 h 10 h 24 h
Raw enrichment broth 18.4 ± 1.9AB 14.7 ± 0.2A 29.1 ± 0.8B 25.0 ± 0.4C 23.6 ± 0.2B
Boiling preparation 24.5a 16.4 ± 0.6BC 13.0 ± 0.3C 13.0 ± 0.4B 32.5 ± 0.9 27.1 ± 0.4C 25.0 ± 0.8C 19.2 ± 0.6D
Two-step centrifugation 22.2a 18.7 ± 1.0B 18.2 ± 0.4B 13.5 ± 0.2B 30.5 ± 0.1B 21.4 ± 0.6C
Two-step centrifugation followed by boiling 20.6 ± 1.2b 14.8 ± 0.7C 12.5 ± 0.5C 12.2 ± 0.5C 27.9 ± 0.2BC 21.1 ± 0.4D 19.4 ± 0.3D
PrepMan Ultra reagent 21.7 ± 1.9 15.6 ± 1.0C 13.0 ± 0.3C 13.0 ± 0.2B 33.47a 27.1 ± 1.1C 20.5 ± 1.3D 19.6 ± 0.3D
FTA Elute 21.6 ± 3.3A 19.7 ± 0.4A 13.1 ± 0.7B 35.7 ± 0.7A 34.1 ± 0.6A 24.6 ± 0.1A
Open in a new tab
a

One out of three repeats was detected by LAMP or qPCR.

b

Two out of three repeats were detected by LAMP.

c

In each column within the data for LAMP or qPCR, Tt or CT values followed by different uppercase letters are significantly different (P < 0.05).

In other spinach varieties, the minimum enrichment time required for LAMP detection was 6 h (8 h for qPCR) when the method with two-step centrifugation followed by boiling or the PrepMan Ultra sample preparation reagent was used (data not shown). However, at least 10 h of enrichment was required for both LAMP and qPCR when using FTA Elute and at least 24 h of enrichment was required for the other three methods (raw enrichment broth, boiling preparation, and two-step centrifugation). In two lettuce varieties, regardless of the DNA extraction method, samples enriched for 6 to 8 h were accurately detected by LAMP and qPCR (data not shown). In three sprouts varieties spiked with 1.2 × 102 to 1.8 × 102 CFU/25 g of STEC cells, 6 to 8 h of enrichment was sufficient for LAMP and qPCR detection, except for alfalfa and clover sprouts by FTA Elute, which needed 10 h (data not shown).

Rapid detection of low levels of STEC in spiked produce.

Table 4 summarizes the LAMP and qPCR results in baby spinach samples spiked with 1.2 to 1.8 CFU/25 g of STEC strains and tested after cold storage and various enrichment periods. After 6 h of enrichment, all LAMP assays in the suite successfully detected such low levels of STEC in baby spinach, except for the stx1 and O157 LAMPs and the stx2 and eae LAMPs in strains carrying stx2c and eae-β, respectively. Significantly higher Tt values were observed for samples enriched for 6 or 8 h than those enriched for longer periods (P < 0.05). A similar trend of detection was observed for qPCR, although the O45 qPCR rather than the stx1 qPCR failed to detect STEC with the 6-h enrichment period (Table 4).

TABLE 4.

Comparison of the LAMP suite and qPCR assays with baby spinach samples spiked with 1.2 to 1.8 CFU/25 g of STEC strains and tested after cold storage and various enrichment periodsc

Assay target Target gene and subtype Strain Avg LAMP Tt (min) after enrichment for:
 Avg qPCR CT (no. of cycles) after enrichment for:
6 h 8 h 10 h 24 h 6 h 8 h 10 h 24 h
Stx1 stx1a SJ9 24.6 ± 2.3A 19.9 ± 1.0B 17.1 ± 0.6B 33.9 ± 0.7A 29.8 ± 0.5B 25.8 ± 0.9C 19.1 ± 0.9D
Stx2 stx2a SJ9 39.3a 33.4 ± 0.8A 28.7 ± 0.6B 26.0 ± 1.0C 34.47a 30.8 ± 0.4A 26.4 ± 0.4B 18.9 ± 0.4C
Stx2 stx2c CVM 9790 34 ± 0.6b 31.0 ± 1.3A 29.0 ± 1.6A 28.8 ± 0.3A 23.8 ± 1.0B 21.1 ± 0.6C
Stx2 stx2d CVM 9322 34.0 ± 1.2A 30.5 ± 0.6B 27.9 ± 0.6C 27.0 ± 0.8C 32.8 ± 0.5A 29.1 ± 0.4B 24.3 ± 0.6C 22.7 ± 0.4D
Intimin eae-β SJ9 27.4 ± 1.1A 23.1 ± 0.8B 19.4 ± 0.3C 30.8 ± 0.5A 26.6 ± 0.9B 19.6 ± 0.4C
Intimin eae-γ1 CVM 9790 16.7 ± 0.2A 14.1 ± 0.8B 12.5 ± 0.3C 12.0 ± 0.3C 31.9 ± 0.4A 27.8 ± 0.6B 23.7 ± 1.1C 21.6 ± 0.6D
Intimin eae-ε CVM 9322 25.4 ± 0.5A 20.6 ± 0.3B 17.2 ± 0.6C 15.7 ± 0.4D 32.4 ± 0.6A 28.3 ± 0.4B 23.7 ± 0.8C 22.0 ± 0.6D
Intimin eae-θ SJ13 25.8 ± 1.4A 23.3 ± 0.3B 21.8 ± 0.4B 22.0 ± 1.4B 29.4 ± 0.5A 26.7 ± 0.9B 25.1 ± 0.8C 25.0 ± 0.1C
O26 wzy (O26) SJ3 28.6 ± 1.7A 22.0 ± 1.1B 19.8 ± 0.7C 18.5 ± 0.5C 31.1 ± 0.4A 24.6 ± 0.1B 20.6 ± 0.3C 17.0 ± 0.8D
O45 wzy (O45) SJ9 28.4 ± 2.6A 22.9 ± 2.6B 20.3 ± 1.7B 18.8 ± 1.4C 30.5 ± 0.7A 25.9 ± 0.5B 19.0 ± 0.3C
O103 wzx (O103) CVM 9322 37.0 ± 3.9A 30.6 ± 2.5B 26.5 ± 1.8B 25.5 ± 0.8C 32.3 ± 0.7A 28.0 ± 0.6B 23.3 ± 0.4C 21.9 ± 0.4D
O111 wzy (O111) SJ13 28.3 ± 1.2A 24.7 ± 0.4B 24.2 ± 1.0B 24.2 ± 1.8B 29.8 ± 0.8A 26.7 ± 1.1B 24.7 ± 1.1C 24.4 ± 0.2C
O121 wzy (O121) ESC0601 25.5 ± 1.0A 21.8 ± 0.8B 19.7 ± 0.7C 19.6 ± 0.7C 32.5 ± 0.5A 28.1 ± 0.9B 25.5 ± 0.1C 24.5 ± 0.7C
O145 wzx (O145) CVM 9790 35.5 ± 2.8A 28.8 ± 1.5B 24.0 ± 1.2C 22.5 ± 1.9C 31.8 ± 0.8A 27.5 ± 1.9B 21.8 ± 1.9C 18.7 ± 0.8D
O157 wzy (O157) MDL 3562 31.9 ± 3.0A 25.9 ± 5.3B 25.1 ± 3.3B 28.4 ± 0.9A 20.4 ± 1.9B 19.2 ± 1.6C
Open in a new tab
a

One out of three repeats was detected by LAMP or qPCR.

b

Two out of three repeats were detected by LAMP.

c

In each row within the data for LAMP or qPCR, Tt or CT values followed by different uppercase letters are significantly different (P < 0.05).

In other spinach and lettuce varieties, both LAMP and qPCR achieved accurate detection after 6 to 8 h of enrichment, except for stx2 LAMP in savoy spinach spiked with CVM 9790 (stx2c), which required 10 h of enrichment (data not shown). In contrast, neither LAMP nor qPCR detected such low levels (1.2 to 1.8 CFU/25 g) of STEC strains in sprouts varieties even after 24 h of enrichment (data not shown). The 102-CFU/25 g inoculum resulted in positive LAMP and qPCR results in sprouts after 6 to 8 h of enrichment for 10 target gene and subtype combinations (stx1a, stx2a, stx2c, stx2d, eae-β, eae-γ1, O26, O45, O103, and O145). All 15 targets/subtypes were detected in sprouts samples spiked with 103 CFU/25 g after 6 to 8 h of enrichment (data not shown).

DISCUSSION

Upon comprehensive evaluation using 156 bacterial strains and eight produce items, the STEC LAMP suite was demonstrated to be rapid (10 to 45 min), reliable (no false-positive or false-negative results), and robust (under conditions mimicking real-world contamination events, e.g., surface contamination, cold storage). Coupled with an effective DNA extraction method, the assays accurately detected low levels (1.2 to 1.8 CFU/25 g) of STEC in lettuce and spinach varieties (but not sprouts) after 6 to 8 h of enrichment. A similar trend of detection was observed for qPCR. This is the first study comparing LAMP and qPCR for detecting STEC strains with various stx and eae subtypes in multiple types of high-risk produce using conditions close to those from real-world applications.

Few studies have closely examined the ability of PCR and qPCR to detect STEC stx or eae subtypes (39–41), and to our knowledge, only one recent study explored the ability of LAMP to detect eae variants (42). There are currently 3 stx1 subtypes (stx1a, stx1c, and stx1d), 7 stx2 subtypes (a through g), and about 30 eae subtypes (α, β, γ, ε, etc.) (37, 43). The presence of eae subtypes (primarily β, γ, and θ) and certain closely related stx2 subtypes (stx2a, stx2c, and stx2d) is strongly associated with severe human illness, such as HUS (36, 43). In a recent produce survey of STEC, the stx1a and stx2a subtypes were the most common, followed by stx2d and stx2c, and only 2 to 3% of strains had the stx2e and stx2g subtypes, while the eae subtypes present were β, γ, and ε (43). The LAMP suite was capable of detecting all of the major target gene subtypes evaluated in this study, consisting of two stx1 subtypes (stx1a and stx1c), four stx2 subtypes (stx2a, stx2c, stx2d, and stx2g), and four eae subtypes (β, γ, ε, and θ). The ability to detect all 27 eae variants tested was also reported in a recent eae LAMP study (42). Such broad specificity is not unexpected, since both eae LAMP assays employed primers in the highly conserved N-terminal region (670 amino acids) of all intimin subtypes (36). In contrast, the eae qPCR primers were located outside this region, increasing the assay's vulnerability for false-negative results. Due to the inability of their qPCR assays to detect some eae and stx subtypes (stx1d, stx2e, stx2f, and stx2g), the current USDA Microbiology Laboratory Guidebook 5B has adopted eae and stx primers/probes developed elsewhere, modified the O145 probe, and included a commercially available qPCR (BAX system; Dupont Qualicon) as an alternative method (44). Other studies also reported the failure of some PCR/qPCR assays to detect genetically distant stx subtypes, e.g., stx2b, stx2f, and stx2g (40, 45). A recent evaluation of seven commercially available qPCR systems for stx subtypes also returned variable results (41). Therefore, careful selection and evaluation of primers/probes are critical in developing these NAATs.

Besides the inability of the stx2 qPCR to detect two strains containing stx2g, the specificities of LAMP and qPCR did not obviously differ for the 156 strains tested. Upon initial development, LAMP assays targeting STEC virulence genes (32) and seven O serogroups (33) were shown to be 100% specific by testing 90 and 120 strains, respectively. However, the specificity of qPCR assays evaluated in this study was not reported when initially developed (35), except that the O157 qPCR targeting the wzy gene was reported to be 100% specific (34). Such high specificity also agreed with that for other LAMP assays for STEC and E. coli O157 (20, 24, 29, 30). Interestingly, significant differences in LAMP Tt values among virulence gene-based assays or serogroup-specific assays or among strains possessing different stx2 or eae subtypes were observed in this study, suggesting variations in assay amplification efficiency. Despite this, all Tt values observed in positive LAMP assays were below 31 min, indicating robust detection (46).

An entirely new set of seven STEC strains was selected to evaluate assay sensitivity. The detection limits of the LAMP suite (1 to 20 CFU per reaction in pure culture) closely mimicked those in previous studies (32, 33) and also fell within the range (0.7 to 100 CFU per reaction) reported for other STEC LAMP assays (20–24, 29–31). One important finding in the present study is that the stx2 and eae LAMPs were less sensitive (up to 100-fold) in strains containing stx2c, eae-β, and eae-θ subtypes, likely due to mismatches on LAMP primer sequences between these subtypes and those used as prototypes for primer design, i.e., sequences with GenBank accession numbers X07865 (stx2a) and Z11541 (eae-γ1) (32). Previously, eae LAMP showed inferior sensitivity in STEC strains 97-3250 and 3215-99 (32), which have now been confirmed by PCR-RFLP, similar to other strains used in this study, to harbor the eae-β and eae-θ subtypes, respectively (data not shown). Conversely, qPCR assays were capable of detecting various stx2 and eae subtypes without compromised sensitivity, possibly due to the use of fewer pairs of primers in qPCR.

Lettuce, spinach, and sprouts were tested since they have historically been involved in STEC outbreaks (10–13). The recent produce survey also identified spinach, lettuce, and cilantro (not sprouts) to be most problematic in STEC contamination (43). The seven serogroup-specific LAMP assays have been evaluated in baby spinach and romaine lettuce (33) but not virulence-based assays (32). In previous studies, inoculation usually occurred in produce homogenates rather than on the surface of intact produce, and samples were tested directly without cold storage (33). In the present study, with surface inoculation and aging treatment, the LAMP suite achieved robust and accurate detection of 105 to 106 CFU/25 g (i.e., 103 to 104 CFU/g, equivalent to 2 to 20 CFU per reaction) of STEC in all produce varieties without enrichment, which was comparable to the detection limits reported previously (33). Consistent with pure culture sensitivity data, inferior sensitivity was also observed when testing strains with certain target gene subtypes (stx2c, eae-β, and eae-θ) in produce. The small variations (10-fold) observed among varieties of the same produce type indicated the effect of produce variety on sample processing and downstream assay, underscoring the need for commodity-specific evaluation.

With 6 to 8 h of enrichment, the LAMP suite detected 1.2 to 1.8 CFU/25 g of STEC in lettuce and spinach varieties, even among strains containing target gene subtypes (stx2c, eae-β, and eae-θ) for which assay sensitivity was inferior. Similar detection capabilities using PCR or qPCR for STEC in leafy greens were obtained, usually with 24 h of enrichment (47, 48). In the present study, for samples enriched for longer than 8 h, more stable (i.e., statistically insignificant) and shorter Tt values were observed. In contrast, enrichment periods longer than 6 h were found to be disadvantageous for the recovery of STEC from salad samples with low levels of contamination (49). The temperature difference used for enrichment (37°C versus 42°C) may account for the discrepancy. In sprouts varieties, both LAMP and qPCR failed to detect this low level (1.2 to 1.8 CFU/25 g) of STEC with up to 24 h of enrichment. Previous studies in Japan applying LAMP in alfalfa and radish sprouts reported 90 to 100% detection rates in samples contaminated with 1 to 20.8 CFU/25 g of STEC after overnight enrichment (20, 25, 28). A multiplex PCR was also recently shown to detect 7 to 58 CFU/25 g of STEC in alfalfa, leek, and soy sprouts (50). It is noteworthy that sprouts samples had rather abundant natural flora, which was at levels 2 to 3 log units higher than those in leafy greens. Further studies are needed to examine whether this normal flora in sprouts or natural compounds released from sprouts during processing may affect STEC survival during enrichment. If necessary, procedures such as acid treatment or immunomagnetic separation may be incorporated to enhance STEC detection in sprouts (51).

DNA extraction is a critical step in molecular-based NAATs. The PrepMan Ultra sample preparation reagent yielded the best results, but the other five methods also generated satisfactory results for LAMP and qPCR with samples enriched for up to 24 h. Previously, the PrepMan Ultra reagent has been widely used to extract STEC DNA from a wide range of food samples, including fresh produce (52). FTA Elute had the advantage of preserving sample DNAs for up to 2 years and without centrifugation steps; however, the final DNA amount extracted was approximately 100-fold less concentrated than that obtained by the other methods, thus requiring more cells in the enrichment broth or a prolonged enrichment time.

In conclusion, the LAMP suite was demonstrated to be a rapid, reliable, and robust method for detecting STEC in produce during routine sampling and testing, providing a valuable tool for the produce industry and regulatory agencies to better identify contamination risks and ensure produce safety, therefore reducing produce-related STEC outbreaks and illnesses.

Supplementary Material

Supplemental material
supp_80_8_2516__index.html (1.1KB, html)

ACKNOWLEDGMENTS

We thank Baoguang Li and Narjol Gonzalez-Escalona from the U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition (FDA/CFSAN), for providing some of the STEC strains and the LA-320C real-time turbidimeter, respectively.

This work was supported in part by a research grant (SCB11064) from the Center for Produce Safety at the University of California, Davis.

Footnotes

Published ahead of print 7 February 2014

Supplemental material for this article may be found at http://dx.doi.org/10.1128/AEM.04203-13.

REFERENCES

  • 1.Karmali MA, Gannon V, Sargeant JM. 2010. Verocytotoxin-producing Escherichia coli (VTEC). Vet. Microbiol. 140:360–370. 10.1016/j.vetmic.2009.04.011 [DOI] [PubMed] [Google Scholar]
  • 2.CDC. 2013. Surveillance for foodborne disease outbreaks—United States, 2009-2010. MMWR Morb. Mortal. Wkly. Rep. 62:41–47 http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6203a1.htm [PMC free article] [PubMed] [Google Scholar]
  • 3.Gould LH, Walsh KA, Vieira AR, Herman K, Williams IT, Hall AJ, Cole D. 2013. Surveillance for foodborne disease outbreaks—United States, 1998–2008. MMWR Surveill. Summ. 62(SS-2):1–34 http://www.cdc.gov/mmwr/preview/mmwrhtml/ss6202a1.htm [PubMed] [Google Scholar]
  • 4.Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson MA, Roy SL, Jones JL, Griffin PM. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. 10.3201/eid1701.P11101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Johnson KE, Thorpe CM, Sears CL. 2006. The emerging clinical importance of non-O157 Shiga toxin-producing Escherichia coli. Clin. Infect. Dis. 43:1587–1595. 10.1086/509573 [DOI] [PubMed] [Google Scholar]
  • 6.CDC. 2011. Vital signs: incidence and trends of infection with pathogens transmitted commonly through food—foodborne diseases active surveillance network, 10 U.S. sites, 1996–2010. MMWR Morb. Mortal. Wkly. Rep. 60:749–755 http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6022a5.htm [PubMed] [Google Scholar]
  • 7.CDC. 2012. FoodNet 2011 surveillance report. CDC, Atlanta, GA: http://www.cdc.gov/foodnet/PDFs/2011_annual_report_508c.pdf Accessed 1 December 2013 [Google Scholar]
  • 8.CDC. 2013. Incidence and trends of infection with pathogens transmitted commonly through food—foodborne diseases active surveillance network, 10 U.S. sites, 1996–2012. MMWR Morb. Mortal. Wkly. Rep. 62:283–287 http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6215a2.htm [PMC free article] [PubMed] [Google Scholar]
  • 9.USDA. 2012. Shiga toxin-producing Escherichia coli in certain raw beef products. Fed. Regist. 77:31975–31981 http://www.fsis.usda.gov/OPPDE/rdad/FRPubs/2010-0023FRN.pdf [Google Scholar]
  • 10.FDA. 2007. FDA finalizes report on 2006 spinach outbreak. FDA, Rockville, MD: http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/2007/ucm108873.htm Accessed 1 December 2013 [Google Scholar]
  • 11.CDC. 2010. Investigation update: multistate outbreak of human E. coli O145 infections linked to shredded romaine lettuce from a single processing facility. CDC, Atlanta, GA: http://www.cdc.gov/ecoli/2010/ecoli_o145/index.html Accessed 1 December 2013 [Google Scholar]
  • 12.Buchholz U, Bernard H, Werber D, Bohmer MM, Remschmidt C, Wilking H, Delere Y, an der Heiden M, Adlhoch C, Dreesman J, Ehlers J, Ethelberg S, Faber M, Frank C, Fricke G, Greiner M, Hohle M, Ivarsson S, Jark U, Kirchner M, Koch J, Krause G, Luber P, Rosner B, Stark K, Kuhne M. 2011. German outbreak of Escherichia coli O104:H4 associated with sprouts. N. Engl. J. Med. 365:1763–1770. 10.1056/NEJMoa1106482 [DOI] [PubMed] [Google Scholar]
  • 13.CDC. 2013. Reports of selected E. coli outbreak investigations. CDC, Atlanta, GA: http://www.cdc.gov/ecoli/outbreaks.html Accessed 1 December 2013 [Google Scholar]
  • 14.Daniels W. 2012. Nationwide produce outbreak: a moment you never forget. Food Saf. Mag. 17:40–45 http://www.foodsafetymagazine.com/magazine-archive1/december-2011january-2012/nationwide-produce-outbreak-a-moment-you-never-forget/ [Google Scholar]
  • 15.Ge B, Meng J. 2009. Advanced technologies for pathogen and toxin detection in foods: current applications and future directions. J. Lab. Autom. 14:235–241. 10.1016/j.jala.2008.12.012 [DOI] [Google Scholar]
  • 16.Whitaker RJ. 2011. Product testing: considering the issues. Produce Marketing Association, Newark, DE [Google Scholar]
  • 17.Wang F, Yang Q, Kase JA, Meng J, Clotilde LM, Lin A, Ge B. 2013. Current trends in detecting non-O157 Shiga toxin-producing Escherichia coli in food. Foodborne Pathog. Dis. 10:665–677. 10.1089/fpd.2012.1448 [DOI] [PubMed] [Google Scholar]
  • 18.Mori Y, Notomi T. 2009. Loop-mediated isothermal amplification (LAMP): a rapid, accurate, and cost-effective diagnostic method for infectious diseases. J. Infect. Chemother. 15:62–69. 10.1007/s10156-009-0669-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Notomi T, Okayama H, Masubuchi H, Yonekawa T, Watanabe K, Amino N, Hase T. 2000. Loop-mediated isothermal amplification of DNA. Nucleic Acids Res. 28:E63. 10.1093/nar/28.12.e63 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hara-Kudo Y, Nemoto J, Ohtsuka K, Segawa Y, Takatori K, Kojima T, Ikedo M. 2007. Sensitive and rapid detection of Vero toxin-producing Escherichia coli using loop-mediated isothermal amplification. J. Med. Microbiol. 56:398–406. 10.1099/jmm.0.46819-0 [DOI] [PubMed] [Google Scholar]
  • 21.Kouguchi Y, Fujiwara T, Teramoto M, Kuramoto M. 2010. Homogenous, real-time duplex loop-mediated isothermal amplification using a single fluorophore-labeled primer and an intercalator dye: its application to the simultaneous detection of Shiga toxin genes 1 and 2 in Shiga toxigenic Escherichia coli isolates. Mol. Cell. Probes 24:190–195. 10.1016/j.mcp.2010.03.001 [DOI] [PubMed] [Google Scholar]
  • 22.Maruyama F, Kenzaka T, Yamaguchi N, Tani K, Nasu M. 2003. Detection of bacteria carrying the stx2 gene by in situ loop-mediated isothermal amplification. Appl. Environ. Microbiol. 69:5023–5028. 10.1128/AEM.69.8.5023-5028.2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Zhu SR, Chen Y, Wang ZG, Jin DZ, Lu QY, Yao PP. 2009. Detection of enterohemorrhagic Escherichia coli O157:H7 by loop-mediated isothermal amplification. Zhonghua Yu Fang Yi Xue Za Zhi 43:803–808 (In Chinese.) [PubMed] [Google Scholar]
  • 24.Zhao X, Li Y, Wang L, You L, Xu Z, Li L, He X, Liu Y, Wang J, Yang L. 2010. Development and application of a loop-mediated isothermal amplification method on rapid detection Escherichia coli O157 strains from food samples. Mol. Biol. Rep. 37:2183–2188. 10.1007/s11033-009-9700-6 [DOI] [PubMed] [Google Scholar]
  • 25.Hara-Kudo Y, Konishi N, Ohtsuka K, Hiramatsu R, Tanaka H, Konuma H, Takatori K. 2008. Detection of Verotoxigenic Escherichia coli O157 and O26 in food by plating methods and LAMP method: a collaborative study. Int. J. Food Microbiol. 122:156–161. 10.1016/j.ijfoodmicro.2007.11.078 [DOI] [PubMed] [Google Scholar]
  • 26.Hara-Kudo Y, Niizuma J, Goto I, Iizuka S, Kaji Y, Kamakura K, Suzuki S, Takatori K. 2008. Surveillance of Shiga toxin-producing Escherichia coli in beef with effective procedures, independent of serotype. Foodborne Pathog. Dis. 5:97–103. 10.1089/fpd.2007.0031 [DOI] [PubMed] [Google Scholar]
  • 27.Ohtsuka K, Tanaka M, Ohtsuka T, Takatori K, Hara-Kudo Y. 2010. Comparison of detection methods for Escherichia coli O157 in beef livers and carcasses. Foodborne Pathog. Dis. 7:1563–1567. 10.1089/fpd.2010.0585 [DOI] [PubMed] [Google Scholar]
  • 28.Kanki M, Seto K, Harada T, Yonogi S, Kumeda Y. 2011. Comparison of four enrichment broths for the detection of non-O157 Shiga-toxin-producing Escherichia coli O91, O103, O111, O119, O121, O145 and O165 from pure culture and food samples. Lett. Appl. Microbiol. 53:167–173. 10.1111/j.1472-765X.2011.03085.x [DOI] [PubMed] [Google Scholar]
  • 29.Wang D, Liu F, Huo G, Ren D, Li Y. 2009. Development and evaluation of a loop-mediated isothermal amplification method for detecting Escherichia coli O157 in raw milk. J. Rapid Methods Autom. Microbiol. 17:55–66. 10.1111/j.1745-4581.2008.00151.x [DOI] [Google Scholar]
  • 30.Jiang R, Long B, Zeng G, Wang D, Fan H, Wu X. 2012. Loop-mediated isothermal amplification for detecting enterohemorrhagic Escherichia coli O157:H7: a comparison with PCR. Nan Fang Yi Ke Da Xue Xue Bao 32:1026–1030 (In Chinese.) [PubMed] [Google Scholar]
  • 31.Zhao X, Wang J, Forghani F, Park JH, Park MS, Seo KH, Oh DH. 2013. Rapid detection of viable Escherichia coli O157 by coupling propidium monoazide with loop-mediated isothermal amplification. J. Microbiol. Biotechnol. 23:1708–1716. 10.4014/jmb.1306.06003 [DOI] [PubMed] [Google Scholar]
  • 32.Wang F, Jiang L, Ge B. 2012. Loop-mediated isothermal amplification assays for detecting Shiga toxin-producing Escherichia coli in ground beef and human stools. J. Clin. Microbiol. 50:91–97. 10.1128/JCM.05612-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Wang F, Jiang L, Yang Q, Prinyawiwatkul W, Ge B. 2012. Rapid and specific detection of Escherichia coli serogroups O26, O45, O103, O111, O121, O145, and O157 in ground beef, beef trim, and produce by loop-mediated isothermal amplification. Appl. Environ. Microbiol. 78:2727–2736. 10.1128/AEM.07975-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Fratamico PM, Bagi LK, Cray WC, Jr, Narang N, Yan X, Medina M, Liu Y. 2011. Detection by multiplex real-time polymerase chain reaction assays and isolation of Shiga toxin-producing Escherichia coli serogroups O26, O45, O103, O111, O121, and O145 in ground beef. Foodborne Pathog. Dis. 8:601–607. 10.1089/fpd.2010.0773 [DOI] [PubMed] [Google Scholar]
  • 35.Fratamico PM, Debroy C. 2010. Detection of Escherichia coli O157:H7 in food using real-time multiplex PCR assays targeting the stx1, stx2, wzyO157, and the fliCh7 or eae genes. Food Anal. Methods 3:330–337. 10.1007/s12161-010-9140-x [DOI] [Google Scholar]
  • 36.Ramachandran V, Brett K, Hornitzky MA, Dowton M, Bettelheim KA, Walker MJ, Djordjevic SP. 2003. Distribution of intimin subtypes among Escherichia coli isolates from ruminant and human sources. J. Clin. Microbiol. 41:5022–5032. 10.1128/JCM.41.11.5022-5032.2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Scheutz F, Teel LD, Beutin L, Pierard D, Buvens G, Karch H, Mellmann A, Caprioli A, Tozzoli R, Morabito S, Strockbine NA, Melton-Celsa AR, Sanchez M, Persson S, O'Brien AD. 2012. Multicenter evaluation of a sequence-based protocol for subtyping Shiga toxins and standardizing Stx nomenclature. J. Clin. Microbiol. 50:2951–2963. 10.1128/JCM.00860-12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Lang MM, Harris LJ, Beuchat LR. 2004. Survival and recovery of Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes on lettuce and parsley as affected by method of inoculation, time between inoculation and analysis, and treatment with chlorinated water. J. Food Prot. 67:1092–1103 [DOI] [PubMed] [Google Scholar]
  • 39.Wasilenko JL, Fratamico PM, Narang N, Tillman GE, Ladely S, Simmons M, Cray WC., Jr 2012. Influence of primer sequences and DNA extraction method on detection of non-O157 Shiga toxin-producing Escherichia coli in ground beef by real-time PCR targeting the eae, stx, and serogroup-specific genes. J. Food Prot. 75:1939–1950. 10.4315/0362-028X.JFP-12-087 [DOI] [PubMed] [Google Scholar]
  • 40.Feng PC, Jinneman K, Scheutz F, Monday SR. 2011. Specificity of PCR and serological assays in the detection of Escherichia coli Shiga toxin subtypes. Appl. Environ. Microbiol. 77:6699–6702. 10.1128/AEM.00370-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Margot H, Cernela N, Iversen C, Zweifel C, Stephan R. 2013. Evaluation of seven different commercially available real-time PCR assays for detection of Shiga toxin 1 and 2 gene subtypes. J. Food Prot. 76:871–873. 10.4315/0362-028X.JFP-12-365 [DOI] [PubMed] [Google Scholar]
  • 42.Zhang X, Ye Q, Liu Y, Li B, Ivanek R, He K. 2013. Development of a LAMP assay for rapid detection of different intimin variants of attaching and effacing microbial pathogens. J. Med. Microbiol. 62:1665–1672. 10.1099/jmm.0.054551-0 [DOI] [PubMed] [Google Scholar]
  • 43.Feng PC, Reddy S. 2013. Prevalences of Shiga toxin subtypes and selected other virulence factors among Shiga-toxigenic Escherichia coli strains isolated from fresh produce. Appl. Environ. Microbiol. 79:6917–6923. 10.1128/AEM.02455-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.USDA. 2013. Microbiology laboratory guidebook 5B.04: detection and isolation of non-O157 Shiga toxin-producing Escherichia coli (STEC) from meat products and carcass and environmental sponges. USDA, Washington, DC: http://www.fsis.usda.gov/wps/wcm/connect/7ffc02b5-3d33-4a79-b50c-81f208893204/MLG-5B.pdf?MOD=AJPERES Accessed 1 December 2013 [Google Scholar]
  • 45.Beutin L, Jahn S, Fach P. 2009. Evaluation of the ‘GeneDisc' real-time PCR system for detection of enterohaemorrhagic Escherichia coli (EHEC) O26, O103, O111, O145 and O157 strains according to their virulence markers and their O- and H-antigen-associated genes. J. Appl. Microbiol. 106:1122–1132. 10.1111/j.1365-2672.2008.04076.x [DOI] [PubMed] [Google Scholar]
  • 46.Yang Q, Wang F, Prinyawiwatkul W, Ge B. 2014. Robustness of Salmonella loop-mediated isothermal amplification assays for food applications. J. Appl. Microbiol. 116:81–88. 10.1111/jam.12340 [DOI] [PubMed] [Google Scholar]
  • 47.Kase JA, Maounounen-Laasri A, Son I, Deer DM, Borenstein S, Prezioso S, Hammack TS. 2012. Comparison of different sample preparation procedures for the detection and isolation of Escherichia coli O157:H7 and non-O157 STECs from leafy greens and cilantro. Food Microbiol. 32:423–426. 10.1016/j.fm.2012.08.006 [DOI] [PubMed] [Google Scholar]
  • 48.Gill A, Martinez-Perez A, McIlwham S, Blais B. 2012. Development of a method for the detection of verotoxin-producing Escherichia coli in food. J. Food Prot. 75:827–837. 10.4315/0362-028X.JFP-11-395 [DOI] [PubMed] [Google Scholar]
  • 49.Tzschoppe M, Martin A, Beutin L. 2012. A rapid procedure for the detection and isolation of enterohaemorrhagic Escherichia coli (EHEC) serogroup O26, O103, O111, O118, O121, O145 and O157 strains and the aggregative EHEC O104:H4 strain from ready-to-eat vegetables. Int. J. Food Microbiol. 152:19–30. 10.1016/j.ijfoodmicro.2011.10.009 [DOI] [PubMed] [Google Scholar]
  • 50.Verstraete K, Robyn J, Del-Favero J, De Rijk P, Joris M-A, Herman L, Heyndrickx M, De Zutter L, De Reu K. 2012. Evaluation of a multiplex-PCR detection in combination with an isolation method for STEC O26, O103, O111, O145 and sorbitol fermenting O157 in food. Food Microbiol. 29:49–55. 10.1016/j.fm.2011.08.017 [DOI] [PubMed] [Google Scholar]
  • 51.Fedio WM, Jinneman KC, Yoshitomi KJ, Zapata R, Weagant SD. 2012. Efficacy of a post enrichment acid treatment for isolation of Escherichia coli O157:H7 from alfalfa sprouts. Food Microbiol. 30:83–90. 10.1016/j.fm.2011.12.003 [DOI] [PubMed] [Google Scholar]
  • 52.Elizaquivel P, Aznar R. 2008. Comparison of four commercial DNA extraction kits for PCR detection of Listeria monocytogenes, Salmonella, Escherichia coli O157:H7, and Staphylococcus aureus in fresh, minimally processed vegetables. J. Food Prot. 71:2110–2114 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental material
supp_80_8_2516__index.html (1.1KB, html)
AEM.04203-13_zam999105271so1.pdf (16.4KB, pdf)

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES