2581

Journal of Food Protection, Vol. 69, No. 11, 2006, Pages 2581–2586 Copyright ᮊ, International Association for Food Protection

Microflora of Minimally Processed Frozen Sold in Gaborone, Botswana

TINNA A. MANANI,1 ERNEST K. COLLISON,2 AND SISAI MPUCHANE2*

1University of Malawi, Lilongwe, Malawi; and 2University of Botswana, Gaborone, Botswana

MS 06-181: Received 28 March 2006/Accepted 13 June 2006

ABSTRACT Downloaded from http://meridian.allenpress.com/jfp/article-pdf/69/11/2581/1679939/0362-028x-69_11_2581.pdf by guest on 23 September 2021

Two hundred samples of minimally processed, frozen, and prepacked potato chips, , corn, and a variety of combined vegetables from supermarkets in Gaborone, Botswana, were examined microbiologically. Determination of aerobic mesophilic plate count, aerobic psychrotrophic plate count, lactic acid bacteria, yeasts and molds, coliforms, Listeria spp., and Staphy- lococcus aureus were done. Chips had the lowest mean log values for all of the microorganisms enumerated except yeasts and molds. The mean log values for single vegetables ranged from 3.6 to 9.1, 3.4 to 8.9, 2.9 to 5.6, and 2.1 to 6.5 log CFU/ g aerobic mesophilic plate count, aerobic psychrotrophic plate count, lactic acid bacteria, and yeasts and molds, respectively. The microbial profiles of peas and corn were almost similar (P Ͻ 0.001). The mean values for combined vegetables were clustered within 4.6 and 5.4 and 4.2 and 5.2 log CFU/g aerobic mesophilic plate count and aerobic psychrotrophic plate count, respectively. All of the vegetables had a coliform population distribution ranging from 0 to Ͻ104 most probable number per g. The predominant gram-negative bacteria isolated included members of Enterobacteriaceae and Pseudomonaceae (86.2%). Escherichia coli was not detected in all of the samples. The organisms isolated included those responsible for spoilage in frozen vegetables, namely Pseudomonas, Klebsiella, Corynebacterium, lactic acid bacteria, and Flavobacterium. The predom- inant lactic acid bacteria were Lactobacillus spp. (55.9%). Other spoilage organisms were yeasts, and Cryptococcus spp. (55.4%) was predominant. Pathogens, namely Listeria monocytogenes, were also isolated at a rate of 2 to 10%, of which 4% was from corn, 2% each from peas and country crop, and 10% from stir-fry. Bacillus cereus was also isolated and accounted for 7.7% of the microorganisms from corn. S. aureus was isolated from all of the vegetables. Enterotoxigenic strains were from corn, peas, mixed vegetables, and stir-fry, and all of them produced enterotoxin A. In addition, the isolates from stir-fry vegetables also produced enterotoxins B and C. The study reveals the presence of pathogens and emerging opportunistic pathogens in the ready-to-use or ready-to-eat vegetables. If E. coli is the only indicator for safety and acceptability, consumers may be exposed to foodborne diseases. Inclusion of other groups as indicator organisms is suggested. Retailers are urged to invest in standby generators to maintain the cold chain.

The consumer demand for high-quality foods requiring MP vegetables is from industrialized countries, and infor- only a minimal amount of effort and time for preparation mation is lacking regarding microbial profiles, ecology, has led to the introduction of minimally processed (ready- quality, and safety of MP vegetables in developing coun- to-use or ready-to-eat) convenient foods preserved by mild tries where the use of such food is on the increase. Thus, methods (1). Minimally processed (MP) vegetables consist this study was undertaken to determine the microbial qual- of raw fresh cut produce, which has undergone minimal ity of a variety of MP vegetables sold in retail supermarkets processing such as trimming, peeling, slicing, shredding, in Botswana. cutting, washing, and blanching before packaging in sealed Most of these products are imported into the country pouches or in plastic trays. and transported by road. The efficiency of the maintenance The potential for microbial contamination is high be- of the cold chain to wholesalers and retailers and conditions cause of the wide variety of conditions to which the pro- during sale would have a bearing on microbiological qual- duce is exposed during growth, harvest, processing, and ity of the consumed vegetables. distribution (15). Slicing and shredding procedures as well as improper refrigeration during storage have been associ- MATERIALS AND METHODS ated with an increase in the number of mesophilic and psy- Sample collection. Two hundred MP vegetables were pur- chrotrophic aerobic microorganisms (2). One of the psy- chased in frozen state from various supermarkets in Gaborone. chrotrophic bacterial pathogens is Listeria monocytogenes, These vegetables are packed in sealed plastic bags, which are which is capable of growing at chill temperatures that are usually displayed in open freezer units. The samples comprised heavily relied upon by the food industry to maintain prod- 50 each of combined vegetables, corn, peas, and potato chips. The uct quality (23) and to extend product shelf life (1). samples were purchased from five supermarkets, and from each Most published information on the microbial quality of supermarket, 10 samples of corn, peas, and chips were bought. The numbers of combined vegetables were determined by avail- * Author for correspondence. Tel: ϩ267 3552598; Fax: ϩ267 3909476; ability from different supermarkets. The samples were transferred E-mail: [email protected]. to the laboratory in ice coolers within an hour of collection and 2582 MANANI ET AL. J. Food Prot., Vol. 69, No. 11 sampling was done within 24 h of collection. Combined bation at 35ЊC for 24 and 48 h. Tubes showing growth and gas samples refer to a mixture of different types of vegetables in one were considered positive for total coliforms. Fecal coliforms were pack. The packaged vegetables used were 10 samples of stir-fry also enumerated by the three-tube MPN method using lauryl tryp- (cabbage, beans, carrots, and onions); 10 samples of mixed veg- tose broth, incubated at 44ЊC for 24 and 48 h. A loopful of culture etables (carrots, onions, baby marrows, beans, peppers, and cab- showing growth and gas was streaked on eosin methylene blue bage); 16 samples of country crop (cauliflower florets, carrot roun- agar (Oxoid). Typical E. coli colonies produced a green metallic dels, cut green beans, and broccoli cuts); 5 samples of vegetable sheen color on eosin methylene blue agar. curry (cauliflower, carrots, peas, corn, beans, potatoes, and curry E. coli. The samples were cultured using a standard plate spices); 9 samples of Mexican mix (broccoli, cauliflower, corn, technique on sorbitol MacConkey agar (HiMedia, Mumbai, India) red peppers, and spices); and 50 samples each of single vegeta- for the isolation of E. coli. Purified isolates were subjected to bles, which refers to those packaged as single entities and include various morphological and biochemical tests including API 20E potato chips, corn, and peas. galleries. Identification was carried out using the ATB reader (bioMe´rieux)

Sample preparation. The packaged contents of the vegeta- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/69/11/2581/1679939/0362-028x-69_11_2581.pdf by guest on 23 September 2021 bles were mixed, and a total of 50 g was aseptically transferred LAB. Lactic acid bacteria (LAB) were enumerated on MRS into a sterile stomacher bag, and 450 ml of sterile 0.1% peptone (Oxoid) agar (14). The samples were cultured using the pour plate water was added. The sample-diluent mixture was blended for 4 method with an overlay (11) and were incubated at 37ЊC for 24 min in a Stomacher 400 laboratory blender (Seward Medical, Lon- to 48 h, after which representative colonies were isolated, purified, don, UK). Serial, 10-fold dilutions 10Ϫ1 to 10Ϫ6 were made using and maintained on MRS agar slants at 4ЊC for identification. the blended sample and 90-ml aliquots of 0.1% peptone water for Gram positives, which were catalase negative and were iso- analysis of aerobic (mesophiles and psychrotrophs) bacterial, co- lated from MRS agar, were treated as presumptive LAB. These liform, yeast and mold, and lactic acid bacterial counts. Samples isolates were characterized and identified using API 50CH and were also analyzed for the presence of Escherichia coli, Staphy- API 50CHL medium (bioMe´rieux), according to the manufactur- lococcus aureus, and Listeria. er’s specifications. Enumeration and detection of microorganisms. Aerobic S. aureus. Triplicate spread plate technique was used on mesophilic plate counts and aerobic psychrophilic plate counts Baird-Parker (Merck, Whitehouse Station, N.J.) supplemented were done as follows: Six replicate plates were prepared with plate with egg yolk tellurite emulsion FD 046 (HiMedia) and the plates count agar (Oxoid, Basingstoke, UK) using the standard pour plate were incubated at 37ЊC for 24 h. Typical representative colonies Њ method. One set of triplicate plates was incubated at 37 C for 24 from each plate were picked and subcultured on nutrient agar Њ h for mesophilic counts and the other set at 7 C for 10 days for slants (Oxoid) for further tests. psychrotrophic counts (24). After incubation, colonies were count- Gram-positive cocci isolated as presumptive S. aureus from ed, and representative colonies were isolated, purified, and sub- Baird-Parker medium were tested for coagulase activity using rab- Њ cultured on nutrient agar slants and were maintained at 4 C until bit coagulase plasma PL850 (Pro-Lab, Austin, Tex.). Organisms further characterization. that showed a positive result for coagulase tests were subjected to Biochemical characterization, identification, and confir- API STAPH (bioMe´rieux) for confirmation. The isolates con- mation. A total of 727 colonies were isolated (348 were from firmed as S. aureus were tested for toxin production using SET- plate count agar, 167 from Baird-Parker agar, 111 from deMan RPLA staphylococcal enterotoxin test kit TD900 (Oxoid) accord- Rogosa Sharpe [MRS], and 101 from dichloran rose bengal chlor- ing to the manufacturer’s specifications. amphenicol [DRBC] agar base [Oxoid]). Three to four represen- Listeria spp. A 10-g sample from each product was added tative colonies were selected based on colony morphology from to 90 ml of Listeria selective enrichment broth (Oxoid) supple- a plate of between 30 and 300 colonies. Initial characterization mented with Listeria selective enrichment supplement SR141E involved Gram staining, oxidase test using oxidase identification (Oxoid), and then blended in a Stomacher 400. Following enrich- sticks (Oxoid), and catalase test using 3% hydrogen peroxide. ment at 30ЊC for 48 h, 0.1 ml of the sample was directly spread Spore staining was done on gram-positive rods. Voges-Proskauer, plated on Listeria selective agar base (Oxoid) and incubated at motility, indole, and urease tests were done on gram-negative or- 30ЊC for 48 h. Suspect colonies were treated as presumptive Lis- ganisms (5). Gram-negative organisms were characterized as en- teria colonies and were subcultured on tryptic soy agar (Merck) teric or nonenteric by inoculating on triple sugar iron agar slants with 0.5% yeast extract for purity (23) and characterized. (Oxoid), which were examined for yellow butt and red slants after Organisms identified as Listeria by the Biolog system were Њ incubation at 37 C for 24 h. confirmed using the Listeria rapid test FT40IM (Oxoid) according to the manufacturer’s specifications. Presumptive Listeria species Gram-positive and gram-negative bacteria. The Biolog were inoculated in 10 ml of Listeria selective broth (Oxoid) sup- system was used to identify and characterize gram-negative and plemented with Oxoid SR 166M half Fraser supplement for pri- gram-positive rods including Bacillus spp. and presumptive Lis- mary enrichment. All supplements were reconstituted and added teria spp. according to the manufacturer’s specifications. The fre- to the broths in accordance with the manufacturer’s instructions. quency of the occurrence of each organism was calculated as the number of isolates of a species over the total number of isolates. Yeast and mold culture. Yeasts and molds were isolated on Using this frequency, the pathogens most commonly associated DRBC agar base supplemented with chloramphenicol selective with the ready-to-use or ready-to-eat vegetables were identified. supplement SR078E. Incubation was done at 25ЊC for 5 days, Members of the families Pseudomonaceae and Enterobacteria- after which yeast isolates were purified and maintained by sub- ceae were confirmed using API 20E strips (bioMe´rieux, Marcy- culturing on malt extract agar (Oxoid). Counts were made of mold l’Etoile, France). colonies. Coliforms. The three-tube most-probable-number (MPN) Identification of yeasts. Yeast classification involved deter- method using lauryl tryptose broth (Oxoid) was used with incu- mining morphological characters such as pigmentation of the col- J. Food Prot., Vol. 69, No. 11 MINIMALLY PROCESSED FROZEN VEGETABLES 2583

TABLE 1. Microbial mean total viable counts in chips, corn, and TABLE 2. Microbial mean total viable counts in combined ve- peasa getablesa Sample type Chips Corn Peas Mixed Vegetable Country (type of organism) (n ϭ 50) (n ϭ 50) (n ϭ 50) Sample type vegetable Stir-fry curry crop (type of organism) (n ϭ 10) (n ϭ 10) (n ϭ 5) (n ϭ 16) Mesophiles 3.6 Ϯ 0.7 4.6 Ϯ 0.7 4.5 Ϯ 0.8 Psychrotrophs 3.4 Ϯ 0.8 4.8 Ϯ 0.8 4.2 Ϯ 0.9 Mesophiles 5.3 Ϯ 0.4 4.9 Ϯ 0.7 5.4 Ϯ 0.3 4.6 Ϯ 0.8 LAB 2.9 Ϯ 1.2 4.4 Ϯ 0.9 3.8 Ϯ 0.9 Psychrotrophs 5.2 Ϯ 0.5 4.5 Ϯ 0.3 4.2 Ϯ 0.4 4.5 Ϯ 0.5 Yeasts and molds 3.4 Ϯ 0.9 2.1 Ϯ 0.7 2.4 Ϯ 1.4 LAB 3.9 Ϯ 0.6 3.8 Ϯ 0.7 3.6 Ϯ 0.6 4.8 Ϯ 0.4 Yeasts and molds 4.0 Ϯ 0.8 2.5 Ϯ 1.1 1.7 Ϯ 0.4 3.7 Ϯ 0.6 a Values are means Ϯ standard deviations expressed in log CFU per gram. a Values are means Ϯ standard deviations expressed in log CFU per gram. ony on both DRBC agar base supplemented with chloramphenicol Downloaded from http://meridian.allenpress.com/jfp/article-pdf/69/11/2581/1679939/0362-028x-69_11_2581.pdf by guest on 23 September 2021 selective SR078E and malt extract agar and cellular morphology ations. In Botswana, it has been observed that some people of vegetative cells using light microscopy. Yeast species were fi- use frozen peas without heating. The peas are thawed and nally classified and identified using physiological tests (4), which added to various salads or stews. Inadequate processing involve assimilation of different carbohydrates by yeasts (11). leading to survival of thermotolerant bacteria prior to freez- This identification was done using API 20 C AUX (bioMe´rieux) ing could result in a high microbial load that could be a per manufacturer’s instructions. health hazard when pathogens are present. Statistical analysis. The mean values obtained were ana- Total aerobic viable counts of combined vegetables. lyzed using SigmaStat software, version 2.0, by applying Stu- The lowest aerobic mesophilic plate count was 4.6 log Ͻ dent’s t test in order to determine significant differences (P CFU/g from country crop and the highest was 5.4 log CFU/ 0.001) between pairs of single and combined vegetables. g from vegetable curry (Table 2). The psychrotrophic val- RESULTS AND DISCUSSION ues were between 4.2 and 5.2 log CFU/g. The ranges for mean log values for LAB and yeasts and molds were 3.6 Total aerobic viable counts of single vegetables. to 4.8 and 1.7 to 4.0 log CFU/g, respectively. Mexican mix Vegetables are liable to contamination by soilborne micro- had the highest LAB mean value of 4.8 log CFU/g, and the organisms. However, preparation before minimally process- difference was statistically significant (P Ͻ 0.001) from ing the vegetables reduces the microbial population (19). values of all other combined vegetables. The processing of vegetables prior to packaging and freez- The number of microorganisms in frozen mixed veg- ing is critical to their safety and quality. Chips had the etables can range from below 102 CFU/g to above 107 lowest mean count for all of the organisms enumerated with CFU/g but usually from 104 to 106 CFU/g (21). This is the exception of yeasts and molds (Table 1). A comparison corroborated by this study in which mean values of aerobic between pairs of vegetables showed that the difference in plate counts for combined vegetables lie within the range mesophilic, psychrotrophic, and yeast and mold mean val- of 4.2 to 5.4 log CFU/g (Table 2). ues between corn and peas were not statistically significant The differences in microbial profiles between these (P Ͻ 0.001). Lactic acid bacterial mean values were all vegetables could be due to the differences in the vegetable statistically significant when pairs of the vegetables were combination. Some combinations contained vegetables that compared. cannot be blanched like onions and peppers (22). Thus, Chips had the lowest mean values of 3.6, 3.4, and 2.9 reduction in microbial load of these vegetables depends on log CFU/g mesophiles, psychrotrophs, and LAB, respec- good manufacturing practices from the field to the shop tively (Table 1). The lower microbial load could be due to shelves. different thermal treatments during blanching (70.4 to 78.7ЊC). However, the microorganisms present could orig- Coliforms. Total coliforms were detected in all of the inate from several subsequent processing steps, which in- vegetables. The MPN values for the chips, corn, , mixed clude handling, packaging, and weighing. The ranges for vegetable, stir-fry, country crop, vegetable curry, and Mex- mean log aerobic mesophilic plate count of chips in this ican mix samples had the following ranges, respectively: study are comparable to those of other studies (6) where Ͻ3to9,Ͻ3 to 240, Ͻ3 to 4,600, 93 to 2,400, 4 to 460, mean log aerobic plate count ranges of Ͻ2.0 to 4.3 log 9 to 1,100, 4 to 4, and 93 to 460. Fecal coliforms were not CFU/g and Ͻ2.0 to 6.7 log CFU/g, respectively, were re- detected by the MPN method probably due to low numbers ported. but were isolated from eosin methylene blue agar. E. coli Peas and corn were of the same microbial quality; the was also not detected in all samples. Because the Interna- differences in mean values were not statistically significant tional Commission on Microbiological Specifications for (P Ͼ 0.001). Peas and corn just like chips undergo blanch- Foods has proposed the use of E. coli as an indicator for ing. The presence of microorganisms therefore suggests microbial quality acceptability of blanched vegetables (20), post blanching and processing contamination. In addition, its absence in these vegetables may indicate good vegetable temperature fluctuations in retail shops and possibly during quality or sanitary processing conditions. Occurrence of transportation and loading from manufacturers outside the pathogenic or opportunistic organisms suggests that inclu- country to the distributors could allow microbial prolifer- sion of other indicators of food safety is required. This is 2584 MANANI ET AL. J. Food Prot., Vol. 69, No. 11

TABLE 3. Numbers of Enterobacteriaceae, Pseudomonaceae, and include Pseudomonas fluorescenes, Flavobacterium spp., gram-positive bacteria isolated from different samples and their and Trichosporon spp. frequency of occurrence (percentage) on various vegetables LAB. The most predominant LAB of the 111 isolates Combined Organism(s) vegetables Corn Chips Peas identified were the Lactobacillus spp. (55.9%). Other LAB identified were the Leuconostoc spp. (36.9%) and Lacto- Acinetobacter spp. 2 (1.7) 4 (6.2) 2 (2.0) coccus spp. (7.2%). Of the chip samples, 18% had Lacto- Aeromonas sp. 2 (3.1) bacillus plantarum, 16% had Lactococcus lactis subsp. lac- Aureobacterium sp. 20 (20.4) tis, and 8% had Leuconostoc mesenteroides. In the corn Bacillus spp. 7 (5.8) 15 (22.1) 9 (16.7) 20 (20.4) samples, 6% had Lactobacillus paracasei subsp. paracasei, Brevibacterium spp. 7 (5.8) 5 (7.7) 36% had L. plantarum, 6% had Leuconostoc citreum, 14% Carnobacterium sp. 7 (5.8) had Leuconostoc lactis, and 22% had L. mesenteroides. In Cellulomonas sp. 7 (5.8) Chryseomonas sp. 2 (1.7) 2 (2.0) peas, 18% were Lactobacillus brevis, 12% were L. plan- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/69/11/2581/1679939/0362-028x-69_11_2581.pdf by guest on 23 September 2021 Corynebacterium spp. 7 (5.8) 18 (33.4) 20 (20.4) tarum, 4% were L. citreum, 12% were Leuconostoc lactis, Curtobacterium sp. 14 (11.7) 9 (16.7) and 16% were L. mesenteroides. In the combined vegeta- Enterobacter spp. 10 (8.3) 2 (3.1) 13 (24.1) 10 (10.1) bles, Lactobacillus collinoides appeared in 4% of the sam- Flavobacterium spp. 2 (1.7) 2 (3.7) 4 (4.1) ples, L. paracasei subsp. paracasei in 24%, and Lactoba- Klebsiella spp. 15 (6.7) 4 (6.2) 2 (2.0) cillus rhamnosus in 6% of the samples. Listeria spp. 14 (11.7) 27 (41.6) 10 (10.2) Spoilage by LAB is achieved through fermentation. Pantoea spp. 10 (8.3) 4 (6.2) 2 (3.7) 2 (2.0) Homofermentative LAB produce lactic acid, whereas het- Pseudomonas spp. 8 (6.6) 2 (3.1) 4 (7.4) 4 (4.1) erofermentative LAB produce acetic acid, formic acid, and Serratia spp. 4 (3.3) 2 (2.0) CO in addition to lactic acid (16). Some strains of LAB Staphylococcus spp. 16 (4.8) 7 (4.2) 1 (0.6) 8 (4.8) 2 have remarkable inhibitory effect on growth dynamics of Stenotrophomonas sp. 4 (3.3) Vibrio spp. 6 (11.1) microflora associated with MP vegetables during refriger- ation. Strains isolated in this study that produce bacteriocins include Lactococcus lactis spp. lactis (13) and Lactobacil- lus casei (25). These strains may have uses as biopreser- imperative as a result of the large percentage of the popu- vatives when introduced in sufficient quantities in food and lation that is immunocompromised as a result of HIV/AIDS their activities are controlled. and the general poor nutrition of many in the population. Yeasts and molds from single and combined vege- Most gram-negative bacteria (86.2%) were classified as tables. The number of molds was insignificant compared members of Enterobacteriaeceae and Pseudomonaceae. with the yeasts that were cultured. This has been reported Sporeformers such as members of the genus Bacillus and in previous studies (18). Yeasts isolated belong to six gen- pathogens like L. monocytogenes were identified among the era and Cryptococcus spp. were the most dominant (60 of gram-positive isolates (Table 3). Pseudomonas aeruginosa 101; 59.4%). The percentage distribution of the other gen- was present in all of the vegetables. The organisms isolated era were as follows: Candida, 22.8%; Saccharomyces, from the vegetables can be categorized as common contam- 11.88%; Trichosporon, 2.97%; Rhodotorula, 1.98%; and inants, spoilage organisms, opportunistic pathogens, and Pichia, 0.99%. pathogenic organisms. The common contaminants include Eight percent of the chips sampled had Candida spp., those whose natural habitat is the soil and those that are and 42% had Cryptococcus spp. In the samples of peas commonly associated with vegetables. analyzed, 10% had Candida spp., 4% had Cryptococcus Gram-positive bacteria are more resistant to blanching spp., and 2% had Rhodotorula spp. From the corn samples, temperatures than gram negatives. Also, gram positives are 20, 18, 12, and 2% had Candida spp., Cryptococcus spp., more resistant to freezing than gram negatives (8, 19). That Saccharomyces spp., and Trichosporon spp., respectively. may be the reason why gram-negative bacteria were less Combined vegetables had the most numerous flora with 8, frequently identified than gram positives. The bacteria iso- 60, 2, 2, 12, and 4% having Candida spp., Cryptococcus lated are those that have been reported as present in MP spp., Pichia spp., Rhodotorula spp., Saccharomyces spp., vegetables with the exception of Flavobacterium, Pantoea, and Trichosporon spp., respectively. This work therefore and Aureobacter species. The predominant gram-negative corroborates findings of previous work (21). An insignifi- bacteria were the Enterobacteriaceae and Pseudomona- cant number of molds were isolated and no further work ceae. Enterobacter and some of the genera of Enterobac- was done with them. teriaceae are part of the natural microflora of vegetables (9), and pseudomonads are regarded as native microflora of Occurrence of pathogens. Pathogens isolated were fresh produce (14). Bacillus cereus from corn, S. aureus from combined veg- The spoilage organisms include Pseudomonas, Kleb- etables, corn, chips, and peas, and L. monocytogenes from siella, Flavobacterium, Corynebacterium, LAB, and yeasts. corn, combined vegetables, and peas (Table 4). These or- These have been reported to cause spoilage in frozen foods ganisms are considered as pathogens of most concern when (17). Populations of pectinolytic microorganisms have been isolated from vegetables, because they have been implicat- determined as spoilage agents (21), and in this study they ed in foodborne diseases in many countries. Using E. coli J. Food Prot., Vol. 69, No. 11 MINIMALLY PROCESSED FROZEN VEGETABLES 2585

TABLE 4. Pathogens and emerging opportunistic pathogens iso- as low as 2ЊC and can thrive in cool, wet areas in process- lated from MP vegetables and their frequency (percentage) on ing facilities (3). L. monocytogenes is an etiologic agent of various vegetables a foodborne listeriosis. Nonpregnant healthy individuals Combined who are not immunosuppressed are highly resistant to in- Organism(s) vegetables Corn Chips Peas fection by this organism. However, neoplasm, AIDS, al- coholism, diabetes (type 1), cardiovascular diseases, renal Bacillus cereus 5 (7.7) transplant, and corticosteroid therapy are known to predis- Corynebacterium diphtheria 9 (16.7) pose to adult listeriosis and to be significant in mortality Enterobacter sakazaki 2 (2.0) Klebsiella pneumoniae 4 (6.2) rate (12). In Botswana, many of these ailments predominate Listeria monocytogenes 14 (11.7) 5 (7.7) 10 (10.2) and proper management of the quality of food is critical. Pseudomonas aeruginosa 8 (6.6) 2 (3.1) 4 (7.4) 4 (4.1) Freezing remains one of the most important methods Staphylococcus aureus 16 (4.8) 7 (4.2) 1 (0.6) 8 (4.8) for retaining quality during long-term storage of MP veg- Vibrio spp. 6 (11.1) etables. The absence of E. coli in this study indicates good Downloaded from http://meridian.allenpress.com/jfp/article-pdf/69/11/2581/1679939/0362-028x-69_11_2581.pdf by guest on 23 September 2021 vegetable quality or good, sanitary processing conditions compared with incidence levels of 12.7 to 30% reported in as the only indicator for the quality of MP vegetables would another study (21). Therefore, the results show that, using not address the potential exposure to these pathogens. only E. coli as an indicator of sanitary quality, the conclu- sion would be that the frozen prepacked MP vegetables S. aureus. Twenty-four (14.4%) of the 167 Staphylo- analyzed were acceptable. However, the absence of E. coli coccus isolates were confirmed as S. aureus. Other staph- does not guarantee the safety of food. A good example is ylococci isolated were identified as S. hominis, S. sciuri, S. the vegetables sampled in this study. It would therefore be warneri, S. lugdunensis, S. lentus, and S. capitis. The fre- preferable if other indicators that are closely associated with quency of isolation of staphylococci was higher in peas and environmentally acquired opportunistic or pathogenic or- corn and the organisms were less frequent in chips and ganisms are also included in assessing and monitoring safe- country crop. ty of MP vegetables. The occurrence of S. aureus, a major cause of food- The presence of human pathogens such as L. mono- borne illness around the world from the country crop, peas, cytogenes, B. cereus, and S. aureus in these vegetables stir-fry, vegetable curry, mixed vegetables, corn, and chip should be considered as having significant public health samples was 6.3, 16, 20, 40, 60, 14, and 2%, respectively. implications, particularly among the immunocompromised Food handlers have been incriminated for contaminating population who are at greater risk. There is therefore a need ready-to-eat vegetables by S. aureus, which has been iso- to include the presence of other organisms especially those lated from potatoes and potato products (6), carrots, lettuce, from various environmental sources in our safety guide- seed sprouts, radishes (3), and many other vegetable prod- lines. The need for good manufacturing and handling prac- ucts (19). tices cannot therefore be overlooked, particularly because Enterotoxigenic S. aureus does not compete well with these products are transported for long distances in some- other microbes normally present on fresh produce. The times extremely high temperatures. The maintenance of the most common enterotoxin detected was A and is commonly cold chain after processing is critical in assuring good qual- involved in food poisoning. S. aureus enterotoxins are sig- ity products as is proper handling of the food and education nificant because they are unusually heat resistant, with- of food handlers and consumers. standing boiling in food for 30 min, with type B toxin Education of retailers and vigilance of retail assistants showing the greatest stability (7). on the importance of maintaining the cold chain cannot be Enterotoxin production was tested in 20 of the 24 con- overemphasized, considering the occurrence of power dis- firmed S. aureus isolates, and 6 (30%) were positive for ruptions that occur now and then and the high ambient tem- enterotoxin A, 1 (5%) for enterotoxin B, and 1 (5%) for peratures experienced. Retailers should be encouraged to enterotoxin C. S. aureus producing enterotoxins B and C invest in stand-by generators to serve during periods when were isolated from stir-fry. Enterotoxins B and C were pro- power failures occur. duced by the same organism, which also produced entero- toxin A. Two of the enterotoxin A–producing S. aureus ACKNOWLEDGMENTS were isolated from corn, two from peas, one from mixed vegetables, and one from stir-fry. Enterotoxin D was not We are grateful to the Belgian Overseas Corporation for funding Tinna’s expenses while pursuing her studies at the University of Botswana detected. and to the technical staff in the Department of Biological Sciences for L. monocytogenes. This organism was detected in 4% their assistance. of the corn samples, 6.3% of the country crop samples, 2% REFERENCES of the peas, and 10% of the stir-fry samples. The isolation rate of this organism in this study was 2 to 10% compared 1. Ahvenainen, R. 2000. Ready-to-use fruits and vegetables. Flair-Flow with the isolation rate of 0 to 19% in another study (23). Europe Technical Manual F-FE 376A/00. Available at: www.exp.hispeed.com/flair/360a00.pdf. Accessed May 2000. L. monocytogenes has been isolated on cabbage, po- 2. Babic, I., S. Roy, A. E. Watada, and W. P. Wergin. 1996. Changes tatoes, ready-to-eat salads, lettuce, tomatoes, broccoli, and in microbial populations on fresh cut . Int. J. Food Microbiol. cauliflower (10). The organism can grow at temperatures 31:107–119. 2586 MANANI ET AL. J. Food Prot., Vol. 69, No. 11

3. Beuchat, L. R. 1995. Pathogenic microorganisms associated with 15. Madden, J. M. 1992. Microbial pathogens in fresh produce—the reg- fresh produce. J. Food Prot. 59:204–216. ulatory perspective. J. Food Prot. 55:821–823. 4. Campbell, I. 1988. Culture, storage, isolation and identification of 16. Marth, E. H. 1998. Extended shelf life refrigerated foods: microbi- yeasts, p. 1–8. In I. Campbell and J. H. Duffus (ed.), Yeast: a prac- ological quality and safety. Food Technol. 52:57–62. tical approach. IRL Press, Oxford. 17. Mayes, T., and G. Telling. 1993. Product safety from factory to con- 5. Cheesbrough, M. 1984. Medical laboratory manual for tropical coun- sumer, p. 93–122. In C. P. Mallett (ed.), technology. tries, vol. II. Microbiology. Tropical Health Technology, Butter- Blackie Academic and Professional, London. worths, London. 18. Mislivec, P. B., and M. E. Stack. 1984. Enumeration of yeasts and 6. Doan, C. H., and P. M. Davidson. 2000. Microbiology of potatoes molds and production of toxins, p. 19.00–19.10. In U.S. Food and and potato products: a review. J. Food Prot. 63:668–683. Drug Administration Bacteriological analytical manual, 6th ed. As- 7. Forsythe, S. J., and P. R. Hayes. 1998. Food hygiene, microbiology, sociation of Official Analytical Chemists, Arlington, Va. and HACCP, 3rd ed. Aspen Publishers, Gaithersburg, Md. 19. Mpuchane, S. F., and B. A. Gashe. 1995. Prevalence of coliforms in 8. Golden, D. A., and L. Arroyo-Gallyoun. 1997. Relationship of fro- traditionally dried leafy vegetables sold in open markets and food zen-food quality to microbial survival, p. 174–193. In M. C. Erick- stores in Gaborone, Botswana. J. Food Prot. 59:28–30. 20. National Advisory Committee on Microbiological Criteria for Foods

son and Y. Hung (ed.), Quality in frozen food. Chapman and Hall, Downloaded from http://meridian.allenpress.com/jfp/article-pdf/69/11/2581/1679939/0362-028x-69_11_2581.pdf by guest on 23 September 2021 International Thomson Publishing, New York. (NACMCF). 1999. Microbiological safety evaluations and recom- 9. Harrigan W. F. 1998. Laboratory methods in food microbiology, 3rd mendations on fresh produce. Food Control 10:117–143. ed. AP Academic Press, San Diego, Calif. 21. Nguyen-the, C., and F. Carlin. 2000. Fresh and processed vegetables, p. 620–684. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould 10. Heisick, J. E., D. E. Wagner, M. L. Nierman, and J. T. Peeler. 1989. (ed.), The microbiological safety and quality of food, vol. I. Aspen Listeria spp. found on fresh market produce. Appl. Environ. Micro- Publishers, Gaithersburg, Md. biol. 55: 1925–1927. 22. Roberts, T., and R. Cox. February 1999. Drying fruits and vegeta- 11. Jacxsens, L., F. Devlieghere, C. Van der Steen, and J. Debevere. bles. Virginia Cooperative Extension, Virginia State University. 2001. Effect of high oxygen modified atmosphere packaging on mi- Available at: www.ext.vt.edu/pubs/foods/348࿞597/348-597.html. Ac- crobial growth and sensorial qualities of fresh-cut produce. Int. J. cessed 16 October 2001. Food Microbiol. 71:197–210. 23. Szabo, E. A., K. J. Scura, and J. M. Burrows. 2000. Survey for 12. Jay, J. M. 1996. Modern food microbiology, 5th ed. Chapman and psychrotrophic bacterial pathogens in minimally processed lettuce. Hall, New York. Lett. Appl. Microbiol. 30:456–460. 13. Kelly, W. J., G. P. Davey, and L. J. Ward. 1998. Characterization of 24. Vanderzant, C., and D. F. Splittstoesser (ed.). 1992. Compendium Lactococci isolated from minimally processed fresh fruit and vege- methods for the microbiological examination of foods, 3rd ed. Amer- tables. Int. J. Food Microbiol. 45:85–92. ican Public Health Association, Washington, D.C. 14. Liao, C., and W. F. Fett. 2001. Analysis of native microflora and 25. Vescovo, M., C. Orsi, G. Scolari, and S. Torriani. 1995. Inhibitory selection of strains antagonistic to human pathogens on fresh pro- effect of selected lactic acid bacteria on microflora associated with duce. J. Food Prot. 64:1110–1115. ready-to-use vegetables. Lett. Appl. Microbiol. 21:121–125.