E. Coli in Raw Produce: A Public Health Risk for Consumers

Raw produce and sprouted seeds may harbor pathogenic strains of Escherichia coli, including the enteric Shiga toxin-producing E. coli (STEC), or enterohemorrhagic E. coli (EHEC). In addition to the rates of mortality and mobidity already associated with this food-borne pathogen, hydrolyzing enzymes endemic to the strain, extended spectrum β-lactamases (ESBLs), may actually increase these rates and trigger an acute threat to public health and consumer safety.

For this reason, Skočková et al.¹ analyzed raw vegetables and sprouted seeds for the presence of E. coli and related antimicrobial resistance, including the presence of genes that encode resistance to tetracycline and β-lactam antibiotics and genes that encode specific virulence. To do this, they collected 91 samples from retailers in the Czech Republic. Of these, there were 39 domestic samples, 52 imported samples, 17 organic samples, and 74 conventional samples. For detection, the researchers performed enrichment using buffered peptone water (Thermo Scientific) followed by aerobic culture on Tryptone Bile X-Glucuronide agar (Thermo Scientific) and MacConkey agar. From each sample that was positive for E. coli., they selected 1 to 3 isolates for confirmation and characterization.

Overall, the researchers identified E. coli in 24 (26.4%) of the samples, including multiple strains per sample, for a total of 27 isolates. They noted a high detection rate among sprouted seed samples (53.3%) and lettuce samples (38.9%). Other samples with positive results for E. coli were cucumber (1), spring onion (2), cauliflower (2), and asparagus (1). Interestingly, only 3 (17.6%)  organic samples tested positive while 21 (28.4%) conventional samples were positive. Of the contaminated samples, 17.9% were domestic, and 32.7.% were imported.

The team used the disk diffusion method to determine antibiotic susceptibility. For this, they used Mueller-Hinton agar and antibiotic disks (Thermo Scientific) to make classifications of susceptible, intermediate resistant, or resistant. They used the E-test (Thermo Scientific) to establish the minimum inhibitory concentration (MIC) of antimicrobials in isolates that were resistant to ampicillin, tetracycline, and trimethoprim.

The researchers found resistance to ampicillin (13 isolates, 48.1%), tetracycline (5 isolates, 18.5%), and, to a lesser degree, trimethoprim, amoxicillin/clavulanic acid, sulfamethoxazole/trimethoprim, streptomycin, and nalidixic acid. For ampicillin, the E-test with 2006 CLSI criteria confirmed resistance in 3 strains (MIC 128–>256 μg.ml⁻¹), intermediate resistance in 7 strains (MIC 9–31 μg.ml⁻¹), and sensitivity in 3 strains (MIC ≤8 μg.ml⁻¹). However, the application of 2013 EUCAST criteria to the E-test confirmed ampicillni resistance in 10 isolates (MIC >9 to >256 μg.ml⁻¹). They confirmed tetracyline resistance in all 4 isolates (MICs of 70–128 μg.ml⁻¹). E-test also verified resistance to trimethoprim and sulfamethoxazole/trimethoprim (MIC of >32 μg.ml⁻¹), amoxicillin/clavulanic acid (MIC 128 μg.ml⁻¹) and streptomycin (MIC 32 μg.ml⁻¹).

To determine ESBL production, Skočková et al. used the double disk synergy test and chromogenic selective medium Brilliance ESBL (Thermo Scientific). They performed polymerase chain reaction (PCR) to identify the genes bla_TEM, bla_SHV, and bla_CTX, which confer the ESBL phenotype, as well as the genes eaeA, hly, stx₁, and stx₂, which encode virulence factors, and the genes tet(A), tet(B), tet(C), and tet(G), which encode tetracycline resistance. The team also performed pulsed field gel electrophoresis (PFGE) on all E. coli isolates, in concordance with the macrorestriction analysis recommendations of the US-based Centers for Disease Control Food-borne Bacterial Disease Surveillance entity.

In this study, the researchers detected bla_TEM in 2 isolates with MICs >256 μg.ml−1 but did not detect bla_SHV or bla_CTX. For tetracycline resistance genes, they found tet(B) in 3 isolates and tet(A) in one isolate. The team did not detect the Shiga-encoding genes stx₁ and stx₂ nor the hly gene. They did detect eaeA, which produces the virulence factor intimin, in 3 isolates (11.1%). The presence of intimin-encoding genes suggest potential enteropathogenicity but requires additional confirmation.

Skočková et al. concluded that raw produce represents a potential risk to consumer health. They also note that the macrorestriction data revealed marked heterogeneity (25 PFGE profiles) among the samples with possible clonal relationships between a few samples with a common distributor. This indicates that contamination with food-borne pathogens is a risk at all stages of process.

Reference: 

1 Skočková, A. et al. (2013) ‘Characteristics of Escherichia coli from raw vegetables at a retail market in the Czech Republic.’ International Journal of Food Microbiology 167, 196–201.