Assessing the growth of Staphylococcus aureus and Escherichia coli on fruits and vegetables.

INTRODUCTION
The number of registered foodborne diseases involving fresh produce is a preoccupation in many countries. For this reason, the aim of this study was to better understand the growth of Staphylococcus aureus and Escherichia coli, two indicators of hygienic and sanitary conditions, on fruits and vegetables that were exposed at different temperatures.


METHODOLOGY
The main salads served at the buffets of commercial restaurants were artificially contaminated with separate pools of both pathogens and subsequently exposed at 10, 20 and 30 °C and at different time intervals. Then, the growth potential of S. aureus and E. coli on each fruit and vegetable was determined.


RESULTS
There was no significant S. aureus and E. coli growth on all evaluated foods exposed at 10 °C until 6 hours. When comparing both microorganisms, E. coli demonstrated higher growth potential than S. aureus on all analysed salads. Peculiarly, E. coli had the highest growth rate for the tomato (α = 6.43 at 30 °C), a fruit with low pH.


CONCLUSION
We suggest that fruits and vegetables should be distributed at temperatures equal to or lower than 10 °C and should not be kept for more than 2 hours at room temperature.


Introduction
The consumption of fresh fruits and vegetables is expected to increase worldwide between 8-9 % over the next 5 years [1].As consumption increases, the number of registered foodborne diseases involving fresh produce is also becoming more common in many countries, including Brazil [2,3].
According to official records on foodborne diseases kept between 2000 and 2018, food services (restaurants and bakeries) in Brazil were the second most frequent establishments involved with foodborne outbreaks, only behind private residences, where the Sanitary Surveillance Services has no access.Among the main etiological agents involved with these outbreaks have been Escherichia coli and Staphylococcus aureus [2], both frequently used as indicators of hygienic and sanitary conditions [4,5].Even though the majority of E. coli is not pathogenic, some strains can cause foodborne diseases.For example, pathotype enterotoxigenic E. coli (ETEC) is responsible for infantile diarrhea in developing countries, while the pathotype enterohaemorrhagic E. coli (EHEC) is regularly associated with food poisoning outbreaks in the developed world [6].Fresh fruits and vegetables that are exposed to these strains could represent a risk for public health.S. aureus and both E. coli (ETEC and EHEC) have been isolated on fruits and salads, including strawberries, fruit salads, radish, cabbage, lettuce, cucumber, carrot, spinach and sprouts [7][8][9][10][11][12][13].These foods can be contaminated and microbial multiplication can occur at various stages throughout the processing chain, including during harvest, storage, transport and distribution [14].Inadequate temperatures during these stages can allow pathogenic bacteria growth, which is one of the main reasons for the occurrence of foodborne outbreaks involving fruits and vegetables [15,16].
According to Brazilian food-service legislation [17], cold food should be maintained at < 5 °C, while hot food should be maintained at > 60 °C for less than 6 hours.This legislation does not mention the cold chain distribution time and is an important topic to be explored.Therefore, the aim of this study was to assess the S. aureus and E. coli growth on fruits and vegetables in order to achieve a better understanding of what may happen during their distribution in food services.

Commercial restaurants data
An informal interview approach was used to identify the fruits and vegetables most frequently served in 50 commercial restaurants in Porto Alegre, Southern Brazil.We elaborated questions regarding the number of daily meals, the fruits and vegetables most frequently served at buffets, and the served or distributed regularity of these foods per week.The data were grouped and ordered, and a descriptive statistical analysis was carried out to identify the frequency, means and standard deviations of each fruit and vegetable served/distributed at buffets.

Bacterial cultures
Five strains of S. aureus and five strains of E. coli were selected to compose two bacterial pools, which were artificially and separately inoculated on fruits and vegetables.The strains used were S. aureus 4668/03 (isolated from a foodborne outbreak that occurred in Southern Brazil), S. aureus S6 (isolated from a stainless-steel surface in a poultry slaughterhouse), S. aureus S8 (isolated from a cutting board of a poultry slaughterhouse), S. aureus ATCC 2998 and S. aureus ATCC 25923.The strains of E. coli were E. coli CQ (isolated from a hot dog in Southern Brazil), E. coli ECHC (isolated from human cystitis), E. coli DH5α (laboratory strain), E. coli ATCC 8739 and E. coli ATCC 25922.

Inoculum preparation
Each strain of S. aureus and E. coli was grown in Brain Heart Infusion broth (BHI, HiMedia, Mumbai, India) at 37 °C for 18-24 hours.After incubation, 2 mL of each strain of S. aureus and E. coli were mixed separately in a sterile tube.Next, each bacterial pool was centrifuged for 5 min at 14000 rpm; the supernatants were then discharged, and the pellets were washed with 0.1% peptone water.This procedure was repeated twice.In the second repetition, cells were resuspended in 0.1% peptone water and the final cell concentration was adjusted through optical density 0.5 (OD630nm) and plate count at 10 8 CFU/mL.Decimal serial dilutions using 0.1% peptone water were prepared and the pools were separately inoculated, reaching final concentrations on fruits and vegetables of approximately 1000 CFU/g (3 log CFU/g).

Inoculation and preparation of vegetables and fruits
All fruits and vegetables were acquired from a supermarket in Porto Alegre, Brazil.The water activity (aw) was performed using an AquaLab water activity meter (Model 3TE, Decagon Devices, USA).For the pH, the foods were homogenized at the ratio of 1:2 (food:water) and were then measured in a pHmeter (Model Q400A, Quimis, Diadema, Brazil).
Each food was submitted to different treatments and processing (Table 1).Then, 10 g of each sample was transferred to a sterile plastic bag.Each fruit and vegetable surface-piece was separately contaminated with 1 mL of each bacterial pool.Subsequently, samples were incubated at different temperatures, simulating the following scenarios: (1) refrigeration using inadequate temperature (10 °C), (2) distribution at room temperature (20 °C) and (3) distribution at room temperature during a Brazilian summer day (30 °C).Samples were taken at 0, 2, 4, 6 hours (simulating the time exposure for these foods at a buffet) and 24 hours (considering inadequate foods reuse).In order to evaluate the microbial contamination of fruits and vegetables, 10 g of each food was homogenized in 90 mL of 0.1% peptone water in a stomacher plastic bag, which was followed by decimal dilution.S. aureus counts were carried out using Baird-Parker Agar (BP, Oxoid, Hampshire, United Kingdom).E. coli counts were done using Violet Red Bile Agar (VRB, Merck, Darmstadt, Germany) and mesophilic microorganisms were also quantified using Brain Heart Infusion Agar (BHI, HiMedia, Mumbai, India).The microorganisms were quantified by drop method [18] with a detection limit of 1.69 log CFU/g (50 CFU/g).Not-inoculated controls were included to verify the initial microbial quality of fruits and vegetables that were used.Experiments were performed at least three times with three repetitions and the results were expressed as log CFU/g.484

Determination of growth potential (α)
The growth potential (α) of S. aureus and E. coli on each fruit and vegetable was estimated by calculating the log difference between the initial counts and the counts at 24 hours.S. aureus and E. coli were considered unable to grow on the foods when α value was negative or lower than 0.5 log [19].

Statistical analysis
The S. aureus and E. coli growth on all fruits and vegetables was statistically analysed via examination of variance (ANOVA) by applying the Tukey test (p < 0.05).The experimental data were analysed by using Statistic 13.0 software (STATSOFT Inc., Tulsa, OK, USA).

Commercial restaurants evaluation
Among the restaurants visited, 34.0% served up to 150 daily meals; 38.3% served between 150 and 300 daily meals, and 27.7% served more than 300 daily meals.Observing that if the restaurants had maximum attendance five days a week, then fruits and vegetables would have high distribution an average of four days a week.The fruits and vegetables more often exposed at the buffets included cucumber (3.61 ± 1.45 days), green cabbage (3.96 ± 1.33 days), cooked broccoli (4.06 ± 1.20 days), watermelon (4.21 ± 1.32 days), grated carrot (4.31 ± 1.16 days), papaya (4.67 ± 0.88 days) and tomato (4.91 ± 0.41 days).Due to these high rates of exposure, these foods were chosen for this study.

S. aureus, E. coli and mesophilic microorganism growth on fruits and vegetables
Samples from each fruit and vegetable used in this study were analysed for the presence of E. coli and S. aureus before inoculation.These pathogens were not detected on any of the analysed samples.Results shown in Table 2 indicate that S. aureus did not grow on any evaluated fruit or vegetable stored at 10 °C until 24 hours.For tomatoes, S. aureus did not grow at any evaluated temperature, indicating that the pool used for this pathogen did not adapt to this food (Table 2).At 20 °C, there was not S. aureus growth on grated carrot and cabbage during 24 hours.At the same temperature, S. aureus growth was detected in cucumber and watermelon after 6 hours; however, at 30 °C, the growth of this pathogen was significant after only 2 hours in both foods.For broccoli, at 20 and 30 °C, S. aureus growth began in less than 2 hours (p < 0.05), with 24 hours demonstrating the highest final populations (8.14 log CFU/g at 20 °C and 9.44 log CFU/g at 30 °C).
Growth for mesophilic bacteria was similar for all foods evaluated at 10 °C in relation to S. aureus, but at 20 and 30 °C the total count was higher.
There was not E. coli growth in all evaluated fruits and vegetables exposed at 10 °C until 6 h (Table 3).In contrast to S. aureus growth, E. coli was able to grow on tomatoes in only 4 hours at 30 °C.Like with the tomatoes, there was E. coli growth on grated carrot and green cabbage after 4 hours at 30 °C and after 6 hours at 20 °C.Papaya, cucumber, watermelon and broccoli all demonstrated faster growth of E. coli.At 20 °C, this microorganism was able to grow on these fruits and vegetables in 4 hours, while E. coli was able to grow at 30 °C in 2 hours (p < 0.05).Mesophilic bacteria growth was similar growth for these foods (Table 3).

S. aureus and E. coli growth potential on fruits and vegetables
At 10 °C, no S. aureus growth potential was observed until 24 hours for all fruits and vegetables evaluated (Figure 1A).At the same temperature, the E. coli growth potential was insignificant for tomato, carrot and cabbage (Figure 1B).However, E. coli had a surprisingly growth rate for tomatoes at 20 and 30 °C, with α = 4.80 and α = 6.43, respectively.The carrot also drew attention because it had negative S. aureus growth potential at 10 and 20 °C.

Discussion
The interaction between intrinsic and extrinsic factors is fundamental to guarantee the quality and safety of food.In this work, foods with different pH and aw were kept for 24 hours at different temperatures to determine safe parameters for their distribution in food 485 services.We chose to start with the temperature 10 °C because it is more easily attained in food services.The results showed that both S. aureus (Table 2) and E. coli (Table 3) did not grow at this temperature for 24 hours.Similarly, Likotrafiti et al. [20] showed that Listeria monocytogenes and Escherichia coli O157:H7 also did not grow on cucumber stored at 10 °C for 24 hours, suggesting that this temperature can be used to maintain fruits and vegetables for a short period of time.
The absence of S. aureus growth at 20 °C on grated carrot and cabbage may be partially explained by these vegetables composition.Babic et al. [21] have showed that extracts of peeled and shredded carrots have an antimicrobial effect against foodborne microorganisms like S. aureus and L. monocytogenes.They have suggested that this antimicrobial activity is due to apolar compounds present in purified and active extracts of carrots.As for green cabbage, Rúa et al. [22] have demonstrated that various phenolic compounds naturally present in plants showed antimicrobial activity against several S. aureus strains.Although there was not S. aureus growth on the grated carrots and green cabbage, there was a significant mesophilic bacteria growth in 24 hours (Table 2).According to McLandsborough [23], S. aureus is a relatively poor competitor with food microbiota.Thus, besides the food matrix, the growth of other bacteria may have inhibited S. aureus growth.In the same way, S. aureus did not grow on papaya at 20 °C for 6 hours, which can be explained by this fruit composition.Studies have suggested that the high quantity of proteolytic enzymes present in papaya, including papain and chymopapain, can inhibit S. aureus growth [24,25].However, S. aureus was able to grow on cucumber, watermelon and broccoli and this growth may be explained by the high presence of vitamins, high water activity and pH of these vegetables and the lack of inhibiting enzymes (Table 1).Broccoli demonstrated higher final populations and was the food that had pH closer to neutrality (6.71), thus probably favouring S. aureus growth.Another factor that may have contributed to this result was the fact that the broccoli was cooked, which may have resulted in competitive flora elimination.
For the tomatoes, S. aureus did not grow at any evaluated temperature.Although the tomato has high water activity, its low pH (Table 1) may constitute an unfavourable environment for microorganism growth.However, the E. coli was able to grow on this food, surprisingly reaching a final population of 9.7 log CFU/g at 30 °C in 24 hours (Table 3).While in our study, E. coli grew after 4 hours at 20 ° C, the Food and Drug Administration (FDA) [26] has reported greater lag times for Salmonella on cut Beefsteak tomatoes (pH = 4.54), 5.29 to 7.49 hours at 22.2 °C.Such findings demonstrate the importance of the time and temperature controls for fruits and vegetables in food services.
We calculated the S. aureus and E. coli growth potential on the evaluated fruits to better understand their microbial behaviour in 24 hours (Figure 1).According to Sant'Ana et al. [19], the pathogens growth is not considered if growth potential (α) is equal to or lower than 0.5 log CFU/g.When comparing both microorganisms, E. coli demonstrated higher growth potential than S. aureus among the fruits and vegetables analysed (Figure 1).The negative growth potential with carrots (Figure 1A) may have occurred because the S. aureus population was reduced in 24 hours, probably because of the competition with mesophilic bacteria, which presented higher values at 10 and 20 °C (Table 2).We suggest that the E. coli strains used in this work have more significant growth in fruits and vegetables than the S. aureus strains.Furthermore, E. coli showed to be more resistant against the competitive microbiota existing in the analysed foods.
To ensure food microbiological safety and quality, predictive microbiology can forecast the response of microorganism growth in relation to factor variations such as temperature, storage conditions, humidity and pH.Preliminary studies highlight the necessity for improving the use of predictive microbiology because this tool can assist food services and regulatory agencies, thus maintaining the quality and microbiological safety of fresh produce [27,28].From this global study, we can choose the fruits and vegetables most susceptible to pathogen growth in order to build predictive models and to better control the time and temperature of distribution in food services.

Conclusion
Based on our results, fruits and vegetables kept at 10 °C did not demonstrate S. aureus and E. coli growth, suggesting that foods maintained at temperatures equal to or lower than 10 °C would be safely conserved, at least during the period of distribution (6 hours).However, our results demonstrated that E. coli was able to grow in less than 2 hours at 30 °C, indicating fruits and vegetables have to be kept at refrigeration temperatures, especially on warm days.This approach may be also a suggestion to food safety managers for better defining the control measures to be adopted in food services to prevent foodborne diseases transmitted by fruits and vegetables.However, it is important to note that other pathogens, especially psychrotolerantes like Listeria monocytogenes, should be explored.

Table 1 .
Treatments, processing applied, pH and water activity (aw) of foods analyzed.