Several outbreaks of food poisoning have been linked to the consumption of contaminated fresh vegetables and fruits. Enterohemorrhagic E. coli O157 (O157) and Salmonella are the most common causes of these illnesses. Listeria is also supposed to be a cause of food-borne illness. Detection methods for these three pathogenic bacteria from fresh vegetables and fruits were studied by enrichment culture procedure, PCR detection and selective detection. The first step was enrichment by BPW (buffered peptone water) for 20-24 hr at 36°C. After 6 hr of enrichment, 1ml of enrichment solution was transferred to NmEC (EC broth with Novobiocin and bile acid) for separation of O157. After enrichment for 20-24 hr, 1 ml of enrichment solution was treated as a PCR sample. Fewer than 10 cells of O157 or Salmonella per 25 g food samples could be detected by our methods. Listeria (1, 000 cells/25 g) were also detectable by these methods. Criteria for the multiplex PCR (O157, Salmonella and Listeria) were key factors for further testing (Limitations of No. detected were2.3×104, 2.7×105 and2.2×105 cfu/ml, respectively). Selective cultures were used when PCR was positive for pathogenic bacteria. These methods are economical and effective for detection of these pathogenic bacteria from fresh vegetables and fruits.
Modified EC broth (mEC) with novobiocin (20 mg/L) is widely used for the isolation of Escherichia coli O157: H7 from foods. The mEC with novobiocin at 20 mg/L (mEC+20n) is usually prepared by autoclaving mEC without novobiocin at 121°C for 15 min and then adding a filter sterilized aqueous solution of sodium novobiocin after cooling to produce a final concentration of 20 mg/L. In Japan, mEC containing 25 mg/L novobiocin [25n mECA (Eiken) and 25n mEC-B (Kyokuto)] autoclaved at 121°C for 15 min is used for the isolation of E. coli O157: H7 from foods. We compared bacterial growth in 10 concentrations of novobiocin and mEC with inoculums of 10 cfu and 105 cfu using 6 strains of E. coli O157: H7 at 42°C for 18 hr. At the 10 cfu inoculum level growth was inhibited in 25n mEC-A and mEC with novobiocin 25 mg/L (added after autoclaving). We then compared 4 concentrations of novobiocin and mEC at recovering E. coli O157: H7 from food samples with both direct plating and IMS plating (immunomagnetic separation method). The eighteen food samples inoculated with E. coli O157: H7 (0.1-23.3 cfu/g or ml) were incubated at 42°C for 18 hr. E. coli O157: H7 was isolated from 18 (100%) with 20n mEC, 16 (89%) with mEC + 20n, 15 (83%) with 25n mEC-A, and 16 (89%) with 25n mEC-B using the IMS plating. The stability after 3 years storage of 20n mEC at 40°C, 30°C, room temperature, 10°C, and 4°C was examined for growth of E. coli O157: H7, for inhibition of other bacteria, for pH, and for moisture (%). The growth of E. coli O157: H 7 and the inhibition of Grampositive bacteria and Serratia marcescens on mEC with novobiocin plate were the same as the control medium. The pH and moisture after 3 years storage for 20n mEC were almost the same as the original medium under room temperature condition. We have found 20n mEC is less selective than mEC + 20n and more selective than mEC. The 20 n mEC is easy to prepare without adding a filter-sterilized aqueous solution of sodium novobiocin to produce a final concentration of 20 mg/ L after the medium cooled. The 20n mEC is a practical medium for screening large numbers of food samples as well as samples involved in disease outbreaks.
As a part of the basic research for the improvement of the method for detection of Shigalike toxin producing Escherichia coli O157: H7 (hereinafter called “E. coli O157”) from raw meat, we examined the enrichment culture method for E. coli O157 injured by heating, by combination of culture medium (mEC-Nb, TSB), incubation time (5, 10, 20 hr), temperature (37°C, 42°C), and method (static, shaken). The examination also tested the efficacy of novobiocin for some species of bacteria which compete with E. coli O157. The results of the examination are as follows. 1) It was shown that the enrichment from very few (10-3 cfu/ml) E. coli O157 injured by heating to 106 cfu / ml or more in an enrichment culture using mEC-Nb medium prepared by the official method took 20 hours, regardless of the incubation temperature or method. 2) After 10 hours or more of enrichment in TSB, the cfu count of E. coli O157 injured by heating was greater at 42°C than at 37°C, irrespective of the incubation method. 3) It was confirmed that the growth of bacteria which would compete with E. coli O157 in the enrichment was inhibited by the addition of 20 mg/l novobiocin to the culture medium.
We examined 82 imported raw vegetables and 14 imported raw fruits for contamination with Salmonella, Enterohemorrhagic Escherichia coli O 157: H 7, and Listeria. Of 12 bean sprout samples, one was contaminated with L. monocytogenes serotype 1/2a, and two with L. innocua. One of two kiwi fruit samples was contaminated with L. seeligeri. Aeromonas caviae was isolated from one of five young field pea samples. One of three alfalfa samples was contaminated with Enteropathogenic E. coli 08. Three months later, we examined an additional 10 imported bean sprout samples obtained from the store that had sold the L. monocytogenes-contaminated bean sprouts described above. L. monocytogenes serotype 1 /2a was isolated from three samples, and L. innocua from one. The AFLP technique (Amplified Fragments Length Polymorphism) was used for genetic analysis of four isolates of L. mono-cytogenes serotype 1/2a and no homology was found among the isolates. The high L. monocytogenes-contamination rate for bean sprouts indicates poor hygiene control in growth, harvest, washing, distribution, and storage of the product. The washing and handling of imported raw vegetables and fruits should be improved to prevent food-borne disease.