Direct detection of toxigenic Bacillus cereus in dietary complement for children and cassava starch Detección directa de Bacillus cereus toxigénicos en complementos dietarios para niños y en almidón de yuca Detecção direta de Bacillus cereus toxigênicos em suplementos alimentares para crianças e ami

Bacillus cereus is a food contaminant and a known human pathogen that can cause emetic and diarrheal syndromes. In this study we evaluated the presence of toxigenic B. cereus by multiplex PCR directly in dietary complement for children and cassava starch samples collected on Medellin, Colombia. Of 75 dietary complement for children samples evaluated, 70.7% were contaminated with toxigenic B. cereus and four different toxigenic consortia were detected: I: nheA, hblC, cytK (9.8%), II: nheA, hblC (2%), III: hblC, cytK (41.2%), IV: hblC (47%). Of 75 cassava starch samples, 44% were contaminated with toxigenic B. cereus and four different toxigenic consortia were determined: I: nheA, hblC, cytK (48.5%), II: nheA, hblC, cytK, cesB (3%), III: hblC, cytK (30.3%), IV: hblC (18.2%). In general, in dietary complement for children only enterotoxigenic consortia were detected while in cassava starch the enterotoxigenic consortia predominated over the emetic. Multiplex PCR was useful to detect toxigenic B. cereus contamination allowing direct and simultaneous detection of all toxin genes in foods. This study is the first in Colombia to evaluate toxigenic B. cereus, providing information of importance for microbiological risk evaluation in dried foods. Jennifer A. Sánchez1, Margarita M. Correa1, Angel E. Aceves-Diez2, Laura M. Castañeda-Sandoval3, * 1Grupo Microbiología Molecular, Escuela de Microbiología, Universidad de Antioquia UdeA, Calle 70 No. 52-21, Medellín, Colombia. 2Departamento de Investigación y Desarrollo, Laboratorios Minkab S.A. de C.V., Av. 18 de Marzo No. 546, Guadalajara, Jalisco, México. 3Escuela de Microbiología, Universidad de Antioquia UdeA, Calle 70 No. 52-21, Medellín, Colombia.Teléfono: (574) 219 5492. *Autor para correspondencia: laura.castaneda@udea.edu.co. Recibido: 19 de junio de 2014. Aceptado: 20 de julio de 2014.


Introduction
Bacillus cereus is a spore-forming, aerobic to facultative, Gram-positive and motile rod bacteria that can be commonly found in many types of environments and is also a known human pathogen that can cause emetic and diarrheal syndromes (1).Emetic syndrome occurs after ingestion of food containing a preformed toxin called cereulide or emetic toxin (2), which is encoded by the ces gen cluster, that comprises seven coding DNA sequences, cesH, cesP, cesT, cesA, cesB, cesC and cesD, located on a plasmid (3).
Three different enterotoxins responsible for the diarrheal syndrome have been described: hemolysin BL (HBL), encoded by the hbl operon that comprises the hblC, hblD and hblA genes; nonhemolytic enterotoxin (NHE), encoded by the nhe operon composed by the nheA, nheB and nheC genes; and cytotoxin K (CytK) encoded by the cytK gene.These genes are located on the bacterial chromosome (4).
Different foods such as rice, meat, pasta, chicken, fruits, grain, spices, and vegetables may be contaminated by cell or spores of B. cereus (5).Moreover dried foods such as powder milk and starch foods are frequently contaminated by B. cereus; this is because of the presence of starch degrading amylases and spores that can survive to treatments such as drying and heat, factors that eliminate other competing microorganisms present.Spores can germinate when in contact with water during food preparation, leading to spoilage or food poisoning (6).
The detection of B. cereus is traditionally performed by plating and biochemical assays that are time-consuming and do not allow detecting the toxigenic potential and diversity of the strains (2).Molecular approaches currently available, for example multiplex PCR, are inexpensive, easy to perform, and allow the evaluation of the toxigenic potential of the strains (7).In addition, some tests do not require isolation of the microorganism in pure culture (8).
In Colombia, the toxigenic potential of B. cereus in foods is unknown, and a rapid detection test for this pathogen in foods is not available, which impede the application of rapid quality control measures to eliminate B. cereus from food.Therefore, the objective of this study was to evaluate the presence of toxigenic B. cereus by a simple and rapid test, a multiplex PCR, directly in dietary complement for children and cassava starch samples collected in Medellin, Colombia.This study is the first in Colombia to evaluate toxigenic B. cereus and can provide an approximation of the type of toxins that strains can cause in food.Also, this study helps to better understand the toxigenic B. cereus in dried foods and provides information for microbiological risk evaluation.

Dried food samples
The following dried foods products were selected for analysis: cassava starch (n=75) and dietary complement for children (n=75), which were collected in public and private educational institutions, bakeries and powdered food companies located in Medellin, Colombia.

Extraction of spores and cells of B. cereus from dried foods
Twenty-five grams of the dried food were dissolved in 225 ml sterile distilled water and filtered through Whatman N°1 filter.The resulting liquid portion containing B. cereus spores and cells was centrifuged at 6000 g for 30 min and the pellet used for DNA extraction.

DNA extraction
Total DNA from B. cereus spores and cells was extracted according to the method described by D' Alessandro (9).

Multiplex PCR
To develop the multiplex PCR assays to test for toxigenic B. cereus in dried foods, the selected primer pairs were directed to amplify hblC, nheA, cesB and cytK genes.In addition, 16S rDNA sequence was targeted as the amplification internal control (Table 1).
The final reaction mixture (16 μL) consisted of 0.6 mM dNTPs mix, 4 mM MgCl 2 , 0.2 μM forward and reverse primers for hblC, nheA, cesB and cytK genes and 0.1 μM for ITS1, 1.3 U of Taq platinum polymerase (Invitrogen, Germany), and 1.6 μL 10X reaction buffer.Amplification was performed on a G-Storm GS482 thermocycler with an initial denaturation step at 94 °C for 5 min, followed by 35 cycles of 1 min denaturation at 94 °C, 40 s annealing at 50 °C and 2 min elongation at 72 °C, and final incubation at 72 °C for 10 min.After electrophoresis on a 2% agarose gel, PCR products were stained with ethidium bromide and visualized under ultraviolet light (UVP Gel Doc).

Specificity and sensitivity determination
To

Specificity and sensitivity of the multiplex PCR
Cloned PCR products from B. cereus reference strains corresponding to hblC, nheA, cesB, cytK, and the internal control 16S rDNA were confirmed by sequencing.High identity values were obtained (95-100%) when these sequences were compared with those registered in GenBank, indicating that the amplified products were those expected.The multiplex PCR only amplified toxigenic genes in the B. cereus reference strains and no in other bacterial genera evaluated demonstrating the specificity of the PCR.The DNA detection limit of the multiplex PCR was 100 ng; at this concentration strong bands were visualized for all the evaluated genes (Figure 1).

Direct detection of toxigenic B. cereus from dried foods
Fifty three (70.7%) of 75 dietary complement for children samples evaluated by multiplex PCR were contaminated with toxigenic B. cereus.The most predominant toxin gene was hblC (65.8%), followed by cytK (34.2%), and nheA (7.9%).The emetic gene cesB was not detected.The standardized multiplex PCR allowed the detection of four toxigenic consortia, differing in the pattern of toxic genes present in the samples (Table 2).
Thirty three (44%) of 75 cassava starch samples were contaminated with toxigenic B. cereus.The most predominant toxin gene was hblC (44%), followed by cytK (36%) and nheA (22.7%).Unlike dietary complement for children samples, the emetic gene cesB was detected in 1.3% of cassava starch samples.According to the pattern of toxic genes determined in the samples, four different toxigenic consortia were established (Table 3).
The results of toxin gene consortia present in dietary complement for children and cassava starch samples are in agreement with those previously reported for toxigenic B. cereus detected directly by multiplex PCR in other foods (8,15).In India, a study in meat and meat products evaluated the presence of hblA, hblC, hblD, nheA, nheB, nheC, cytK, and  entFM genes.Six toxigenic consortia were detected: Group I (50.84%), all eight genes were detected; Group II (10.16%) and Group III (8.47%), the foods lacked hbl complex (hblCDA) and cytK, respectively; Group IV (16.94%), foods not containing the hbl complex and cytK; Group V (6.78%), foods in which none, one, two or all the three genes of the hbl complex were present but cytK, entFM and at least one gene of nhe complex (nheABC) were present; Group VI, similar to Group V but lacking cytK gene (15).In a study conducted in Kenya, the presence of hblA, hblC, hblD, nheA, nheB, nheC, and emetic toxin genes were evaluated by multiplex PCR directly in ready-to-eat foods such as rice and milk.Emetic toxin genes and the hblC gene were detected in rice while in milk nheA, hblD, and hblC genes were identified (8).
Only the toxin gene consortium II of cassava starch presented the cesB gene, which is in agreement with previous studies that indicated that the detection of the emetic toxin was rare (16).Its production seems to be restricted to a particular lineage of B. cereus (4).
Few studies have been carried out using multiplex PCR for direct detection of B. cereus in foods, and usually the toxigenic potential of B. cereus strains is determined by strain isolation from foods.In a study in Korea, B. cereus isolates from cereal presented a high frequency of nheA (99%) and hblDC (84%) genes, but cytK gene was less frequent (55%) (17).In Belgium, various strains of the B. cereus group were isolated from marketed food products, none harbored the ces gene required for the production of the emetic toxin, but 52.5% strains carried all seven genes required for the production of the diarrhoeal enterotoxins: haemolytic BL, non-haemolytic enterotoxin and cytotoxin K (18).
In United States 47 (56.6%)B. cereus isolates from rice contained the hblA and hblD genes and 74 (89.1%) isolates the nheA and nheB, but the ces gene was not detected in any of the isolates (19).In Argentina, of 132 B. cereus isolates from honey, 42% harbored the hblABCD genes, 53% the cytK gene and 73% the bceT (20), the latter gene encodes the BceT enterotoxin and its biological activity is not clear yet, therefore, it was not evaluated in the present study (2).In Brazil, 97 foodborne B. cereus sensu stricto strains isolated in the 1980's, 1990's and 2000's were analyzed.The nhe genes were detected in 84.5% strains and hbl genes in 62.9% strains; all strains were negative for ces and the cytK-2 gene was found in 45.4% strains.The predominant toxigenic pattern included enterotoxin genes positive strains but no ces (21).
In countries such as Korea, emetic strains have been isolated from grain, korean rice cake (22) and sunsik (23).It is known that emetic strains have been reported in oriental countries where the emetic syndrome has been associated with the consumption of rice (2).In a study in Argentina only one emetic strain in cooked chicken was reported (24) and this was the first report of emetic strains in the Americas.The finding in our study of a sample containing the cesB gene (consortia II) in a cassava starch sample suggest that frequencies of ces containing strains may be low in Medellin, Colombia, and possibly this is the case for other Latinamerican countries.However, more studies are needed to prove this hypothesis.

Conclusions
The predominant consortium found in dietary complement for children included hblC and cytK genes and the predominant consortium found in cassava starch included nheA, hblC and cytK genes.In general, in dietary complement for children only enterotoxigenic consortia were detected while in cassava starch the enterotoxigenic consortia predominated over the emetic.
The standardized multiplex PCR can be used to test contamination of dried foods by toxigenic B. cereus by simultaneous detection of all toxin genes directly in foods such as dietary complement for children and cassava starch.This assay will help to optimize the time and resources in the laboratory and serves the bases for carrying out survey or studies directed to better understand the epidemiological risk represented by toxigenic B. cereus.
assess the specificity of the multiplex PCR developed in this study, a panel of B. cereus reference strains were included: B. cereus F4810/72 (ces, nhe), B. cereus 1257 (ces, nhe), B. cereus ATCC 10987 (hbl, nhe, cytK), and B. cereus ATCC 14579 (hbl, nhe, cytK).Also, various bacterial strains considered important in food safety available in the reference lab collection, such as Salmonella spp., Shigella spp., Escherichia coli, Staphylococcus aureus and Listeria monocytogenes were evaluated.To assess the minimum amount of B. cereus DNA detectable by the multiplex PCR, 0.1, 0.5, 1, 10, 20, 50, 100, 200 or 500 ng of DNA template from B. cereus reference strains was added separately to tubes that contained multiplex PCR mix.

Table 1 .
Description of primers used for multiplex PCR detection of B. cereus genes.

Table 2 .
Toxigenic consortia in dietary complement for children samples.