The Bacillus cereus group, or B. cereus sensu lato (B. cereus s.l.), are Gram-positive bacteria with a low GC content that belong to the Firmicutes phylum. It is a group that includes bacillus-shaped, spore-forming, aerobic to facultative anaerobic bacteria, with peritrichous flagella involved in locomotion, including at least eight highly related species: B. anthracis, B. cereus, B. thuringiensis, B. mycoides, B. pseudomycoides, B. weihenstephanensis, B. cytotoxicus and B. toyonensis (Enosi Tuipulotu et al., 2021; Liu et al., 2015).

The species of the B. cereus group are endemic soil bacteria that occupy diverse ecological habitats due to the formation of endospores resistant to heat, UV radiation, acids, and desiccation, for which the bacteria can persist in a dormant state (Ehling-Schulz et al., 2019). In addition to spore production, B. cereus resists adverse environmental conditions due to the formation of biofilms on abiotic surfaces and living tissues (Duport et al., 2016). In this sense, it has been reported that B. cereus can produce biofilms in pipes and food storage systems, favoring the systematic contamination of food products (Majed et al., 2016; Shemesh and Ostrov, 2020; Wijman et al., 2007).

In addition to soil, species have been isolated from fresh and frozen foods, invertebrates, and plants (Ehling-Schulz et al., 2019). In this sense, it is estimated that B. cereus is responsible for 1.4%-12% of all food poisonings worldwide (Grutsch et al., 2018). B. cereus is responsible for two types of gastrointestinal syndromes. The emetic type is mainly characterized by nausea and emesis, which appears one hour after consumption of contaminated food and is clinically indistinguishable from Staphylococcus aureus enterotoxin poisoning (Stenfors Arnesen et al., 2008). The diarrheal type of food poisoning is also associated with various foods. The disease mainly manifests in diarrhea and abdominal cramps, like Clostridium perfringes food poisoning, and occurs after approximately 8 to 16 h after consumption (Jessberger et al., 2020; Stenfors Arnesen et al., 2008).

The diarrhoeal syndrome is associated with different enterotoxins produced after germination of spores and growth of vegetative cells, ingested through contaminated food and activated in the small intestine (Schoeni and Wong, 2005, Stenfors Arnesen et al., 2008). B. cereus produces three pore-forming enterotoxins: Nhe (a three-component non-haemolytic enterotoxin), Hbl (a three-component haemolytic enterotoxin) and, the single-component cytotoxin CytK. The emetic syndrome results from poisoning caused by a toxin called cereulide, which is found in food (Schoeni and Wong, 2005). Cereulide is a dodecapeptide composed of alpha amino and alpha hydroxy acids, structurally related to the potassium ionophore valinomycin. Cereulide is produced by a peptide synthetase called ces, which represents a new type of non-ribosomal peptide synthetase (NRPS) (Ehling-Schulz et al., 2004).

Epidemiological data show that rice, pasta, cakes, and noodles are associated with emetic syndrome, whereas vegetables, meat, and dairy products have been associated with diarrheal syndrome (Enosi Tuipulotu et al., 2021). In Mexico, there is currently no health legislation related to the presence of B. cereus in foods, but the presence of B. cereus in vegetables (Castulo-Arcos et al., 2022; Flores-Urbán et al., 2014), ice cream (Adame-Gomez et al., 2019), artisanal cheeses and eggs (Adame-Gómez et al., 2020b; Cruz-Facundo et al., 2022; 2023) has been reported. Rice is a product consumed in Mexico in different preparations, including fried rice and milk dessert with rice. In a preliminary study, B. cereus was isolated from 5% of rice samples, with a higher frequency in rice desserts (8.5%). However, not only the isolation of the microorganism is essential, but also the molecular characterization of different virulence factors, are needed. Therefore, the objective of this study is to identify the toxigenic profile, lytic enzymes, antimicrobial resistance, and biofilm production of B. cereus strains isolated from rice.

In Mexico, the annual per capita consumption of rice is 8.5 kg, considered the second crop with the highest expenditure among mexican families (SAGARPA, 2017). However, B. cereus, a pathogen associated with foodborne diseases worldwide, has been isolated in rice (Dietrich et al., 2021; Enosi Tuipulotu et al., 2021). Rice and its derivatives serve as an ideal growth medium for bacteria, particularly B. cereus, due to their compositional and chemical characteristics (pH close to 7, approximately 79% carbohydrates, 7% protein, and 2% fat). These properties, combined with the resilience of B. cereus spores, which can survive in dehydrated grains and withstand the temperatures typically applied during processing, enable the bacterium to proliferate under a variety of conditions (Jaquette and Beuchat, 1998). Therefore, in this study, strains of B. cereus s.l. were characterized and isolated from different rice-based preparations.

In the different studies of B. cereus in rice, strains with the same toxigenic profiles have not been isolated, and even the frequencies change depending on the type of rice and country (Ankolekar et al., 2009; Kim et al., 2009; 2014; Park et al., 2009), which is why studying the diversity of B. cereus strains in Mexico is essential. A total of six strains of the B. cereus s.l. group previously isolated from rice were characterized. All these strains were positive for the non-hemolytic toxin gene (nhe) and negative for the cereulide toxin gene (ces). Only one strain (B629) was positive for the hemolytic toxin BL gene (hbl) and negative for the cytotoxin K gene (cytK) (Table 1) (Figure 1).

Table 1

Strain	hbl	nhe	cytK	ces
B629	+	+	-	-
B630	-	+	+	-
B631	-	+	+	-
B632	-	+	+	-
B633	-	+	+	-
B634	-	+	+	-

Figure 1

In this study, a remarkable occurrence of the non-hemolytic toxin gene (nheABC) with a 100% frequency was identified in the analyzed strains. Globally, non-hemolytic toxin genes are frequently reported, often reaching values of up to 100%. In Mexico, several food products, including cheese, ice cream, eggs, and vegetables, have shown a high prevalence of non-hemolytic toxin genes (Adame-Gomez et al., 2019; 2020b; Castulo-Arcos et al., 2022; Cruz-Facundo et al., 2022; 2023); it is important to note that these findings do not exclude the possibility that these strains are circulating in rice, in which, the frequency of non-hemolytic toxin genes show a range from 47.3% to 100%. The variability in these results is suggested to be related to the specific type of rice analyzed. For example, in ready-to-eat rice-based foods, a frequency of 47.3% was observed (Chen et al., 2022) while frequencies approaching 100% were identified in different types of rice, including white rice, brown rice, black rice, hard rice, and glutinous rice (Ankolekar et al., 2009; Kim et al., 2009; 2014; Park et al., 2009).

The frequencies of Hbl toxin genes were low (16.6%) compared to the global reported frequencies for these toxin genes (40-70%) (Jessberger et al., 2020). As for Mexico, the frequencies are low (Adame-Gomez et al., 2019; 2020b; Castulo-Arcos et al., 2022; Cruz-Facundo et al., 2022; 2023) as in this study, which does not rule out a regional circulation of B. cereus strains in different food products. Regarding rice, the reported frequencies of Hbl toxin genes are high compared to this study. The lowest frequencies for the Hbl toxin are found in strains isolated from rice-based ready-to-eat foods (36.3 %), and the highest frequencies are found in white and brown rice (83-100%) in China and Korea. In the United States, even when it was isolated from white and brown rice, the frequency of the Hbl toxin was low (56.6%), so it is not surprising that not only the type of product, but also the geographic region, is important in the frequency of enterotoxins. In this sense, the regionalization of diarrheal and emetic syndromes with the dietary habits of each country has even been described (Enosi Tuipulotu et al., 2021; Kotiranta et al., 2000).

Another virulence factor, in addition to spore formation and toxin production by B. cereus, is the ability to form biofilms. Biofilms are bacterial communities that, in the case of B. cereus, have been described in food production areas, becoming a problem because, on one hand, they favor systematic contamination of food. On the other hand, the formation of biofilms prevents the eradication of the microorganisms trapped in it due to the characteristics of resistance to disinfectant agents (Majed et al., 2016).

Regarding biofilm formation, strains B631, B632, B633, and B634 exhibit substantial biofilm production on glass material in contrast to strain B629 (p = 0.0211, p = 0.0445, p = 0.0211, p = 0.0216). Notably, three strains (B632, B633, B634) show the ability to form biofilms on polystyrene. It is noteworthy that none of the strains form biofilms on polyethylene as shown in Figure 2.

Figure 2

In this investigation, prolific biofilm production was observed exclusively on glass material, in contrast to minimal production on polyethylene and, to a lesser extent, polystyrene. The variation in biofilm production on different materials is consistent with previous literature which attribute such differences to the hydrophobic properties and surface roughness of the materials(Adame-Gómez et al., 2020a; De-la-Pinta et al., 2019). In addition, the presence of trace elements, particularly iron, has been implicated in biofilm production. Hayrapetyan et al. (2015) reported increased biofilm production on stainless steel, attributed to iron ions, compared to polystyrene. Within our research group, previous studies have indicated higher biofilm production on PVC compared to other materials (Adame-Gómez et al., 2020a; Cruz-Facundo et al., 2022). It’s worth noting that discrepancies between our results and those of this study may be related to the different origins of the strains. Hayrapetyan et al. (2016) relate the origin of the isolation to the adaptation capacity of the strains to environments with different iron conditions.

Regarding the genes related to biofilm production, only strain B629 has the eps2 operon. All strains have the sipW-tasA-calY operon genes (Figure 3). In B. cereus, the participation of different genes in the production of biofilms has been described; these genes are grouped in the sipW-tasA-calY and eps2 operons (Caro-Astorga et al., 2014; 2020). SipW is a peptidase that participates in the maturation of the tasA protein, which functions in the formation of amyloid-type fibers, which are part of a mature biofilm (Caro-Astorga et al., 2014). CalY fulfills a dual function during biofilm production, participating as an adhesin on glass but not polystyrene in the early stages of biofilm formation and forming fibers during the late phases (Candela et al., 2019). Eps2 is an operon that include genes related to the extracellular matrix, the main component of biofilms (Caro-Astorga et al., 2020). The diverse functions these proteins fulfill explain why most of the strains in this study produce abundant biofilms on glass, except for strain B629, which, despite having all the genes related to biofilms, visually produces a biofilm that easily detaches from the air-liquid interface and therefore is not evident by safranin staining. Dogsa et al. (2013) showed that the width of the biofilm is related to the culture medium and to the presence of the tasA and epsA genes. The absence of one of these genes drastically reduces the thickness of the biofilm, and in the absence of both, the strain does not produce a biofilm. For B629, the production of a thin biofilm could be explained by presence of a non-functional tasA gene, which is reflected in a thin biofilm that easily detaches from the tube. Despite the detailed information about the participation of the sipW-tasA-calY and eps2 operons in biofilm production (Caro-Astorga et al., 2020; 2014), few epidemiological studies include them in molecular characterization (Cruz-Facundo et al., 2023). Since biofilm production is part of the resistance mechanisms and systematic contamination of food products (Enosi Tuipulotu et al., 2021), not only enterotoxigenic and enzymatic profiles should be considered, but also those genetic profiles associated with biofilm production, which are included in this study.

Figure 3

B. cereus not only produces enterotoxins, but also a significant number of extracellular enzymes with degradative activity, such as phospholipases, proteases, chitinases, and amylases (Ivanova et al., 2003). This enzymatic profile is related to the ability of B. cereus to spoilage food and as a nutrient assimilation mechanism (Arslan et al., 2014). In this sense, the B629 strain is the only amylolytic strain in the study, and coincides with being the only strain isolated from fried rice. Strains isolated from sweet rice are capable of producing proteases, which can have a significant impact on the decomposition of the product, since it has been described that proteases are capable of gelatinizing milk (Chen et al., 2003). All strains produce proteases and strains in different proportions produce lecithinase. Only one strain can grow at refrigeration temperatures (B629) (Table 2). In this sense, strains of B. cereus are divided into two groups: psychotrophic and mesophilic. In this study, it is reported that strain B629 is a psychotrophic strain capable of growing in 5 days at a temperature of 10°C, and that the other strains, since they cannot grow at refrigeration temperatures, could be considered mesophilic strains. The presence of psychrotrophic strains, such as B629, impact the food safety of food products that are preserved in the cold, including rice.

Table 2

Strain	Amylase	Protease	Lecithinase	Growth in low temperatures (DT)
B629	+	+	0.71	5
B630	-	+	0.52	-
B631	-	+	0.60	-
B632	-	+	0.50	-
B633	-	+	0.50	-
B634	-	+	0.60	-

Detection time (DT): the day on which growth was observed Tiempo de detección (DT): El día en donde se observa crecimiento

Genetic diversity techniques, such as GTG’5, have allowed strains to be grouped according to different characteristics, including growth temperature. Regarding the genetic diversity of the B. cereus s.l. group, strains were grouped into three clusters. In the first cluster, two clones were identified; the first clone includes strain B629 isolated from fried rice, the only strain that grows at refrigeration temperatures and is amylase positive, and is also the only one positive for the toxin BL gene (hbl); the second clone includes strains B632 and B633, which share most of the characteristics with the clones belonging to the other clusters (Figure 4).

Figure 4

The GTG’5 technique allowed the identification of two clones, one with tolerance to refrigeration temperatures, amylolytic, not a biofilm producer, and isolated from fried rice. Furthermore, the technique allowed us to separate this clone from three remaining clones isolated from sweet rice with similar genotypic characteristics, including the profile of toxins and genes associated with biofilm production. This is evidence that the strains circulating in fried rice are different from the strains found in sweet rice, which allows us to estimate that the risk of poisoning also depends on the product consumed. For example, emetic strains have been linked to the consumption of products such as pasta and rice, while the consumption of meat and vegetables has been linked to enterotoxigenic strains (Enosi Tuipulotu et al., 2021).

Regarding antimicrobial resistance, it was found that all strains are resistant to penicillin, ceftriaxone, clindamycin, and trimethoprim. The strains were susceptible to kanamycin, gentamicin, vancomycin, tetracycline, chloramphenicol, and ciprofloxacin. It is important to note that only B629 has intermediate sensitivity to erythromycin (Table 3).

Table 3

Antibiotic	Sensible (≤ 2µg/ mL)	Intermediate (>2- <16 µg/ mL)	Resistant (≥ 16 µg/ mL)
Penicillin			6 (100%)
Ceftriaxone			6 (100%)
Erythromycin	5 (83.4%)	1 (16.6%)
Clindamycin			6 (100%)
Ciprofloxacin	6 (100%)
Tetracycline	6 (100%)
Chloramphenicol	6 (100%)
Vancomycin	6 (100%)
Gentamycin	6 (100%)
Kanamycin	6 (100%)
Trimethoprim			6 (100%)

Regarding antibiotic resistance, strains of B. cereus s.l. from this study were resistant to beta-lactams, which has been reported in other studies (Chen et al., 2022; Fraccalvieri et al., 2022; Park et al., 2009; Perera and Ranasinghe, 2012). Fraccalvieri et al. (2022) demonstrated that the main mechanism of resistance to beta-lactams is mediated by the beta-lactamases BLA-1 and BLA-2, since all their beta-lactam resistant strains presented these genes. In addition to resistance to beta-lactams, high resistance to trimethoprim has also been reported in other studies in strains of B. cereus s.l. isolated from rice and cheese (Chen et al., 2022; Cruz-Facundo et al., 2023). It is important to mention the reports related to the high susceptibility to different groups of antimicrobials such as aminoglycosides, glycopeptides, tetracyclines, and phenicol’s (Chen et al., 2022; Cruz-Facundo et al., 2023; Park et al., 2009) which is also consistent with what was reported in these strains. However, the presence of a strain with intermediate sensitivity to erythromycin in this study (B629) should not be underestimated.

The Bacillus cereus sensu lato (B. cereus s.l.) strains isolated from rice exhibit enterotoxigenic profiles, indicating their potential to cause food poisoning. Of particular note is the finding that strains isolated from sweet rice have a propensity to form biofilms on glass, and exhibit a distinct enterotoxigenic profile compared to the strain isolated from fried rice. The unique characteristics identified in the strain isolated from fried rice underscore the importance of further investigation and warrant an expansion of the sample size, specifically within this rice type. This expansion aims to systematically explore the presence of strains with similar metabolic and virulence characteristics.