Incidence and growth of Salmonella enterica on the peel and pulp of avocado (Persea americana) and custard apple (Annona squamosa)

International Journal of Food Microbiology 235 (2016) 10–16

Contents lists available at ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

Incidence and growth of Salmonella enterica on the peel and pulp of avocado (Persea americana) and custard apple (Annona squamosa) Ana Carolina B. Rezende, Juliana Crucello, Rafael C. Moreira, Beatriz S. Silva, Anderson S. Sant'Ana ⁎ Department of Food Science, Faculty of Food Engineering, University of Campinas (UNICAMP), Campinas, SP, Brazil

a r t i c l e

i n f o

Article history: Received 1 December 2015 Received in revised form 8 May 2016 Accepted 25 June 2016 Available online 28 June 2016 Keywords: Occurrence Prevalence Tropical fruits Enterobacteriaceae Modeling Predictive microbiology

a b s t r a c t The aim of this study was to assess the incidence and to estimate the growth kinetic parameters (maximum growth rate, μ; lag time, λ; and maximum population, κ) of Salmonella on the peel and pulp of avocado (Persea americana var. americana) and custard apple (Annona squamosa L.) as affected by temperature (10–30 °C). The incidence of Salmonella was assessed on the peel and pulp of the fruits (n = 200 of each fruit), separately, totalizing 800 analyses. Only three samples of custard apple pulp were positive for Salmonella enterica and the three isolates recovered belonged to serotype S. Typhimurium. Salmonella was not recovered from avocado and custard apple peels and from avocado pulp. Generally, the substrate (pulp or peel) of growth did not affect μ values of S. enterica (p N 0.05). Very similar μ values were found for S. enterica inoculated in custard apple and avocado. S. enterica presented the highest λ in the peel of the fruits. The growth of S. enterica resulted in larger λ in custard apple in comparison to avocado. For example, the λ of S. enterica in the pulp of custard apple and avocado were 47.0 ± 0.78 h and 10.0 ± 3.78 h, respectively. The lowest values of κ were obtained at the lower storage temperature conditions (10 °C). For instance, κ values of 3.7 ± 0.06 log CFU/g and 2.9 ± 0.03 log CFU/g were obtained from the growth of S. enterica in avocado and custard apple pulps at 10 °C (p b 0.05), respectively. On the other hand, at 30 °C, κ values were 6.5 ± 0.25 log CFU/g and 6.5 ± 0.05 log CFU/g, respectively. Significantly higher κ were obtained from the growth of S. enterica in the pulp than in the peel of the fruits (p b 0.05). For instance, the growth of S. enterica in the pulp of avocado led to a κ value of 6.5 ± 0.25 log CFU/g, while in the peel led to a κ value of 4.6 ± 0.23 log CFU/g (p b 0.05). In general, growth kinetic parameters indicated that avocado comprises a better substrate than custard apple for the growth of S. enterica. The square root model fitted to the data obtained in this study and to the growth data available in the literature for other tropical low acid fruits indicated high variability in μ and λ of Salmonella. The results obtained in this study show that whole low acid tropical fruits can harbor Salmonella, and that this foodborne pathogen can not only survive but also grow both on the peel and pulp of low acid tropical fruits, such as avocado and custard apple. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The consumption of fruits and vegetables comprises an important component of a healthy diet and a contributing factor for reducing the risk of non-commutable chronic diseases (Boeing et al., 2012). When it comes to fruits, a special interest of consumers on the so-called exotic fruits has been recently observed pushed by health and wellness trends (Sabbe et al., 2008, 2009a, 2009b). Several of the exotic fruits are indigenous from tropical regions and an important portion presents pH N 4.5 (i.e., low acid fruits). Despite of their functional appeal (Berto et al., 2015; Clerici and Carvalho-Silva, 2011; Saini et al., 2015), some low acid fruits have been associated with large foodborne outbreaks (Berger et al., 2010; Gautam et al., ⁎ Corresponding author at: Rua Monteiro Lobato, 80, PO Box 6121, CEP 13083-862, Campinas, SP, Brazil. E-mail address: [email protected] (A.S. Sant'Ana).

http://dx.doi.org/10.1016/j.ijfoodmicro.2016.06.034 0168-1605/© 2016 Elsevier B.V. All rights reserved.

2014; Laksanalamai et al., 2012) and most fruit-borne outbreaks have been caused by Salmonella (Beatty et al., 2004; CDC, 2008, 2011a, 2012a, 2012b; PHAC, 1998; Sivapalasingam et al., 2003; Teoh et al., 1997). For example, in 2011, foodborne outbreaks involving several Salmonella serotypes contained in fruits such, as melon (CDC, 2011a) and papaya (CDC, 2011b), occurred in the United States. In 2012, an outbreak of S. Braenderup associated due to consumption of contaminated mangos from Mexico involved 127 people in 15 US States, with 13 people being hospitalized (CDC, 2012b). Moreover, several food safety incidents involving low acid fruits have also been reported, such as a voluntary warning due to the association of avocado with salmonellosis in Mexico and the recall of a large quantity of avocados shipped to the United States (FDA, 2008). The association of low acid fruits and salmonellosis can be due to their contamination by Salmonella in any step from farm to fork. For example, in the field, contamination takes place when fruits have direct or indirect contact with soil, dust, water, animal feces, wild animals, insects

A.C.B. Rezende et al. / International Journal of Food Microbiology 235 (2016) 10–16

and food handlers (Moretti, 2007; Sant'Ana et al., 2014). At the postharvest steps, fruits may be contaminated by different sources such as the water and surfaces of equipment. During the hydrothermal treatment for disposal of fly larvae, the contamination of water can lead to internalization of pathogens into the fruits (Bordini et al., 2007). In addition, cross-contamination can take place in any step that involves the application of water tanks (e.g. cooling). The contamination of fruits by the surface of equipment is highly relevant as it was pointed out as the cause of the 2011 salmonellosis outbreak linked to cantaloupe (CDC, 2011a). Likewise, the contamination of fruit peels by foodborne pathogens concerns as bacterial pathogens present on the peel could be transferred through cutting tools (i.e., knives) to the edible parts of fruits. This pathogen can further grow in the fruits if adequate conditions (pH and time/temperature) are provided (CDC, 2011a). Although most fruits present low pH, a considerable number of tropical fruits present pH N 4.5 (low acid fruits), thus representing susceptible matrices to the growth of bacterial pathogens such as Salmonella, and a concern from the public health perspective. The growth of Salmonella in low acid fruits, such as melon, persimmon and papaya has been reported (Penteado and Leitão, 2004a; Rezende et al., 2009). Nevertheless, it should be highlighted that little is known regarding the incidence and growth kinetic parameters of Salmonella in tropical fruits. At the exotic tropical fruits segment, açaí, avocado, cupuaçu, jabuticaba, custard apple, among others stand out as commercially relevant fruits, particularly in Brazil. Among these, avocado (Persea americana var. americana) and custard apple (Annona squamosa L.) deserve special attention because of their great market potential, given the diversity of their functional compounds and vitamins (Silva, 2008). These fruits are consumed mainly in natura and both are low acid, with average pH of 6.2 and 5.4, respectively. To date, there is no report in the literature relating the incidence and growth of Salmonella in custard apple and avocado. The availability of data on the prevalence and growth kinetic parameters of Salmonella in avocado and custard apple comprises important information from both food safety and public health perspectives. Bearing this in mind, the current study was performed aiming to assess the incidence and growth kinetic parameters of Salmonella on the peel and pulp of avocado and custard apple stored in a range of temperature these fruits are exposed from farm to fork. 2. Material and methods 2.1. Incidence of Salmonella on the peel and pulp of avocado and custard apple 2.1.1. Sampling A total of 400 fruits (200 of avocado; P. americana var. americana; and 200 of custard apple, also known as sugar apple; Annona squamosa L.) were acquired in different retail and street markets from the city of Campinas, SP, Brazil. The samples were collected between March and August 2014. 2.1.2. Incidence of Salmonella on the peel and pulp of avocado and custard apple In order to determine the incidence of Salmonella on the peel, each unit of fruit was placed into sterile plastic bags containing sterile 0.1% peptone water (Bacto peptone, Oxoid, Basingstoke, UK). The volume of diluent used was equal to the weight of the fruit sample unit tested (1:1). Fruits immersed in the diluent within sterile plastic bags were gently massaged by rubbing the peel with hands for 5 min under aseptic conditions. During this period, the content inside the sterile bags was also submitted to intermittent agitation. After this procedure, the washing solution was then filtered through a 0.45 μm membrane (Millipore; Merck, Darmstadt, Germany) using a vacuum filtration system (Manifold; Kasvi, Curitiba, Brazil). Thereafter, the fruits were sanitized with chlorine solution (150 mg per L, during 15 min) (Rezende et al.,

11

2009). Then, the fruits were aseptically peeled for removal of pulp and further homogenization. An amount of 25 g of homogenized pulp of each fruit was collected for detection of Salmonella. Considering both fruit parts (peel and pulp), 800 analyses of Salmonella were conducted. The detection of Salmonella spp. was performed according to ISO method 6579 (Anonymous, 2002). Membranes and homogenized of the pulp samples were placed separately, in sterile plastic bags containing 225 mL of buffered peptone water (BPW; Merck, Darmstadt, Germany), homogenized for 2 min (Stomacher 400, Seward, London, UK), followed by incubation at 37 °C for 18 ± 2 h. Further, aliquots of BPW were transferred to Rappaport Vassiliadis Broth Soy Peptone (RVS; Merck, Darmstadt, GER) (41.5 ± 1 °C/24 h ± 3) and Tetrathionate Broth Muller-Kauffman with Novobiocin (MKTTn; Oxoid, Basingstoke, UK). After incubation, aliquots of both broths were streaked onto Xylose Lysine Deoxycholate Agar (XLD; Oxoid, Basingstoke, UK) and Mannitol Lysine Crystal Violet Brilliant Green Agar (MLCB; Oxoid, Basingstoke, UK), followed by incubation at 37 °C for 24 ± 3 h. Presumptive colonies grown either on XLD or MLCB agar were transferred to Nutrient Agar plates (NA; Merck, Darmstadt, GER). NA plates were further incubated at 37 °C for 24 ± 3 h and then subjected to glucose fermentation tests, urea hydrolysis, lysine decarboxylation, β-galactosidases production, acetoin production (Voges-Proskauer test) and indole test (Anonymous, 2002). The colonies were also subjected to agglutination reaction using Salmonella polyvalent antiserum (Probac, Sao Paulo, Brazil). The results were expressed as presence or absence of Salmonella in the pulp (per 25 g) and peel (per fruit). 2.1.3. Phenotypic and genotypic characterization of Salmonella Salmonella isolates were phenotypically (serotyping) and genotypically (genetic profiles by pulsed field gel electrophoresis; PFGE) characterized at the Oswaldo Cruz Foundation (Fiocruz, Rio de Janeiro, RJ, Brazil). PFGE was performed according to the protocol of PulseNet, United States (CDC, 2013). 2.2. Determination of growth kinetic parameters of S. enterica on the peel and pulp of avocado (Persea americana) and custard apple (Annona squamosa) 2.2.1. Samples Avocado (Persea americana) and custard apple (Annona squamosa) were obtained at supermarkets of Campinas, SP, Brazil. The fruits were sanitized with chlorine solution (150 mg per L), and peeled using sterile knife and utensils under aseptic conditions (laminar hood). Portions of 2 g (2 cm2) of the peel from each fruit were aseptically removed and placed in sterile Petri dishes. The peels and pulp of the fruits was stored at −20 °C and defrosted when the experiments were carried out. 2.2.2. Inoculum preparation Three different strains of S. enterica were used: S. Typhimurium (isolated from custard apple in this study), and two strains of clinical origin: S. Enteritidis (S64) and S. Montevideo (S129). These last two strains were chosen because they have similar acid and chlorine resistance as compared to S. Typhimurium according to preliminary experiments performed (data not shown). Each Salmonella strain was separately inoculated into Tryptone Soy Broth (TSB; Oxoid, Basingstoke, UK), followed by incubation at 37 °C for 24 h. Then, a loopful of each strain was transferred to 100 mL of TSB broth, followed by incubation at 37 °C for 24 h. After incubation, equal volumes of TSB broth inoculated with each strain were mixed and centrifuged at 2810 ×g at 8 °C for 10 min (Sorvall Legend XTR, Thermo Scientific, Waltam, USA). The supernatant was discarded and the pellet of the cocktail was resuspended in 100 mL of 0.85% (w/v) NaCl solution (LabSynth, Diadema, Brazil). This entire process was carried out in duplicate (Rezende et al., 2014; Sant'Ana et al., 2012, 2013).

12

A.C.B. Rezende et al. / International Journal of Food Microbiology 235 (2016) 10–16

2.2.3. Inoculation of fruits peel and pulp and enumeration of S. enterica Peels and pulps of the fruits were thawed and placed in sterile plastic bags. The peel and pulp were inoculated with the pool of S. enterica strains to reach a concentration of 102–103 CFU/g. The peels were spot inoculated (10 μL) and allowed to settle for 15 min at room temperature. The pulps were spiked with the suspension cells previously prepared and further gently homogenized. The inoculated peel and pulp were stored at 10, 15, 20 and 30 °C. In addition, non-inoculated samples of fruit peels and pulps were periodically collected for water activity (aw) and pH measurements using an Aqualab® (Decagon, USA) (aw) and a pH meter (model B 374, Micronal, Brazil). The water activity (aw) of the fruits and their parts varied between 0.98 and 0.99 (avocado pulp), 0.96–0.98 (avocado peel), 0.98–0.99 (custard apple pulp) and 0.96–0.97 (custard apple peel). The pH values found were 6.4 ± 0.1 (avocado pulp), 6.3 ± 0.1 (avocado peel), 5.4 ± 0.1 (custard apple pulp), 5.2 ± 0.1 (custard apple peel), respectively. At different sampling times, fruit peel samples (2 g) and pulp (25 g) were placed into sterile plastic bags containing 18 and 225 mL of steriled 0.85% (w/v) NaCl solution (LabSynth, Diadema, Brazil), respectively. Then, the content was homogenized for 2 min in a Stomacher (Lab-Blender 400, West Sussex, UK) and further diluted (1:10 volume ratio) in sterile 0.85% NaCl solution (LabSynth, Diadema, Brazil). Aliquots of 1 mL were pour plated on MLCB agar (Oxoid, Basingstoke, UK), followed by incubation at 37 °C for 24 h. Triplicate test portions of peels and pulps were used. Additionally, the experiments were carried out twice with duplicate plating for each time and temperature condition tested.

2009). When not available, growth kinetic parameters were obtained by fitting the Baranyi model to the growth data at different given storage temperature studied (Barany and Roberts, 1994). Then, the variation in growth kinetic parameters as a function of temperature was modeled using the Ratkowsky model (Ratkowsky et al., 1982). Finally, the growth kinetic parameters of Salmonella obtained in this study (peel and pulp of avocado and custard apple) and from previous studies (Castro et al., 2007; Penteado and Leitão, 2004a, 2004b, Rezende et al., 2009) were compared.

2.4. Statistical analysis The growth kinetic parameters of S. enterica were checked for significant statistical differences (p ≤ 0.05) employing one-factor analysis of variances (ANOVA) followed by Scott-Knott test (Granato et al., 2014; Nunes et al., 2015). Statistical analyses were carried out in Assistat version 7.5 free software (Campina Grande, Brazil) (Silva and Azevedo, 2002).

2.2.4. Modeling the growth kinetic parameters of S. enterica on the peel and pulp of avocado and custard apple The Baranyi model was used to fit the data of S. enterica growth on the peel and pulp of each fruit stored at 10, 15, 20 and 30 °C (Barany and Roberts, 1994) (Eqs. (1)–(3)), using DMFit version 2.1 Excel® add-in (www.ifr.ac.uk/safety/DMfit).
 eμ max Aðt Þ −1 ln ðN ðt ÞÞ ¼ ln ðN0 Þ þ μ max Aðt Þ− ln 1 þ ðN −N Þ e max 0

ð1Þ

 ð−μ t Þ  e max þ q0 1 þ q0

ð2Þ

Aðt Þ ¼ t þ

1 μ max

ln

  1 ln 1 þ q0 λ¼ μ max

ð3Þ

where: ln(N(t)) = log of cell concentration at time t [h] (CFU/g); ln(N0) = log of initial cell concentration (CFU/g); μmax = exponential growth rate (1/h); ln(Nmax) = log of maximum cell concentration; q0 [−] = parameter expressing the physiological state of cells when t = t0; λ = lag time (h). Once the growth kinetic parameters of S. enterica were determined, the Ratkowsky model (Ratkowsky et al., 1982) was used to describe μ and λ as a function of the storage temperature (Eq. (4)): √ r or ln ðλÞ ¼ b  ðT−T min Þ

ð4Þ

In this model, √r is the square root of maximum growth rate and ln of (λ) is the natural logarithm of the lag time. b is the slope of the regression line, T is temperature and T0 is a conceptual minimum temperature for microbial growth, in which T was given in °C. 2.3. Model comparisons Data on growth of Salmonella spp. in low acid fruits such as melon, papaya, watermelon, and persimmon were obtained from the literature (Castro et al., 2007; Penteado and Leitão, 2004a, 2004b, Rezende et al.,

Fig. 1. Growth of Salmonella enterica in the avocado pulp (A) and peel (B) stored at 10, 15, 20 and 30 °C, respectively. Points represent a mean value of four results.

A.C.B. Rezende et al. / International Journal of Food Microbiology 235 (2016) 10–16

3. Results and discussion Salmonella spp. is known to tolerate/grow in pH ranging from 4.0 to 9.0 (Holt et al., 1994), and aw varying from 0.94 to 0.99 (Jay et al., 2005). In the current study, the mean pH values were 6.5 (avocado) and 5.5 (custard apple), while aw was 0.98 for both fruits. These data indicate that these fruits are potential substrates for the growth of Salmonella spp. There are few data on the incidence of this pathogen in low acid fruits, such as cantaloupe (8 positive samples among 151 analyzed) (FDA/CFSAN, 2001), melons, watermelons and papaya (all 120 samples analyzed were negative) (Penteado and Leitão, 2004b) and persimmon (5 positive samples among 585 analyzed) (Rezende et al., 2009). A total of 800 analysis for detection of Salmonella on the peel and pulp (separately) of custard apple and avocado was conducted. Despite the large sampling used in this study, Salmonella was not detected on the peel of any 400 fruit samples. However, this pathogen was detected in three out of 200 custard apple pulp samples analyzed, resulting in an incidence of 1.5%. Salmonella was not detected in the pulp of avocado samples. The presence of Salmonella in fruits suggests that harvesting, handling, distribution or storage were carried out under inadequate hygienic conditions (NACMF, 1999). In fruits, the two most important sites that can harbor microbial contaminants are the peel and the pulp. If present on the surface of fruits (peels), Salmonella can be transferred to the edible portions during cutting and may grow until a high number of viable cells that can increase probability of infections. Complementarily, the presence of Salmonella on fruit peels can result in contamination of surfaces of equipment, cross-contamination during handling and during fruit washing or hydrothermal treatment (Bordini et al., 2007; Penteado et al., 2014; Ravishankar et al., 2010). Nevertheless, in the current study, Salmonella was not found on the peel of the 400 samples of custard apple and avocado. Despite this, the presence of this pathogen in the pulp is highly concerning as the pulp represents a suitable site for microbial growth given the pH and physicochemical composition of low acid tropical fruits. The presence of Salmonella in the pulp custard apple can be an indication of internalization of this bacterium into the fruits. This process is often caused by contact of the fruit with contaminated water, as occurred in 1999 and 2001 with imported mangos from Brazil into the USA. In those outbreaks, the water used for the hydrothermal treatment was considered the source of contamination of fruits (Beatty et al., 2004; Sivapalasingam et al., 2003). As mangoes present pH that can vary from 3.4 to 4.8 (Benevides et al., 2008; Bordini et al., 2007), Salmonella can reach the pulp and the microbial growth can occur (Bordini et al., 2007). In the current study, the three Salmonella isolated from pulp samples of custard apple were characterized (serotyping) as S. Typhimurium. These three isolates presented the same clonal origin according to PFGE profile (data not shown) and were isolated from the pulp of the custard apples belonging to the same batch and acquired from the same producer. Therefore, this may be an indicative of contamination through hydrothermal treatment of fruits or environmental contamination (surfaces contacting the fruits during handling and storage). The fact that S. Typhimurium was detected in the pulp but not on the peel of the fruit samples is of concern, as this pathogen may find adequate conditions to grow during transportation, commercialization and consumer storage phases. This finding reinforces the need to test samples for the incidence of pathogens into fruits (pulps). Sampling strategies must be used aiming to generate data on the incidence of pathogens in whole fruits, including in their pulps, allowing the development of further measures to control the incidence of pathogens in whole fruits and their impacts on public health. Although S. Typhimurium has been more associated with foodborne outbreaks involving cattle and poultry-related sources (CDC, 2015), it is known that the 2012 cantaloupe salmonellosis outbreak was caused by this serotype (CDC, 2012a).

13

The growth kinetic parameters of one S. Typhimurium strain isolated from custard apple with S. Enteritidis (S64) and S. Montevideo on the peel and pulp of avocado (Persea americana) and custard apple (Annona squamosa) were estimated considering their low acidity (pH 5.5–6.5) and high aw (0.98). Non-inoculated fruit peel and pulp samples incubated at same temperature conditions throughout the storage period indicated no presence of Salmonella (limit of quantification — 101 CFU/mL). Figs. 1 and 2 show the growth curves of Salmonella spp. on the peel and pulp of avocado and custard apple, respectively. The average R2 values for fitting the Baranyi model to the growth data of S. enterica inoculated on the peel and pulp of avocado (Fig. 1) and custard apple (Fig. 2) were N0.97, indicating a very good fitting of the data. Figs. 1 and 2 show that both the peel and the pulp of the two fruits studied supported the growth of S. enterica in a wide temperature range (Figs. 1–2). In contrast to the results obtained by Ma et al. (2016), the growth of S. enterica on the peel and pulp of all fruits and storage temperature studied was characterized by the presence of λ (Figs. 1–2). The λ was affected by storage temperature, type and portion of fruit (peel or pulp) (Table 1). For instance, the growth of S. enterica in the pulp and peel of custard apple at 10 °C resulted in average λ values of 69.4 ± 5.5 and 75.7 ± 6.1 h, respectively (Table 1) (p N 0.05). On the other hand, when the growth of this pathogen occurred in the pulp and peel of this same fruit stored at 15 °C, average λ values of 47.0 ± 0.8 and 26.2 ± 4.0 h were obtained respectively (p b 0.05) (Table 1). S. enterica

Fig. 2. Growth of Salmonella enterica in the custard apple pulp (A) and peel (B) stored at 10, 15, 20 and 30 °C, respectively. Points represent a mean value of four results.

14

A.C.B. Rezende et al. / International Journal of Food Microbiology 235 (2016) 10–16

Table 1 Growth kinetic parameters (maximum growth rate, μ — 1/h; lag time, λ — h; and maximum population, κ — log CFU/g) of S. enterica on the peel and pulp of avocado and custard apple stored at 10, 15, 20 and 30 °Ca,b. Growth kinetic

Fruits T °C

parameters μ

λ

κ

a b

10 15 20 30 10 15 20 30 10 15 20 30

Avocado

Custard Apple

Pulp

Peel

Pulp

Peel

0.05 ± 0.04 Ca 0.06 ± 0.02 Cb 0.14 ± 0.05 Ba 0.27 ± 0.01 Aa 22.6 ± 1.11 Aa 10.0 ± 3.78 Ba 11.5 ± 0.51 Ba 3.2 ± 0.11 Ba 3.7 ± 0.06 Ca 3.7 ± 0.01 Ca 6.0 ± 0.11 Ba 6.5 ± 0.25 Aa

0.01 ± 0.01 Ca 0.16 ± 0.04 Ba 0.18 ± 0.08 Ba 0.44 ± 0.21 Aa 39.2 ± 7.87 Aa 11.8 ± 0.31 Ba 10.5 ± 0.51 Ba 1.8 ± 0.11 Cb 3.3 ± 0.17 Ca 4.1 ± 0.03 Ba 4.1 ± 0.10 Bb 4.6 ± 0.23 Ab

0.06 ± 0,06 Ba 0.17 ± 0.07 Ba 0.18 ± 0.04 Ba 0.27 ± 0.00 Aa 69.4 ± 5.48 Aa 47.0 ± 0.78 Ba 8.5 ± 1.10 Ca 3.2 ± 1.10 Cb 2.9 ± 0.03 Da 3.6 ± 0.09 Ca 5.5 ± 0.50 Ba 6.5 ± 0.05 Aa

0.04 ± 0,01 Ba 0.09 ± 0.03 Ba 0.22± 0.17 Aa 0.29 ± 0.06 Aa 75.7 ± 6.08 Aa 26.2 ± 4.00 Bb 12.5 ± 5.88 Ca 8.8 ± 1.42 Ca 2.8 ± 0.20 Ca 4.0 ± 0.80 Ba 4.8 ± 0.20 Aa 5.4 ± 0.67 Ab

Different capital letters in the same column for the same parameter indicate significant differences (p b 0.05) according to Scott-Knott test as a function of storage temperature. Different lower case letters in the same line indicate significant differences (p b 0.05) according to Scott-Knott test as a function fruit part (pulp and peel).

presented, in average, two–three times shorter λ values at 10 °C in the pulp and peel of avocado in comparison to λ values obtained for custard apple (Table 1). This highlights that the fruits can vary in their ability to support the growth of pathogens. In addition, these data indicate that S. enterica can not only survive, but also grow on the peel of fruits in a wide temperature range (Table 1, Figs. 1–2). The increase of storage temperature from 10 to 30 °C resulted in a decrease of λ values of S. enterica inoculated in the pulp and peel of avocado and custard apple of about 7, 21, 21 and 8 times, respectively (Table 1) (p b 0.05). S. enterica presented the shortest λ values when growing in the pulp and peel of avocado. The shortest λ (1.8 h) was recorded when this pathogen was inoculated on the peel of avocado, followed by storage at 30 °C (Table 1). On the other hand, S. enterica presented the largest λ values when growing on the peel of custard apple at 10 °C (Table 1). These results can be explained, in part, by the high pH values of avocado (pH = 6.5) in comparison to custard apple (pH = 5.5). The capacity of Salmonella to grow in the pulp and on the peel of tropical fruits has been demonstrated in few studies. Rezende et al. (2009) evaluated the behavior of S. Enteritidis in the pulp and on the peel of two different varieties of persimmons (“Fuyu” and “Rama Forte”) and observed higher λ times (7.85–17.7 h) for S. Enteritidis, both in the pulp and peel when the fruits were stored at 10 °C. Shorter λ values (3.82 and 4.46 h) were obtained at 30 °C in the pulp and peel of the fruits. These results are consistent with the data observed in the current study: we obtained higher lag values at 10 °C — 22.6, 39.2, 69.4 and 75.7 h in the pulp and peel of avocado and custard apple, respectively, and shorter λ values at 30 °C — 3.2 and 1.8 h for avocado, in the pulp and on the peel and of 3.2 and 8.8 h for the pulp and peel of custard apple, respectively. Penteado et al. (2014) obtained lag phase duration of 19 h for S. Enteritidis in melon pulp stored at 25 °C. This observation is similar to the data for Salmonella incubated at 20 ° C, both in avocado and custard apple pulp (Table 1). No significant differences in μ were found when S. enterica grew on the peel and pulp of the fruits. The only exception observed was when the growth took place at 15 °C. At this condition, a significantly higher μ was obtained on the peel in comparison with μ obtained when the growth took place in the pulp (Table 1). S. enterica presented faster μ (5–7 times) when the temperature was increased from 10 to 30 °C (p b 0.05) (Table 1). With respect to the maximum growth rate, Castro et al. (2007) obtained values of 0.04, 0.20 and 0.41 log 1/h for S. Enteritidis in melon pulp at 10, 20 and 30 °C, respectively. The results of the current study are consistent with those observed by ArvizuMedrano et al. (2001), who reported the ability of Salmonella to grow in the pulp of avocado stored at 22 °C after a short lag phase. Initial levels of Salmonella spp. on the peel and the pulp of avocado and custard apple ranged from 2.5 to 3.0 log CFU/g, reaching maximum

Fig. 3. Relationship between average growth of Salmonella spp. (measured as μ and λ) on the peel and in pulp of avocado and custard apple as affected by temperature compared with literature data. Where A: square root of growth rate (μ) and B: ln of lag time (λ). Data generated in this study were identified as follows: avocado peel (current study) (■), avocado pulp (current study) (♦), custard apple peel (current study) (▲), custard apple pulp (current study) (□). Data obtained from literature were: melon pulp (Castro et al., 2007) (−), melon pulp (Penteado and Leitão, 2004a, 2004b) (◊), papaya pulp (Penteado and Leitão, 2004a, 2004b) (○), watermelon pulp (Penteado and Leitão, 2004a, 2004b) (▬), persimmon “Rama Forte” peel (˟) and pulp (⌂) (Rezende et al., 2009), persimmon “Fuyu” peel (●) and pulp (*) (Rezende et al., 2009).

A.C.B. Rezende et al. / International Journal of Food Microbiology 235 (2016) 10–16

15

Table 2 Secondary models for growth rate (μ) and lag time (λ) of S. enterica inoculated on the peel and pulp of avocado and custard apple stored at 10, 15, 20 and 30 °C compared with literature data for different low acid fruits (pH N 4.6), such as melon, papaya, watermelon and persimmon. Fruit

Fruit portion

Square root

Lag time

Avocado

Pulp Peel Pulp Peel Pulp Pulp Pulp Pulp Pulp Peel Pulp Peel

√μ = 0.0165(T − 2.92) pffiffiffi μ = 0.0287(T − 4.94) pffiffiffi μ = 0.0123(T − 13.8) pffiffiffi μ = 0.0172(T − 3.4) pffiffiffi μ = 0.025(T − 1.9) pffiffiffi μ = 0.0235(T − 1.66) pffiffiffiffi μ = 0.03(T − 6.14) pffiffiffi μ = 0.031(T − 3.86) pffiffiffi μ = 0.0293 (T − 6.09) pffiffiffi μ = 0.0224(T − 0.52) pffiffiffiffi μ = 0.0285(T − 6.19) pffiffiffi μ = 0.0259(T − 3.12)

ln(λ) = −0.0902(T − 43.7) ln(λ) = −0.1456(T − 34.31) ln(λ) = −0.1637(T − 36.15) ln(λ) = −0.1027(T − 48.7) ln(λ) = −0.0671(T − 49.22) ln(λ) = −0.0907(T − 43.84) ln(λ) = −0.1586(T − 35.35) ln(λ) = −0.110 (T − 40.63) ln(λ) = −0.0591(T − 57.08) ln(λ) = −0.0781(T − 47.06) ln(λ) = −0.0589(T − 55.37) ln(λ) = −0.036 (T − 67.38)

Custard Apple Melon Papaya Watermelon Persimmon “Rama Forte” Persimmon “Fuyu”

population (κ) ranging from 3.3 to 4.6 log CFU/g and from 3.7 and 6.5 CFU/g for the peel and pulp of avocado, respectively, and ranging from 2.8 to 5.4 log CFU/g and from 2.9 to 6.5 log CFU/g, respectively for the peel and pulp of custard apple (Table 1). κ values of 4.6 ± 0.2 and 5.4 ± 0.7 were obtained when S. enterica was grown on the peel of avocado and custard apple, respectively. In addition, κ values were significantly higher at 30 °C (p b 0.05) for both fruits and at 20 °C for avocado pulp. The lowest κ values were obtained at 10–15 °C (p b 0.05) (Table 1). These results were expected as higher storage temperature for the growth of Salmonella (30 °C) is close to the optimum growth temperature (37 °C) (Holt et al., 1994). At 10 °C, the Salmonella growth is slower but is not inhibited. Similarly, previous studies also reported a higher increase in population of Salmonella on fresh-cut dragon fruits stored at 28 °C compared to the storage conducted at 12 °C, where κ values reached about 106 log CFU/g (Leverentz et al., 2001; Sim et al., 2013). The determination of the κ value is of paramount importance as infectious dose for Salmonella can vary depending on several factors, such as age, health status of the host and virulence of the strains. Some studies have reported that the infectious dose can be between 15 and 20 cells (United States, 1998), however, others suggest that levels of 106 CFU or more are needed to the onset of symptoms (D'Aust and Maurer, 2007). Temperature is an environmental factor that has a major influence on microbial growth (McMeekin et al., 2008). Given this, the ability to represent the relationship between changes in growth kinetic parameters of bacterial pathogens and temperature through mathematical models becomes highly relevant from the food safety standpoint. This is explained by the fact that fruits are stored and commercialized under variable temperature conditions. The influence of temperature on the growth kinetics parameters (μ and λ) of S. enterica can be seen in Fig. 3. The secondary models (Ratkowsky et al., 1982) represented in this figure were built based on the mean values of the kinetic parameters of S. enterica growth in the pulp and peel of avocado and custard apple (Fig. 3). In addition, secondary models (Ratkowsky et al., 1982) obtained in this study were compared to the growth kinetic parameters of S. enterica obtained in four different temperatures for different low acid fruits, such as melon (Castro et al., 2007; Penteado and Leitão, 2004a, 2004b), papaya, watermelon (Penteado and Leitão, 2004a, 2004b) and persimmon (Rezende et al., 2009). The generation of the secondary model for the growth rate (μ) has been successfully achieved using a linear relationship between the square root of μ and temperature (Ratkowsky et al., 1982). In the case of λ, the natural logarithm transformation (ln) was used as it resulted in a better fitting of the experimental data. Overall, large variabilities regarding μ and λ of S. enterica were observed in all temperature conditions for which data were compared (Fig. 3A and B). However, it is noticeable the larger variability in μ of S. enterica obtained at 30 °C in comparison to the other temperature studied. For instance, μ varied more than 60% when the growth of S. enterica took place in the six fruits studied at 30 °C (Table 1). On the other hand, a large variability in λ of S. enterica was observed

at 10 °C. At this temperature, values of λ varied from two to approximately 4.3 h (Fig. 3B), likely because these were close conditions to the minimal temperature for Salmonella growth (D'Aust and Maurer, 2007). As Fig. 3A–B show, the R2 values for the √μ ranged from 0.91 to 0.99 and from 0.85 to 0.99 for ln (λ). This range of R2 values encompasses variability in √μ and ln (λ) that are inherent of the data collected from different studies, including differences in fruit composition. The secondary models describing the influence of temperature on the variation of growth kinetic parameters (μ and λ) for S. enterica inoculated on the peel and pulp of avocado and custard apple are shown in Table 2. These comprise secondary models that can be applied to predict the growth of S. enterica in pulp and peel of some low acid tropical fruits as affected by temperature. The main implications are that even contamination of the peel at pre- or post-harvest steps may lead to the spread of Salmonella and, for example, on the transference of this pathogen during washing and cutting operations, which may potentially affect the public health. 4. Conclusions The results obtained in this study showed that whole low acid tropical fruits could harbor foodborne pathogens, such as Salmonella. Although the incidence of this pathogen was low in the fruit samples analyzed, the presence of Salmonella in these fruits is of great concern as both the pulp and the peel of avocado and custard apple comprise potential substrates for its growth in a wide storage temperature range. Data indicated that Salmonella growth was more substantial in the pulp of fruits likely because of higher concentration and availability of nutrients in relation to the peel. As data on prevalence of pathogens in low acid fruits and their association with foodborne disease outbreaks is scarce, research on the occurrence and behavior of pathogens in low acid tropical fruits is highly important for the development of management strategies to ensure their microbial safety. Acknowledgements The authors acknowledge the financial support of “Fundação de Amparo a Pesquisa do Estado de São Paulo” (FAPESP) (Grant #13/ 19520-4), “Conselho Nacional de Desenvolvimento Científico e Tecnológico” (CNPq) (Grant #302763/2014-7), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the financial support of this project. References Anonymous, 2002. ISO 6579:2002(E) – Microbiology of Food and Animal Feeding Stuffs – Horizontal Method for the Detection of Salmonella spp. International Organization for Standardization, Switzerland (2002). Arvizu-Medrano, S.M., Iturriaga, M.H., Escartín, E.F., 2001. Indicator and pathogenic bacteria in guacamole and their behavior in avocado pulp. J. Food Saf. 21, 233–241.

16

A.C.B. Rezende et al. / International Journal of Food Microbiology 235 (2016) 10–16

Barany, J., Roberts, A.T., 1994. Mathematics of predictive food microbiology. Int. J. Food Microbiol. 26, 199–218. Beatty, M.E., Laporte, T.N., Phan, Q., Van Duyne, S.V., Braden, C., 2004. A multistate outbreak of Salmonella enterica serotype Saintpaul infections linked to mango consumption: a recurrent theme. Clin. Infect. Dis. 38, 1337–1338. Benevides, S.D., Ramos, A.M., Stringheta, P.C., Castro, V.C., 2008. Quality of the fruits and pulp of Ubá mango. Cienc. Tecnol. Aliment. 28, 571–578. Berger, C.N., Sodha, S.V., Shaw, R.K., Griffin, P.M., Pink, D., Hand, P., Frankel, G., 2010. Fresh fruit and vegetables as vehicles for the transmission of human pathogens. Environ. Microbiol. 12, 2385–2397. Berto, A., Silva, A.F., Visentainer, J.V., Matsushita, M., Souza, N.E., 2015. Proximate compositions, mineral contents and fatty acid compositions of native Amazonian fruits. Food Res. Int. http://dx.doi.org/10.1016/j.foodres.2015.08.018. Boeing, H., Bechthold, A., Bub, A., Ellinger, S., Haller, D., Kroke, A., Leschik-Bonnet, E., Müller, M.J., Oberritter, H., Schulze, M., Stehle, P., Watzl, B., 2012. Critical review: vegetables and fruit in the prevention of chronic diseases. Eur. J. Nutr. 51, 637–663. Bordini, M.E.B., Ristori, C.A., Jakabi, M., Gelli, D.S., 2007. Incidence, internalization and behavior of Salmonella in mangoes, var. Tommy Atkins. Food Control 18, 1002–1007. Castro, M.F.P.M., Schmidt, F.L., Penteado, A.L., 2007. Growth of Salmonella Enteritidis and Listeria monocytogenes in melon pulp: predictive modeling and evaluation of model performance. Braz. J. Food Technol. 10, 176–182. Centers for Disease Control and Prevention — CDC, 2008. Multistate outbreak of Salmonella Litchfield infections linked to cantaloupes (final update). (Available at:) http:// www.cdc.gov/salmonella/2008/cantaloupes-4-2-2008.html (Accessed 10/10/15). Centers for Disease Control and Prevention — CDC, 2011a. Multistate outbreak of Salmonella Panama infections linked to cantaloupe. Salmonella, 23rd of June. (Available at:) http://www.cdc.gov/salmonella/panama0311/062311/index.html (Accessed 04/05/ 15). Centers for Disease Control and Prevention — CDC, 2011b. Salmonella infection [salmonellosis]. Salmonella outbreaks. Whole, fresh imported papayas — Salmonella Agona. August 29, 2011 (final update). Investigation update: multistate outbreak of human Salmonella Agona infections linked to whole, fresh imported papayas. 29 of August. (Available at:) http://www.cdc.gov/salmonella/agona-papayas/082911/index.html (accessed 04/20/15). Centers for Disease Control and Prevention — CDC, 2012a. Multistate outbreak of Salmonella Typhimurium and Salmonella Newport infections linked to cantaloupe (final update). Salmonella, 05 of October. (Available at:) http://www.cdc.gov/salmonella/ typhimurium-cantaloupe-08-12/index.html (Accessed 04/05/15). Centers for Disease Control and Prevention — CDC, 2012b. Multistate outbreak of Salmonella Braenderup infections associated with mangoes (final update). Salmonella, 11 of October. (Available at:) http://www.cdc.gov/salmonella/braenderup-08-12/index. html (Accessed 07/25/15). Centers for Disease Control and Prevention — CDC, 2013. Standard Operating Procedure for PulseNet PFGE of Escherichia coli O157:H7, Escherichia coli non-O157 (STEC). Salmonellaserotypes, Shigella sonnei and Shigella flexneri 1–13. Available at: https://www.cdc.gov/pulsenet/pdf/ecoli-shigella-salmonella-pfge-protocol-508c.pdf. Centers for Disease Control and Prevention – CDC, 2015. Preliminary incidence and trends of infection with pathogens transmitted commonly through food — foodborne diseases active surveillance network, 10 U.S. sites, 2006–2014. May 15, 2015, 64, 495– 499. (Available at:) http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6418a4. htm (Accessed 10/28/15). Clerici, M.T.P.S., Carvalho-Silva, L.B., 2011. Nutritional bioactive compounds and technological aspects of minor fruits grown in Brazil. Food Res. Int. 44, 1658–1670. D'Aust, J.Y., Maurer, J., 2007. Salmonella species. In: Doyle, M.P., Beuchat, L.R. (Eds.), Food Microbiology Fundamentals and Frontiers, third ed. ASM Press, Washington, pp. 187–236 (Cap.10). FDA/CFSAN, 2001. Survey of imported fresh produce. FY 1999. (Available at:) http:// www.cfsan.fda.gov/dms/prodsur6.html (Accessed 09/10/15). Food And Drug Admistration - FDA, 2008. Grande Produce, LTD. CO Recalls Jalapeno Peppers, Serrano Peppers, and Avocados Because of Possible Health Risk. Available at: http://www.fda.gov/Safety/Recalls/ArchiveRecalls/2008/ucm112470.htm. (Accessed 08/02/2014). Gautam, D., Dobhal, S., Payton, M.E., Fletcher, J., Ma, L.M., 2014. Surface survival and internalization of Salmonella through natural cracks on developing cantaloupe fruits, alone or in the presence of the melon wilt pathogen Erwinia tracheiphila. PLoS One 9 (8), e105248. Granato, D., Calado, V.M.A., Jarvis, B., 2014. Observations on the use of statistical methods in food science and technology. Food Res. Int. 55, 137–149. Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T., 1994. Bergey's Manual of Determinative Bacteriology. nineth ed. Lippincott Williams & Wilkins, Baltimore (787 pp.). Jay, J.M., Loessner, M.J., Golden, D.A., 2005. Modern Food Microbiology. seventh ed. Springer, New York (790 pp.). Laksanalamai, P., Joseph, L.A., Silk, B.J., Burall, L.S., Tarr, C., Gerner-Smidt, P., Datta, A.R., 2012. Genomic characterization of Listeria monocytogenes strains involved in a multistate listeriosis outbreak associated with cantaloupe in US. PLoS One 7 (7), e42448. Leverentz, B., Conway, W.S., Alavidze, Z., Janisiewicz, W.J., Fuchs, Y., Camp, M.J., Chighladze, E., Sulakvelidze, A., 2001. Examination of bacteriophage as a biocontrol method for Salmonella on fresh-cut fruit: a model study. J. Food Prot. 64, 1116–1121. Ma, C., Li, J., Zhang, Q., 2016. Behavior of Salmonella spp. on fresh-cut tropical fruits. Food Microbiol. http://dx.doi.org/10.1016/j.fm.2015.10.006.

McMeekin, T., Bowman, J., McQuestin, O., Mellefont, L., Ross, T., Tamplin, M., 2008. The future of predictive microbiology: strategic research, innovative applications and great expectations. Int. J. Food Microbiol. 128, 2–9. Moretti, C.L., 2007. Manual de processamento mínimo de frutas e hortaliças. Empresa Brasileira de Pesquisa Agropecuária, Embrapa Hortaliças, Ministério da Agricultura, Pecuária e Abastecimento, Serviço Brasileiro de Apoio às Micro e Pequenas Empresas, Brasília, DF, p. 257. National Advisory Committee on Microbiological Criteria for Foods – NACMCF, 1999r. Microbiological safety evaluations and recommendations on fresh produce. Food Control 10, 117–143. Nunes, C.A., Alvarenga, V.O., Sant'Ana, A.S., Santos, J.S., Granato, D., 2015. The use of statistical software in food science and technology: advantages, limitations and misuses. Food Res. Int. 75, 270–280. Penteado, A.L., Leitão, M.M.F., 2004a. Growth of Listeria monocytogenes in melon, watermelon and papaya pulps. Int. J. Food Microbiol. 92, 89–94. Penteado, A.L., Leitão, M.M.F., 2004b. Growth of Salmonella Enteritidis in melon, watermelon and papaya pulp stored at different times and temperatures. Food Control 15, 369–373. Penteado, A.L., Castro, M.F.P.M., Rezende, A.C.B., 2014. Salmonella enterica serovar Enteritidis and Listeria monocytogenes in mango (Mangifera indica L.) pulp: growth, survival and cross-contamination. J. Sci. Food Agric. 94, 2746–2751. Public Health Agency of Canada — PHAC, 1998. Salmonella Oranienburg, Ontario. In: Canadian Communicable Disease Report 24, 177–178. Ratkowsky, D.A., Olley, J., McMeekin, T.A., Ball, A., 1982. Relationship between temperature and growth rate of bacterial cultures. J. Bacteriol. 149, 1–5. Ravishankar, S., Zhu, L., Jaroni, D., 2010. Assessing the cross contamination and transfer rates of Salmonella enterica from chicken to lettuce under different food-handling scenarios. Food Microbiol. 27, 791–794. Rezende, A.C.B., Castro, M.F.P.M., Porto, E., Uchima, C.A., Benato, E., Penteado, A.L., 2009. Occurrence of Salmonella spp. in persimmon fruit (Diospyrus kaki) and growth of Salmonella Enteritidis on the peel and in the pulp of this fruit. Food Control 20, 1025–1029. Rezende, A.C.B., Igarashi, M.C., Destro, M.T., Franco, B.D.G.M., Landgraf, M., 2014. Effect of gamma radiation on the reduction of Salmonella strains, Listeria monocytogenes, and Shiga toxin-producing Escherichia coli and sensory evaluation of minimally processed spinach (Tetragonia expansa). J. Food Prot. 77, 1768–1772. Sabbe, S., Verbeke, W., Van Damme, P., 2008. Familiarity and purchasing intention of Belgian consumers for fresh and processed tropical fruit products. Br. Food J. 110, 805–818. Sabbe, S., Verbeke, W., Van Damme, P., 2009a. Confirmation/disconfirmation of consumers' expectations about fresh and processed tropical fruit products. Int. J. Food Sci. Technol. 44, 539–551. Sabbe, S., Verbeke, W., Van Damme, P., 2009b. Perceived motives, barriers and role of labeling information on tropical fruit consumption: exploratory findings. J. Food Prod. Mark. 15, 119–138. Saini, R.K., Nile, S.H., Park, S.W., 2015. Carotenoids from fruits and vegetables: chemistry, analysis, occurrence, bioavailability and biological activities. Food Res. Int. 76, 735–750. Sant'Ana, A.S., Barbosa, M.S., Destro, M.T., Landgraf, M., Franco, B.D.G.M., 2012. Growth potential of Salmonella spp. and Listeria monocytogenes in nine types of ready-to-eat vegetables stored at variable temperature conditions during shelf life. Int. J. Food Microbiol. 157, 52–58. Sant'Ana, A.S., Destro, M.T., Landgraf, M., Franco, B.D.G.M., 2013. Growth potential of Salmonella and Listeria monocytogenes in ready-to-eat lettuce and collard greens packaged under modified atmosphere and in perforated film. J. Food Prot. 5, 888–891. Sant'Ana, A.S., Silva, F.F.P., Maffei, D.F., Franco, B.D.G.M., 2014. FRUITS AND VEGETABLES | Introduction. In: Batt, C.A., Tortorello, M.L. (Eds.), Encyclopedia of Food Microbiology, 2ed. v. 1. Elsevier, Waltham, pp. 972–982 (Org.). Silva, A.C., 2008. Abacate como fonte terapêutica. In: Leonel, S. (Ed.), Abacate: Aspectos técnicos da produção, first ed. Universidade Estadual Paulista-Cultura Acadêmica Editora, São Paulo, pp. 215–239 (Org.). Silva, F.A.S., Azevedo, C.A.V., 2002. Versão do programa computacional Assistat para o sistema operacional Windows. Revista Brasileira de Produtos Agroindustriais Vol. 4, pp. 71–78. Sim, H.L., Hong, Y.K., Yoon, W.B., Yuk, H.G., 2013. Behavior of Salmonella spp. and natural microbiota on fresh-cut dragon fruits at different storage temperatures. Int. J. Food Microbiol. 160, 239–244. Sivapalasingam, S., Barrett, E., Kimura, A., Duyne, S.V., Witt, W.D., Ying, M., Frisch, A., Phan, Q., Gould, E., Shilam, P., Reddy, V., Cooper, T., Hoekstra, M., Higgins, C., Sanders, J.P., Tauxe, R.V., Slutsker, L., 2003. A multistate outbreak of Salmonella enterica serotype Newport infection linked to mango consumption: impact of water-dip disinfestation technology. Clin. Infect. Dis. 37, 1585–1590. Teoh, Y.L., Goh, K.T., Neo, K.S., Yeo, M., 1997. A nationwide outbreak of coconut-associated paratyphoid: a fever in Singapore. Ann. Acad. Med. Singap. 26, 544–548. United States, 1998. Food and Drug Administration. Center for Food Safety and Applied Nutrition. The bad bug book: foodborne pathogenic microorganisms and natural toxins handbook. (Available at:) http://www.cfsan.fda.gov/~mow/chap1.html (Accessed 08/02/2014).