Chili peppers are a very important crop in Mexico. However, these peppers have been associated with Salmonella infection outbreaks in the United States, and Salmonella and diarrheagenic Escherichia coli pathotypes have been isolated from jalapeño and serrano peppers in Mexico. To decrease microbial contamination of fruits and vegetables, chemical agents are commonly used; however, chemical agents used to eliminate pathogenic bacteria on vegetables have a limited antimicrobial effect. Roselle (Hibiscus sabdariffa) calyces have been reported to have an antimicrobial effect on pathogenic bacteria. In the present study, the antibacterial effect of four roselle calyx extracts (water, methanol, acetone, and ethyl acetate), sodium hypochlorite, colloidal silver, and acetic acid against foodborne bacteria was evaluated on contaminated jalapeño and serrano peppers. The 13 types of foodborne bacteria evaluated were Listeria monocytogenes, Shigella flexneri, Salmonella Typhimurium, Salmonella Typhi, Salmonella Montevideo, Staphylococcus aureus, E. coli O157:H7, five E. coli pathotypes (Shiga toxin producing, enteropathogenic, enterotoxigenic, enteroinvasive, and enteroaggregative), and Vibrio cholerae O1. All 13 types attached to both pepper types, with no significant differences in attachment between jalapeño and serrano peppers. Roselle calyx extract treatment resulted in a greater reduction in levels of all foodborne bacteria than did treatment with sodium hypochlorite, colloidal silver, and acetic acid on both pepper types. Roselle calyx extracts may be a useful for disinfection of chili peppers in the field, processing plants, restaurants, and homes.
In recent years, demand for fresh food products has been increasing; however, foodborne illness outbreaks also have become more common with this increased intake of fresh fruit and vegetables (17). A wide spectrum of pathogenic bacteria has been implicated in foodborne disease outbreaks around the world (17, 33). Bacterial contamination of whole or minimally processed fresh vegetables can occur at any one of the processing stages, i.e., harvest, trimming, washing, slicing, soaking, dehydrating, blending, and/or packaging (33). In developing countries such as Mexico, the most common pathogens reported have been Salmonella, Shigella, Staphylococcus aureus, Listeria monocytogenes, and Vibrio cholerae, with diarrheagenic Escherichia coli pathotypes recently coming into prominence (1, 11, 14, 15, 24, 46, 49, 53). V. cholerae O1 is a halophilic marine bacterium commonly isolated from ocean water and marine food. However, V. cholerae O1 has been isolated from sewage water and the environment in Mexico (3, 22, 25, 54) and from various vegetables in other countries (47). Therefore, V. cholerae O1 can contaminate raw vegetables such as peppers via contaminated sewage water or the environment. Chili peppers are major commercial crops in Mexico, accounting for over 2,732,635 tons of production in 2014; 940,210 tons were jalapeño peppers and 261,798 tons were serrano peppers (52). A recent outbreak of Salmonella Saintpaul infection in North America was linked to jalapeño and serrano peppers (16) and affected at least 1,442 people in 43 U.S. states, the District of Columbia, and Canada. The jalapeño peppers were traced back to distributors in the United States that had received produce grown and packed in Mexico.
Salmonella and diarrheagenic E. coli pathotypes also have been isolated from jalapeño and serrano peppers purchased in public markets in Mexico (14, 19). Salmonella was detected on 10 and 12% of jalapeño and serrano peppers, respectively (14), and diarrheagenic E. coli pathotypes, including enterotoxigenic E. coli (ETEC) and Shiga toxin–producing E. coli (STEC), were identified on 36 and 14% of jalapeño and serrano peppers, respectively (19). STEC was isolated from 36% of serrano and 14% of jalapeño samples, and ETEC was isolated from 12% of serrano and 2% of jalapeño samples. Both pathotypes were identified in 14% of serrano and 2% of jalapeño samples. No E. coli O157:H7 was detected in any STEC-positive samples (19).
To decrease microbial contamination of fruits and vegetables, disinfection agents such as organic acids and sodium hypochlorite are commonly used. However, in various studies organic acids and sodium hypochlorite have had a limited or ineffective antimicrobial effect on pathogenic microorganisms on vegetables (13). The use of chemical disinfectants is suspected to be environmentally unsound due to possible occupational and operational hazards, and many of these chemicals are potentially harmful to humans (48). Consequently, novel technologies or chemical agents that are more effective and meet high current environmental standards are needed.
Roselle (Hibiscus sabdariffa) calyces have been reported to have an antimicrobial effect on pathogenic bacteria (41, 45). In previous studies, roselle calyx extracts have been used to disinfect romaine lettuce and alfalfa sprouts (41) and organic leafy vegetables (44). We reported antimicrobial effect of roselle calyx extracts against multidrug-resistant Salmonella strains on carrots (31) and tomatoes (32). Thus, roselle extracts can help control foodborne bacteria on fruits and vegetables. However, no data are available on the antibacterial activity of roselle calyx extracts against foodborne bacteria on chili peppers. The present study objectives were to investigate the ability of 13 types of foodborne bacteria to attach to jalapeño and serrano peppers and to compare the antibacterial effect of roselle calyx extracts, sodium hypochlorite, acetic acid, and colloidal silver against these 13 pathogens on both types of peppers.
MATERIALS AND METHODS
Bacterial strains. The foodborne bacterial strains used in the study were Salmonella Typhimurium (ATCC 14028); L. monocytogenes (ATCC 19115); S. aureus (ATCC 25923); Shigella flexneri (ATCC 12022); Salmonella Typhi (strain SS6, isolated from raw tomatoes (32)); Salmonella Montevideo (strain Z6, isolated from raw carrots (31)); E. coli O157:H7 (strain E09, donated by E. F. Escartín, Universidad Autónoma de Querétaro, Mexico); V. cholerae O1 (strain 87151, serotype Inaba, donated by E. F. Escartín); three strains each of non-O157 STEC, enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), and enteroinvasive E. coli (EIEC); and one strain of enteroaggregative E. coli (EAEC). The three non-O157 STEC strains were STSP41 (locus stx1) and STJP6 (locus stx2) isolated from serrano and jalapeño peppers, respectively (19), and STSCM23 (loci stx1 and stx2) isolated from ready-to-eat salads (12). The three EPEC strains were EPNW7 and EPNC6 isolated from raw nopalitos (Opuntia ficusindica L. (28)) and EPCSA225 isolated from a ready-to-eat cooked vegetable salad (5). The three ETEC strains were ETSP7 and ETJPI isolated from serrano and jalapeño peppers, respectively (19), and ETSAS22 from ready-to-eat salads (12). The three EIEC strains were EISCM13 isolated from ready-to-eat salads (12), EICJB121 from carrot juice (59), and EIMS79 from mung bean sprouts (18). The single EAEC strain was kindly donated by Dr. J. F. Cerna-Cortes (Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, Mexico). Mutant strains resistant to rifampin (1,000 mg/liter; R+) were obtained for all experiment strains (15). These R+ strains were streaked onto Trypticase soy agar (TSA) slants and maintained at 3 to 5°C, with weekly transfers onto TSA. All strains maintained rifampin resistance throughout the study.
Roselle calyx extracts. Roselle calyx extracts were produced following a previously described method (23, 45). A sample (5 kg) of dehydrated roselle (H. sabdariffa var. Criolla de Oaxaca) calyces was used. Dried calyces (500 g) were weighed and placed in sterile glass flasks, and 2 liters of methanol, acetone, or ethyl acetate (the concentration of each solvent was 99.8%) was added to each flask. The flasks were then sealed and stored at room temperature for 7 days, after which the liquid phase was filtered through filter paper and concentrated in a rotary evaporator (V-800 vacuum controller, BÜCHI Labortechnik AG, Flawil, Switzerland). Solvents (methanol, acetone, and ethyl acetate) were eliminated from the concentrates by placing them in a recirculating air incubator (Ambi-Hi-Low Chamber, Labline Instruments, Kochi, Kerala, India) for 24 h at 45 ± 1°C.
An aqueous extract was produced by placing dried calyces (500 g) in a sterile glass flask, adding 5 liters of distilled water, heating the mixture to boiling for 10 min, and then allowing it to cool to room temperature. Water was eliminated from the concentrate as described above. All dried concentrated extracts were stored in sterile plastic bags at room temperature until used.
A solution was prepared from each aqueous, methanol, acetone, and ethyl acetate roselle calyx extract concentrate (5 g each) using deionized water (95 ml) for the aqueous and methanol extracts and a solution of Tween 80:distilled water (100 ml; 2:10, vol/vol) for the acetone and ethyl acetate extracts (final extract concentration of 50 mg/ml).
Disinfectant solutions. Disinfectant solutions were prepared following previously described procedures (31). A total 1-liter solution of sodium hypochlorite (pH 5.0, 200 mg/liter free chlorine) was prepared in deionized water from a sodium hypochlorite solution (10%; Sigma-Aldrich, St. Louis, MO). Free available chlorine was determined by iodometric titration (30). A 1-liter solution of 0.5% acetic acid was prepared using glacial acetic acid (Sigma-Aldrich) and distilled water. A 1-liter solution of colloidal silver (~3.5 mg/liter) was prepared in deionized water from a commercial solution (Microdyn, 0.35% colloidal silver; Microdyn, Tavistock Group, Miguel Hidalgo, Mexico) according to the product label (8 drops per liter of water).
Jalapeño and serrano pepper preparation. Jalapeño and serrano peppers were purchased in a public market of the city of Pachuca (Hidalgo State, Mexico). Upon purchase, the peppers were placed in sterile bags, transported to the laboratory, and processed within 1 h. Only peppers free of visible defects (e.g., bruises, cuts, and abrasions) were used. In the laboratory, peppers were manually cleaned with a cloth to remove dust and then disinfected with a sodium hypochlorite solution (pH 5.0, 200 mg/liter free chlorine). Free available chlorine was determined by iodometric titration (30). Peppers were immersed in the sodium hypochlorite solution for 10 min and then washed for 1 min with sterile tap water to eliminate any sodium hypochlorite residue. Peppers were then dried in a sterilized strainer and placed on a sterile stainless steel tray in a laminar flow biosafety hood at room temperature for 2 h to remove surface moisture. Before inoculation, peppers were maintained at room temperature (~25°C).
Inocula preparation and inoculation. Trypticase soy broth tubes (3 ml) were inoculated with individual R+ Salmonella Typhimurium, Salmonella Montevideo, Salmonella Typhi, L. monocytogenes, S. aureus, S. flexneri, E. coli O157:H7, V. cholerae O1, STEC, EIEC, EPEC, ETEC, and EAEC strains and incubated at 35°C for 18 h. The cultures were washed twice in sterile isotonic saline solution (0.85% NaCl) by centrifugation at 1,507 × g for 20 min, and the pellets were resuspended in sterile 0.1% peptone water at about 109 CFU/ml. For the multiple EPEC, ETEC, EIEC, and STEC strains, an inoculation cocktail for each pathotype was prepared from the three strains by mixing 1 ml of each washed suspension. Individual suspensions of Salmonella Typhimurium, Salmonella Montevideo, Salmonella Typhi, L. monocytogenes, S. aureus, S. flexneri, E. coli O157:H7, V. cholerae O1, and EAEC strains or cocktail suspensions of each EPEC, ETEC, EIEC, and STEC pathotype were diluted in sterile 0.1% peptone to yield approximately 106 CFU/ml (6 log CFU/ml).
Whole jalapeño and serrano peppers were inoculated with approximately 104 CFU (4 log CFU) from each individual suspension of the 13 bacterial types by placing 10 μl (from a bacterial suspension of approximately 106 CFU/ml) inside a circle (approximately 0.5 cm diameter) marked on the external surface (14). After inoculation, peppers were stored at 25 ± 2°C for 30 min to promote bacteria adherence. The inoculated area was then washed for 1 min with sterile tap water from a wash bottle to remove nonadhered inoculated cells.
Disinfection treatments and microbiological counts. Disinfection treatments and microbiological counts were made following the previously described procedures (31). Aliquots (20 ml) of each calyx extract, sodium hypochlorite, acetic acid, colloidal silver, or control isotonic saline solution (ISS; 0.85% sodium chloride) were placed in 1-liter precipitate cups. The inoculated portion of each pepper was immersed in a treatment solution or ISS for 5 min. After removal from the cup, the inoculated and treatment area (marked circle) was washed for 1 min with sterile tap water to eliminate any treatment solution residue. Pathogenic bacteria were quantified by removing the treated and washed area from each pepper with a sterile knife to a depth of approximately 0.5 cm (the portion or piece removed from each pepper was approximately 1 and 0.5 g for jalapeño and serrano peppers, respectively). Each extracted pepper section was placed in a separate sterile bag (each sterile bag was one sample) containing 10 ml of 0.1% sterile peptone water, and the sealed bag was manually massaged for 2 min. Bacterial counts were obtained from appropriate dilutions of the bacterial suspensions spread on TSA plates containing 100 mg/liter rifampin and incubated at 35 ± 2°C for 24 to 48 h.
To confirm the presence of R+ strains, three to five colonies from various plates were transferred to plates of bismuth sulfite agar (Bioxon, BD, Mexico City, Mexico), Salmonella-Shigella agar (Bioxon, BD), Baird-Parker agar (Bioxon, BD), modified Oxford medium (Difco, BD, Sparks, MD), thiosulfate–citrate–bile salts–sucrose agar (Bioxon, BD), sorbitol MacConkey agar (Bioxon, BD) containing tellurite-cefixime (2.5 and 0.05 mg/liter; Sigma-Aldrich), and eosin methylene blue agar (Bioxon, BD) for Salmonella (serovars Typhimurium, Montevideo, and Typhi), Shigella, S. aureus, L. monocytogenes, V. cholera O1, E. coli O157:H7, and the five E. coli pathotypes, respectively. Plates were then incubated at 35 ± 2°C for 24 h. Typical colonies isolated from each selective agar were confirmed using the somatic polyvalent (O) test for Salmonella (Institute of Epidemiological Diagnosis and Reference of the Health Secretary, Mexico). Shigella was biochemically identified with a commercial biochemical kit (API 20E, bioMérieux, Mexico City, Mexico), S. aureus was confirmed with the Staphyloslide latex test (BBL, BD, Sparks, MD), and L. monocytogenes was confirmed with Listeria O antiserum type 4 (Difco, BD). E. coli O157:H7 was confirmed with a latex agglutination test kit (RIM E. coli O157:H7 latex test kit, Remel, Lenexa, KS), and the E. coli pathotypes were confirmed only as E. coli with the API 20E test.
Experimental design. Each experiment (pepper type, disinfectant solution, roselle extract) with each individual R+ Salmonella Typhimurium, Salmonella Montevideo, Salmonella Typhi, L. monocytogenes, S. aureus, S. flexneri, E. coli O157:H7, V. cholerae O1, EAEC suspension or cocktail suspension of each pathotype (EPEC, ETEC, EIEC, and STEC) was run in three independent trials, and each trial was done in triplicate (three inoculated whole peppers per individual or cocktail of each pathogen were analyzed for each sampling time). Counts were done with two replicate plates from each sample.
Statistical analyses. Statistically significant differences (P < 0.05) were calculated with an analysis of variance and Duncan's test, using the STATISTICA ver. 8 program (StatSoft, Tulsa, OK).
RESULTS AND DISCUSSION
Attachment study. Contamination of fresh produce such as jalapeño and serrano peppers can occur as a result of preharvest factors such as the indigenous microbial population in the soil environment and manure added to soil as fertilizer, which ultimately influences the microbial load. Postharvest factors include handling, container sanitation, and food processing procedures (8).
Jalapeño and serrano peppers have a waxy cuticle that primarily functions as a semipermeable barrier against moisture and gas loss. The epicuticular wax covering the outer surface also repels water, which forms beads. Other cuticle functions include scattering short-wave radiation and preventing attachment of microorganisms (42).
In the present study, jalapeño and serrano peppers were inoculated directly on their surfaces; the pathogens were therefore expected to be present all over the peppers. No previous studies have indicated that preliminary disinfection of vegetables with sodium hypochlorite followed by water wash affect the colonization of bacteria.
All 13 types of foodborne bacteria attached to the jalapeño and serrano peppers at level shown in Table 1. In general, the attachment of these bacteria was not significantly different (P > 0.05) on jalapeño versus serrano peppers.
Various studies have been conducted on the behavior and survival of foodborne bacteria on fruits and vegetables. Most studies have focused on Salmonella and E. coli O157:H7 sprayed, immersed, or applied directly onto the foliage of plants, fruits, or vegetables by a range of techniques or applied to seeds, roots, or soil (38, 39, 55, 56, 58). When applied directly to foliage, both Salmonella and E. coli O157:H7 can survive on parsley in the field for 177 and 231 days, respectively (38, 39). On lettuce plants contaminated with E. coli O157:H7, the pathogens can survive for 30 days (55).
Recently, we reported the behavior of ETEC, EPEC, EIEC, and non-O157 STEC on whole jalapeño and serrano peppers (29). All of these bacteria survived well on both peppers for at least 12 days at 25 and 3°C.
The attachment, behavior, and survival of L. monocytogenes, Shigella, and S. aureus in fruits and vegetables has also been previously reported. Nevertheless, limited information is available on the attachment, behavior, and survival of E. coli pathotypes other O157and on V. cholerae O1 on vegetables such as jalapeño and serrano peppers. Attachment is a prerequisite for pathogen colonization and subsequent transmission via the edible parts of fruits and vegetables, such as jalapeño and serrano peppers. Once attached, pathogens are very difficult to remove from contaminated fruit and vegetables by washing (9), and these pathogens may survive throughout the shelf life of the product.
This study is the first to reveal the attachment ability of L. monocytogenes, Shigella, Salmonella Typhimurium, Salmonella Typhi, Salmonella Montevideo, S. aureus, E. coli O157:H7, EAEC, and V. cholerae O1 to jalapeño and serrano peppers.
Disinfection study. Fruits and vegetables for consumption typically are washed with water that contains free chlorine from approximately 0 to 200 ppm. Chlorine and chlorinated compounds have been used for several decades, and these compounds are still the most widely used sanitizers in the food industry (4, 6, 7, 37, 51). However, chlorine and chlorinated compounds may have a limited effect on the reduction of microorganisms on the surfaces of vegetables (13). Many researchers have stated that excessive use of chlorine can be harmful owing to the formation of carcinogenic disinfection by-products such as trihalomethanes, chloramines, haloketones, chloropicrins, and haloacetic acids caused by the reaction of residual chlorine with organic matter (2, 10, 21, 27, 34, 48, 60). Because of the risks posed by the use of chlorine in the food industry, disinfection with these compounds is forbidden in European countries such as The Netherlands, Sweden, Germany, and Belgium (40, 48, 50). In response, chlorine-based compounds are being eliminated from the disinfection and decontamination processes and alternatives, such as antimicrobial agents from plants, are being explored. Many studies are now in progress searching for alternative antimicrobial agents, such as that derived from the roselle calyx.
The effect of the roselle calyx extract on the levels of foodborne bacteria inoculated onto the surface of jalapeño and serrano peppers is shown in Tables 2 and 3. In general, treatments with disinfectant solutions caused a similar reduction in the level of pathogenic bacteria on both types of peppers. Overall, colloidal silver had no significant effect (P > 0.05) for reducing the levels of pathogenic bacteria on either pepper type compared with the control treatment (ISS). Colloidal silver solution is widely used to disinfect fruits and vegetables in restaurants and homes in Mexico, and the Health Secretariat of Mexico promotes use of this solution as a disinfectant for fruits and vegetables. However, little information has been published in scientific journals on the use of colloidal silver as a disinfectant of fruits and vegetables. In Mexico, Microdyn (colloidal silver in gelatin) is sold in supermarkets to disinfect salad vegetables and drinking water. However, in the present study colloidal silver did not significantly reduce the levels of pathogens on both types of peppers. García-Gómez et al. (26) reported limited reduction in levels of Salmonella when they disinfected lettuce and coriander with Microdyn. Soto Beltran et al. (57) also reported a limited effect of Microdyn for reducing levels of Salmonella Typhimurium on bell peppers.
In contrast, treatments with sodium hypochlorite and acetic acid each resulted a significant reduction of pathogenic bacteria (P < 0.05) in both types of peppers compared with the control and with the colloidal silver treatment (Tables 2 and 3). However, no significant difference (P > 0.05) was found between sodium hypochlorite and acetic acid treatments in the reduction of pathogenic bacteria on both types of peppers (Tables 2 and 3). In general, <2-log reductions were obtained for pathogenic bacteria on both types of peppers.
Treatments with calyx extracts in water, ethyl acetate, methanol, and acetone caused a greater (P < 0.05) reduction in the levels of all pathogenic bacteria for both types of peppers than did treatments with colloidal silver, sodium hypochlorite, and acetic acid (Tables 2 and 3). The four roselle calyx extracts resulted in 2.2- to 3.0-log reductions in the pathogenic bacteria for both types of peppers. In general, the methanol and acetone extracts produced a greater reduction (P < 0.05) in the Salmonella cocktail than did the ethyl acetate and water extracts. V. cholerae O1 was the most sensitive microorganism to all treatments in both types of peppers, followed by the E. coli pathotypes EPEC, ETEC, EIEC, and EAEC. These results suggest that roselle calyx extracts may be a useful addition to disinfection procedures for raw jalapeño and serrano peppers in the field and for peppers in processing plants, restaurants, and homes.
The antimicrobial activity of compounds in the roselle calyces constitute a potential alternative for controlling foodborne bacteria on food. This activity has been attributed to compounds such as protocatechuic acid and anthocyanins (43), but no published research has yet identified the specific compounds responsible for this activity. Further research will be needed to identify the active compounds and their effective concentrations.
In previous studies, roselle calyx extracts have been used to disinfect carrots (31), tomatoes (32), romaine lettuce and alfalfa sprouts (41), and organic leafy vegetables (romaine lettuce, iceberg lettuce, and baby spinach) (44). Jaroni and Ravishankar (41) reported a reduction of Salmonella Newport on alfalfa sprouts and romaine lettuce of approximately 1.0 log CFU immediately upon exposure to aqueous roselle extracts. Moore et al. (44) examined leafy greens (romaine lettuce, iceberg lettuce, and baby spinach) contaminated with Salmonella Newport and treated by immersion in roselle extract concentrates (at 10, 20, and 30%) for 2 min. An immediate 1-log reduction of Salmonella Newport populations was noted for all three greens upon exposure to the 30% roselle extract, but no reductions were observed with the 10 and 20% concentrations (44). Some antimicrobial activity of roselle calyx water extracts has been found in other foods such as apple juice (20), milk with various fat concentrations (35), ground beef (20), and hot dogs (36).
To the best of our knowledge, this is the first report of reductions in foodborne bacteria in response to application of roselle calyx extracts on both jalapeño and serrano peppers. Roselle calyx extracts are clearly a promising alternative for reducing or eliminating foodborne bacteria on raw fruits and vegetables. However, truly effective prevention and control of health risks from foodborne bacteria require incorporation and consistent application of good agricultural and manufacturing practices throughout the pepper production process, from crop to harvest to retailer. Proper jalapeño and serrano pepper handling and processing practices must be promoted and implemented among pepper growers and consumers, and these practices, such as thorough washing and disinfection, need to be promoted in homes and restaurants. Roselle calyx extracts may be a useful addition to disinfection procedures in the field, processing plants, restaurants, and homes.
This work was supported by the Fondos Mixtos de Fomento a la Investigación Científica y Tecnológica, Consejo Nacional de Ciencia y Tecnología–Gobierno del Estado de Hidalgo, Mexico, grant 192649.