Occurrence and genetic characterization of Listeria spp . in minimally processed vegetables commercialized in Porto Alegre , Brazil *

Minimally processed vegetables go through many steps before they are refrigerated, selection, washing, peeling, cutting, disinfection and finally packaging. However, if no care is taken at the origin of the raw materials and in the process ing stages, there is a chance of finding pathogenic bacteria, such as Listeria monocytogenes , which are able to grow at low temperatures. The aim of this research was to verify the occurrence of Listeria sp. in minimally processed vegetables sold in Porto Alegre, Brazil, and the genetic relationship among the isolates. Minimally processed salads were sampled monthly from local supermarkets and analyzed by inoculation on Listeria Enrichment Broth and subsequent seeding on two selective media, Palcam and Modified Oxford Agar. The typical colonies were identified to species level and their intergenic region 16S-23S rDNA were amplified in order to verify the genetic variability. Species of Listeria were found in 23 of the 52 processed salad samples analyzed and L. monocytogenes was found in seven. The presence of L. monocytogenes in the samples is a health concern, as these salads are eaten without further treatment by the consumer. The amplification of the intergenic region 16S-23S rDNA, showed a great genetic diversity among the isolates, with 43 different patterns, proving the usefulness of this technique in epidemiologic studies.


INTRODUCTION
The increasing demand for fresh fruit and vegetables, has encouraged the consumption of minimally processed foods [22,23].However, if these ready-to-eat food products meet the claims for practicality of modern society, they may also represent hazard to consumers, once they do not systematically carry out any sanitization process before such products are eaten.When present in minimally processed vegetables, Listeria isolates may be able to survive if no inhibition mechanism is prompted during storage [12], since this bacteria grows at ordinary refrigeration temperatures and in modified atmosphere [5,21].Since these readyto-eat products are quite popular, the investigation of the occurrence of L. monocytogenes in leafy vegetables, especially of the kind used in salad preparations, is particularly important.
The contamination of vegetables may be of diverse sources and take place at different stages of a food processing plant.It may take place in the processing environment as a whole [8,17], either before or even after the food has been sanitized [10,20].In this scenario, molecular methods have proved themselves useful in the establishment of contamination sources throughout a production line [13,14,18,19].With this intent, the amplification of the 16S -23S rDNA spacer regions, whether cleaved or not with restriction enzymes, has been used in many studies [1,3,6,26].
This study aimed (i) to establish the occurrence of Listeria sp. in minimally processed leafy vegetables, (ii) to characterize the isolates by amplification of the 16S -23S rDNA spacer regions., and (iii) to correlate the data obtained with the origin of the product.

MATERIALS AND METHODS
Ready-to-eat salad samples were randomly collected in supermarkets of Porto Alegre (RS, Brazil) from August 2004 to August 2005.Four samples were collected per month, totaling 52 samples that varied for salad mix types and commercially available brands (A, B and C).Each sample was composed by three 200g salad packages of the same brand and salad mix, to make up a 600g analysis unit.Twenty-five grams of each sample were homogenized in 225 mL Listeria enrichment broth (LEB) and incubated at 30°C for 48 hours, the agar plates were inspected for typical colonies for the following 7 days.Afterwards, Listeria colonies were isolated from Modified Oxford Agar (MOX) and Palcam agar (PAL) and seeded in Triptone Soy Agar supplemented with yeast extract (TSA -YE 0.6%) for purification.The following biochemical assays were used to establish genus and identify species: Gram test; catalase; specific motility at 28°C; Triple-Sugar-Iron; methyl red; Voges-Proskauer; nitrate reduction; fermentation of xylose, rhamnose, and manitol; hemolysis and CAMP-test.
Chromosomal DNA was extracted [4] and the amplification of the 16S -23S rDNA spacer region was carried out using primers designed based on the sequences for different Listeria sp.found in GenBank: 16S LisF 5' GCTGGATCACCTCCTTTCTA 3', and 23S LisR 5' GCGCCCTTYCTAACYTAACC 3'.The reactions were carried out in 10mM Tris/HCl, pH 8.3; 50mM KCl; 2 mM MgCl 2 ; 0.2 mM dNTPs; 0.38 µM primer 16S LisF; 0.5 µM primer 23S LisR; 1 U Taq DNA polymerase and 100 ng of template DNA, to a final volume of 25 µL.Amplification reactions were conducted in a thermal-cycler (Mastercycler Personal Eppendorf) as follows: an initial denaturation step at 94°C for 2 min followed by 30 cycles at 94°C for 1 min, at 53°C for 1 min, at 72°C for 1 min, and a final extension step at 72°C for 10 min.For each set of PCR analyses, positive controls were prepared with L. monocytogenes ATCC 15315 and L. innocua ATCC 33090, these reference strains behaved unvaryingly throughout the experiments.The PCR products were electrophoresed in 8% polyacrylamide gels (10 V/cm for 2 h 30 min.)stained with ethidium bromide (0.5 µg/mL), analyzed under UV light and photographed with a digital camera Kodak 1D (version 3.5.2).The molecular marker λ EcoRI / HindIII / ClaI was used as a standard for inspecting fragment size.

RESULTS
In the samples analyzed in the present study Listeria species were isolated in 23 of the 52 samples of minimally processed salad products analyzed.As for the commercial brands chosen, Listeria sp. were detected in 16 samples of brand A (n=24), 5 samples of brand B (n=10), and in 2 samples of brand C (n=18).L. seeligeri was detected in 11 samples, L. monocytogenes in 7, L. welshimeri in 7, L. grayi in 6, L. innocua in 4, and L. ivanovii in 3 samples.No correlation between the presence of Listeria species and the composition of the salad mixes was observed.(Table 1).The highest occurrence of Listeria was between January and April 2005, when Listeria were isolated in 3 samples per collection.The most common species throughout the year were L. seeligeri and L. welshimeri, detected in almost all brands analyzed.L. seeligeri prevailed in the samples as a whole, whereas L. monocytogenes prevailed in warm weather samples.

DISCUSSION
In vegetables, the occurrence of Listeria genus varies greatly in product range, especially in readyto-eat foods [7,9,16,26].The genus Listeria can be found in different environments and L. monocytogenes is the species most associated with food borne diseases.However, the presence of other Listeria species indicates that environmental conditions are ideal to microbial growth.The presence of L. monocytogenes in ready-to-eat produce raises health concerns, as these food products are extensively consumed without further treatment.The amplification of the 16S -23S rDNA spacer regions in Listeria isolates has proved its genetic diversity, with the detection of 43 patterns.Such high number of patterns may occur due to the quantity of operons in the 16S -23S rDNA spacer regions present in the Listeria genome [11].The genetic diversity among strains, as detected in the present study, may be explained by several aspects involved in the contamination route, such as the conditions of soils, manure, and of the water destined to irrigation and to washing.These aspects may vary considerably between food processing units, between farms, and even between seasons, which makes contamination difficult to track down for a specific point in the production chain.
Recurring strains were observed in the samples investigated, proving that these strains may be present in these food processing units or, at least, that they recur in the vegetable raw materials therein processed.This last hypothesis is plausible in the light of the fact that these recurring strains are detected in different commercial brands that process raw food material of identical origin.Recurring Listeria strains have also been identified by other authors [2,15,24,27], who suggested that the main contamination source is ultimately the processing unit itself.Interestingly, the occurrence of the same genotypic profile in isolates of distinct origin evinces the role played by one same strain in the contamination of different products.Nevertheless, it remains to be established whether this given strain has recently been dispersed or is a common strain adapted to the food processing environment.

CONCLUSIONS
All six Listeria species were detected and isolated from the salad samples analyzed, roughly throughout the entire sampling period.These findings may reveal the degree of extensive distribution and resistance of this microorganism during the processing and storing procedures adopted for this kind of food product.Though simple, the molecular method used in the present study afforded the discrimination between isolates of the same species.The method described may be equally useful in the investigation of disease outbreaks, in the identification of contamination sources, and in the characterization of inter-and intraspecies genetic diversity.The mapping of isolate dispersal, both in food processing environments and in food items themselves, will assist the outlining of hygiene measures against Listeria sp. in the production of safe foods.

Table 1 )
). L. monocytogenes and L. seeligeri presented the highest number of patterns in common, patterns 18, 35, and 38.Fourteen patterns were found in more than one species, and 7 patterns were detected in one species exclusively.As for the commercial brands chosen for investigation, more specifically, Brands A and B shared different patterns for L. grayi (pattern 13) and L. seeligeri (patterns 18, 24, 35, and 39).Brands A and C shared pattern 35 for L. monocytogenes, whereas Brands B and C did not share any patterns.Recurring strains were observed, .The pattern found for the isolate L. grayi subspecies murrayi was the same as observed for the 15 isolates of L. grayi subspecies grayi (pattern 28).

Table 1 .
Distribution of Listeria species isolated from minimally processed vegetables commercialized in Porto Alegre, Brazil, and their molecular patterns.The sampling period was from August 2004 to August 2005.Capital letters correspond to the brand and numbers, to the salad mix type.* and (n*) correspond to isolates that didn't recovered for DNA extraction.** Salad brand A and B, original from Porto Alegre, Rio Grande do Sul, salad brand C, original from São José dos Pinhais, Paraná #