Attachment and Survival of Salmonella enterica and Listeria monocytogenes on Tomatoes (Solanum lycopersicum) as Affected by Relative Humidity, Temperature, and Storage Time.

ABSTRACT: Tomatoes (Solanum lycopersicum) are one of the most commonly consumed fruits worldwide. The fruit can become contaminated with Salmonella and Listeria monocytogenes at various stages of the production and supply chain, and these pathogens may survive under various storage conditions. The effects of relative humidity, temperature, and duration of storage on the attachment and survival of both pathogens on the surface of tomatoes were investigated. Fresh whole Roma tomatoes were inoculated with a cocktail of Salmonella or L. monocytogenes strains and stored at 5, 12, 25, 30, or 35°C for up to 10 days. Every day during storage, relative humidity and temperature were measured and tomatoes were removed to enumerate pathogens cells that were loosely attached (LA; cells were detached from the tomato surface by rinsing) and strongly attached (SA; sonication was required to detach cells from the tomato surface). The attachment strength (SR) was calculated to express the proportion of surviving SA cells on the tomato surface. The initial levels of Salmonella and L. monocytogenes on the tomato surface after inoculation were 6.6 and 6.5 log CFU per tomato for LA cells and 5.1 and 5.6 log CFU per tomato for SA cells, respectively. For both pathogens, the LA levels were higher (P < 0.05) than the SA levels. The LA and SA levels differed significantly as a function of temperature, relative humidity, and duration of storage. The SR for Salmonella was affected by storage time but not temperature, whereas the SR for L. monocytogenes was affected by storage time and temperature and relative humidity (P < 0.05). An understanding of the attachment and survival of Salmonella and L. monocytogenes on tomatoes stored under various temperature conditions may be useful for preventing or reducing the establishment of pathogens and for designing improved decontamination methods.

Simultaneous and individual quantitative estimation of Salmonella, Shigella and Listeria monocytogenes on inoculated Roma tomatoes (Lycopersicon esculentum var. Pyriforme) and Serrano peppers (Capsicum annuum) using an MPN technique.

Simultaneous and individual enumeration of Salmonella, Shigella and Listeria monocytogenes was compared on inoculated Roma tomatoes and Serrano peppers using an Most Probable Number (MPN) technique. Samples consisting of tomatoes (4 units) or peppers (8 units) were individually inoculated with a cocktail of three strains of Salmonella, Shigella or L. monocytogenes, or by simultaneous inoculation of three strains of each pathogen, at low (1.2-1.7 log CFU/sample) and high (2.2-2.7 log CFU/sample) inocula. Samples were analyzed by an MPN technique using universal pre-enrichment (UP) broth at 35 °C for 24 ±â€¯2 h. The UP tubes from each MPN series were transferred to enrichment and plating media following adequate conventional methods for isolating each pathogen. Data were analyzed using multifactorial analysis of variance (p < 0.05) and LSD multiple rang test. There were differences (p < 0.05) in recovery of simultaneous and individual bacteria inoculated (individual > simultaneous), type of bacteria (Salmonella > Shigella and L. monocytogenes), type of sample (UP broth > pepper and tomato), and inoculum level (high > low). The MPN technique was effective for Salmonella on both commodities. Shigella counts were higher on tomatoes compared to peppers, (p < 0.05), and for L. monocytogenes on peppers (p < 0.05).

Use of a novel medium, the Polymyxin Ceftazidime Oxford Medium, for isolation of Listeria monocytogenes from raw or non-pasteurized foods.

Polymyxin Ceftazidime Oxford Medium (PCOM), a novel selective and differential plating medium for Listeria monocytogenes was compared with Modified Oxford Agar (MOX) for efficacy to isolate L. monocytogenes and other Listeria spp. naturally present in non-pasteurized Mexican-style cheese (n = 50), non-pasteurized fresh squeezed orange juice (n = 50), raw beef chunks (n = 36), and fresh cabbage (n = 125). Samples were collected from retail markets and farms in Mexico and tested following the US Department of Agriculture enrichment technique. Listeria spp. were isolated from 23.4% of analyzed samples, and from those, 75.0% corresponded to raw beef chunks, 38.0% to non-pasteurized Mexican-style cheese, and 30.0% to fresh squeezed orange juice. No Listeria spp. were isolated from fresh cabbage samples. L. monocytogenes was recovered from 15.3% of food samples analyzed. Non-pasteurized Mexican-style cheese showed the highest proportion of L. monocytogenes positive samples (36.0%), followed by orange juice (26.0%) and raw beef (25.0%). The frequency of isolation of Listeria spp. and L. monocytogenes was not different (P > 0.05) between PCOM and MOX. The advantages of using PCOM when comparing to MOX, include the easier way to identify Listeria species, the lower cost per plate and the availability of its ingredients for Latin-American countries.

Quantitative distribution of Salmonella spp. and Escherichia coli on beef carcasses and raw beef at retail establishments.

Salmonella is a foodborne pathogen that commonly inhabits the gastrointestinal tract of a healthy feedlot cattle and can be transferred to the carcass surface during hide removal and evisceration procedures. Numerous investigations on Salmonella prevalence throughout different stages of the beef chain have been conducted. In contrast, limited studies are available on quantitative determinations of Salmonella at different steps in raw meat production. Quantitative data, particularly for pathogenic bacteria such as Salmonella are important for quantitative risk assessment. Salmonella spp. and Escherichia coli populations were enumerated on beef carcass samples collected at abattoirs and also in beef chunks and ground beef samples collected from butcher's shops at retail in Jalisco State, Mexico. Sponge samples from beef carcass sides (n=142) were collected immediately after final water wash and before chilling at three non-federally inspected abattoirs following USDA-FSIS sampling protocols. Beef chunks (n=84) and ground beef (n=65) samples were obtained from 86 butcher's shops. Salmonella enumeration was conducted by the Most Probable Number method and E. coli counts were determined using Petrifilm plates. Salmonella was isolated from 18% of beef carcasses, 39% of beef chunks and 71% of ground beef samples. Salmonella mean counts were 1.3±0.9 Log MPN/300 cm(2) on beef carcasses, 1.9±0.9 and 2.3±1.1 Log MPN/25 g in beef chunks and ground beef samples, respectively. Twenty-six Salmonella serotypes and 11 serogroups were identified among 432 isolates recovered. Salmonella typhimurium (14%), Salmonella sinstorf (12%) and S. Group E1 monophasic (10%) were the most frequent. Escherichia coli was present on 97, 84 and 100% of beef carcasses, beef chunks and ground beef samples, respectively. Escherichia coli mean counts were 3.2±0.7 Log CFU/300 cm(2), 3.9±1.1 and 4.5±1.2 Log CFU/25 g on beef carcasses, beef chunks and ground beef, respectively. Salmonella prevalence and mean counts found in raw beef were higher than previously reported in studies from other countries. The data collected in this study show a trend in the prevalence of Salmonella to be higher as meat processing is extended at retail. This, together with the diversity of serotypes found, indicates that raw meat is exposed to multiple contamination sources during slaughter and retail processing and highlights the necessity to implement Sanitation Standard Operating Procedures for those establishments. Finally, this study provides quantitative information for future risk assessments associated with the risk of human salmonellosis.

The Polymyxin Ceftazidime Oxford Medium as an alternative selective and differential medium for isolation of Listeria monocytogenes from raw or unpasteurized food.

The Polymyxin Ceftazidime Oxford Medium (PCOM) was developed to recover Listeria monocytogenes from raw or unpasteurized foods. It contains esculin-ferric ammonium citrate as indicator system for Listeria growth, and ceftazidime and polymyxin B as selective agents, which are available in several Latin American countries. Comparison of PCOM, Modified Oxford Medium (MOX) and Tryptic Soy agar with 0.6% yeast extract (TSAYE) indicated that both selective media were equally effective at recovering four individual strains of L. monocytogenes (Scott A, V7, California and broccoli), and a mixture of these strains (LMM) (P > 0.05). The ability of PCOM, MOX, TSAYE and TSAYE supplemented with 4% NaCl to recover heat, acid and freeze-damaged LMM was similar for all media (P > 0.05). The PCOM proved to be effective at isolating colonies of LMM from inoculated raw beef chunks, unpasteurized orange juice, cabbage, and Mexican-style cheese by direct plating and by the US Department of Agriculture's Food Safety and Inspection Service enrichment method. Differentiation of L. monocytogenes colonies was easier on PCOM than on MOX for foods with high levels of background microbiota. Based on the evaluations performed on foods naturally contaminated with L. monocytogenes, PCOM was a more economical alternative than MOX for selective and differential isolation of Listeria from raw or unpasteurized foods.

Effect of the use of a neutralizing step after antimicrobial application on microbial counts during challenge studies for orange disinfection.

The effects of using a neutralizer after applying antimicrobial treatments and the effect of time lapse between treatment application and subsequent recovery and enumeration of Escherichia coli O157:H7 and Salmonella were investigated in Valencia oranges. Inoculated oranges surfaces were washed with distilled water for 15 s and then sprayed with a solution containing 200 mg/liter sodium hypochlorite (pH 6.5) for 15 s; they were then dipped in L-lactic acid (2.0% at 55°C) for 1 min or in distilled water at 80°C for 1 min. Posttreatment, oranges were divided into two groups. In the first group, oranges were dipped in neutralization treatment: 270 ml of buffered peptone water for 2 min for lactic acid-treated oranges, 270 ml of Dey-Engley broth for 2 min for chlorine-treated oranges, or 3.7 liters of tap water (25°C) for 10 s for hot water-treated oranges. The second group of treated oranges was not subjected to any neutralizer. All oranges then were kept at room temperature (average 26.2°C) and sampled at 0, 7.5, and 15 min for enumeration of surviving Salmonella and E. coli O157:H7. The orange surface (30 cm(2)) was excised for pathogen enumeration. The presence of free chlorine and changes in pH and temperature on the orange surface were determined in uninoculated, treated oranges. Free chlorine was detected on oranges after treatment; the change in temperature of orange surfaces was greater during treatment with hot water than with lactic acid. Nevertheless, pathogen enumeration did not show any impact of neutralizer use on the residual activity of antimicrobials or any impact of the time elapsed between antimicrobial treatment and recovery of bacterial pathogens from inoculated oranges (P ≥ 0.05). The results of this study indicate that the lack of a neutralizing step before enumeration of pathogens is not likely to affect the accuracy of results during challenge studies to test pathogen reduction strategies on oranges.

Comparison of different washing treatments for reducing pathogens on orange surfaces and for preventing the transfer of bacterial pathogens to fresh-squeezed orange juice.

The objectives of this study were to compare the effectiveness of various washing treatments for reducing Escherichia coli O157:H7, Salmonella sp., and Listeria monocytogenes populations on orange surfaces and to measure the effect of some of these treatments in preventing the transfer of pathogens during juice extraction. Orange surfaces inoculated with L. monocytogenes or a mixture of E. coli O157:H7 and Salmonella Typhimurium were washed by water spray and then sprayed with or dipped in water at 80°C for 1 min, 70% ethanol for 15, 30, or 45 s or 1, 2, or 4 min, 2 or 4% lactic acid solution at 55°C for 15, 30, or 45 s or 1, 2, or 4 min, or 200 mg/liter hypochlorite at pH 6.5 or 10 for 15 s. The surviving populations of these pathogens on the oranges were enumerated after each treatment. In a further stage, the ability of these pathogens to be transferred to the juice during extraction was tested. Juice was obtained from inoculated oranges that were subjected to selected treatments using chlorine, lactic acid, ethanol, and hot water as described above, and then bacterial counts in orange juice were determined. The application of these treatments reduced the populations of pathogens on orange surfaces by 1.9 to >4.9 log, 1.9 to >4.6 log, and 1.4 to 3.1 log cycles for E. coli O157:H7, Salmonella Typhimurium, and L. monocytogenes, respectively. The treatments using hot water or lactic acid showed greater reductions than other treatments. The time, antimicrobial concentration, and form of application affected the bacterial reduction. All treatments resulted in undetectable counts in the juice. Nevertheless, pathogens were recovered by the enrichment-plating method. Treatment of oranges before juice extraction may reduce the risk associated with consuming orange juice.

Spread of bacterial pathogens during preparation of freshly squeezed orange juice.

To study the potential of three bacterial pathogens to cross-contaminate orange juice during extraction, normal operation conditions during juice preparation at food service establishments were simulated. The spread of Salmonella enterica serovar Typhimurium, Escherichia coli O157:H7, and Listeria monocytogenes from inoculated oranges to work surfaces and to the final product was determined. The transference of these three bacterial pathogens to orange juice made from uninoculated oranges with the use of contaminated utensils was also studied. Fresh oranges were inoculated with a marker strain of rifampicin-resistant Salmonella Typhimurium, E. coli O157:H7, or L. monocytogenes. Final pathogen levels in juice were compared as a function of the use of electric or mechanical juice extractors to squeeze orange juice from inoculated oranges. Pathogen populations on different contact surfaces during orange juice extraction were determined on sulfite-phenol red-rifampicin plates for Salmonella Typhimurium and E. coli O157:H7 and on tryptic soy agar supplemented with 0.1 g of rifampicin per liter for L. monocytogenes. After inoculation, the average pathogen counts for the orange rind surface were 2.3 log10 CFU/cm2 for Salmonella Typhimurium, 3.6 log10 CFU/cm2 for E. coli O157:H7, and 4.4 log10 CFU/cm2 for L. monocytogenes. This contamination was spread over all utensils used in orange juice squeezing. Mean pathogen counts for the cutting board, the knife, and the extractor ranged from -0.3 to 2.1 log10 CFU/cm2, and the juice contained 1.0 log10 CFU of Salmonella Typhimurium per ml, 2.3 log10 CFU of E. coli O157:H7 per ml, and 2.7 log10 CFU of L. monocytogenes per ml. Contact with contaminated surfaces resulted in the presence of all pathogens in orange juice made from uninoculated oranges. These results give emphasis to the importance of fresh oranges as a source of pathogens in orange juice.