Colonial morphology of the studied strains.
The characterization the E. coli strains analyzed included the colonial morphology on agar, in this were observed colonies between 2–4 mm in diameter of white color, shiny surface, opaque density, and convex elevation. In the blood agar some strains showed hemolysis around the colonies. In MacConkey agar all the strains except negative control (HB101) were lactose positive and the biochemical test (Koneman 1983) showed a IMViC (+,+,-,-) for all strains.
In vitro biofilm formation.
The biofilm formation on polystyrene plates was positive for all E. coli strains except for the negative control (HB101) the three incubation times. Except for the strain E66438 which was a weak biofilm former, the rest of DAEC strains were classified as strong biofilm formers. The extraintestinal UPEC strain CFT073 was classified as moderate biofilm former ant the environmental (bovine-derived) strains were weak biofilm formers (Table 2).
Table 2
Formation of biofilm in vitro by strains of E. coli at three different incubation times in two culture media
Group | Pathotype | Strain code | MEM (߃± SD) | LBB (߃± SD) | CF |
24h | 48h | 72h | 24h | 48h | 72h | |
Diarrheagenic | DAEC | E66438 | 0.215 ± 0.003a | 0.208 ± 0.002a | 0.264 ± 0.049 a | 0.233 ± 0.011 a | 0.235 ± 0.010 a | 0.317 ± 0.030 a | WBF |
| EAEC1 | 49766 | 5.49 ± 0.378c | 6.055 ± 0.680 c | 6.36 ± 0.177 c | 5.524 ± 0.423 c | 5.949 ± 0.554 c | 6.367 ± 0.179 c | SBF |
| EHEC | 9330 | 3.223 ± 0.763 b | 4.208 ± 0.071 b | 4.363 ± 0.173 b | 3.385 ± 0.571 b | 4.211 ± 0.087 b | 4.388 ± 0.198 b | SBF |
| EHEC | DL933 | 2.42 ± 0.019 ab | 2.85 ± 0.101 ab | 3.60 ± 0.398 ab | 0.257 ± 0.006 ab | 2.963 ± 0.226 ab | 3.72 ± 0.277 ab | SBF |
Extraintestinal | UPEC | CFT073 | 1.219 ± 0.138 a | 1.850 ± 0.087 a | 1.906 ± 0.007 a | 1.209 ± 0.099 a | 1.880 ± 0.044 a | 1.719 ± 0.395 a | MBF |
Environmental2 | | SPEC DC+ | 0.140 ± 0.049 a | 0.166 ± 0.015 a | 0.185 ± 0.013 a | 0.150 ± 0.046 a | 0.169 ± 0.017 a | 0.197 ± 0.026 a | WBF |
| | 3FCL + 2 | 0.176 ± 0.017 a | 0.231 ± 0.103 a | 0.300 ± 0.029 a | 0.191 ± 0.022 a | 0.216 ± 0.080 a | 0.332 ± 0.033 a | WBF |
| | IFL + 6 | 0.445 ± 0.048 a | 0.680 ± 0.046 a | 0.720 ± 0.045 a | 0.452 ± 0.052 a | 0.645 ± 0.055 a | 0.792 ± 0.019 a | WBF |
| | LPL+. | 0.265 ± 0.040 a | 0.317 ± 0.029 a | 0.328 ± 0.028 a | 0.345 ± 0.036 a | 0.375 ± 0.030 a | 0.390 ± 0.007 a | WBF |
| | 3PL + C | 0.294 ± 0.052 a | 0.351 ± 0.036 a | 0.363 ± 0.031 a | 0.299 ± 0.053 a | 0.371 ± 0.020 a | 0.395 ± 0.002 a | WBF |
| | 6PBL+. | 0.496 ± 0.104 a | 0.547 ± 0.038 a | 0.613 ± 0.015 a | 0.506 ± 0.016 a | 0.560 ± 0.035 a | 0.612 ± 0.051 a | WBF |
Non-pathogenic3 | | HB101 | 0.169 ± 0.043 a | 0.186 ± 0.031 a | 0.282 ± 0.025 a | 0.170 ± 0.039 a | 0.196 ± 0.031 a | 0.289 ± 0.023 a | NBF |
1Strain used as positive control. 2E. coli isolated from the surfaces of bovine carcasses. 3Non-pathogenic strain of E. coli generated in the laboratory from E. coli K12. MEM: Minimum Essential Medium, LBB: Luria Bertani Broth. CF: Biofilm forming capacity, NBF: Non biofilm former, WBF: Weak Biofilm Former, MBF: Moderate Biofilm Former, SBF: Strong Biofilm Former. Different letters represent statistically significant differences.
In this study no statistically significant differences in biofilm formation were observed between the two-growth media evaluated (glucose-enriched MEM and LBB). Studies by Cáceres et al (2018) reported an increase in the formation of biofilms by E. coli and other enterobacteria in cell culture medium glucose enriched, they proposed that glucose is useful as a substrate for the exopolysaccharide matrix formation, and therefore increases the total biomass. The potential effect of glucose on biofilm development was confirmed in two of the EHEC O157:H7 strains (Table 2). On the other hand, Faleiro (2010) reported a stronger biofilm formation in minimal media as the LBB used in this study. Some authors mention that bacterial adherence and the biofilms formation are stimulated under conditions of scarce nutrients in the medium (Reisner et al. 2006; Skyberg et al. 2006; Yang et al. 2004). Similarly, Pratt et al. (1998), reported an increase in biofilm formation by E. coli strains grown in LBB and low production when using a minimal broth supplemented with a carbon source such as glucose or glycerol. When comparing the nature of both media, it can be concluded that the characteristics of the culture medium and the expression of the exopolysaccharide depend to a large extent on the microorganism and the capacity for biofilm formation in any external condition (Cáceres et al., 2018), which gives them an excellent capacity for adaptation and survival.
Biofilm expression in vitro on tomato epidermis
In this assay was observed that the serotypes O157:H7 (EHEC) and OND: H10 (EAEC) were strong biofilm forming bacteria on tomato epidermis. It is important to point out that the strain OND: H10 of the EAEC group was isolated from the autopsy of a child (Eslava et al., 1992) and is used as a positive control in biofilm assays on abiotic surfaces (polyethylene plates). The biofilm formation on the tomato epidermis of OND: H10 beginning at 24 h (Fig. 1a), which reached its maximum expression at 72h (Fig. 1b).
With relation to the time of expression, the specific behaviors of each of the strains evaluated showed statistically significant differences; in this respect, it was observed that at 24h the reference strain DL933 (O157:H7) showed 66% higher production (considering the highest peak as 100% OD) compared to the positive control strain (OND:H10), this indicates that the EAGG strains are excellent biofilm formers on biotic surfaces under natural conditions, making the aggregative E. coli pathotype especially relevant as potential triggers of epidemic outbreaks. About in the 2011 year there was an outbreak in Germany related with an E. coli strain O104:H4, which carried genes both of EAEC and EHEC (Yang, 2015). This fact lends greater impact to the present work given that this is the first time that the ability of EAEC strains to colonize raw consumption vegetables has been reported.
Our results coincide with previous studies that mention that the flagella, pili, outer membrane proteins and biofilms production, allow the bacteria to initially interact with the surfaces and then adhere in a specific manner through cell receptors (González, 2005, Ryu et al. 2004). Bacteria can communicate using chemical signals to detect cellular density and coordinate gene expression (Hughes, 2008), a process known as quorum sensing (QS). E. coli O157:H7 has been shown to utilize QS signals to communicate with plants and to regulate the expression of virulence and flagella genes (Carey et al., 2009). Cell-cell signals between the bacteria and its host are regulated by Acil-Homoserin Lactones (AHL) (HughesSperandio 2008). Carey et al. (2009), point out that these factors are involved in the formation and mobility of biofilm, resulting in the colonization of different horticultural products, which demonstrate the results of this study. (Adator et al. 2018; Lindsay et al. 2008; Park et al. 2006; Reid et al. 2001; Skyberg et al. 2006; Zhi et al. 2019).
In vivo biofilm expression under greenhouse conditions
To evaluate biofilm formation under greenhouse conditions tomato fruits at two stages of maturity and different incubation times were inoculated with different E. coli strains. The obtained results showed biofilm formation obtaining 95% confidence intervals for each of the attained means with statistically significant differences. The formation of biofilm for both strains of E. coli (O157:H7 and OND: H10) occurred in the PM and CM stages (Fig. 2a).
One important observation during the CM stage of the fruits was the fact that they presented internal changes in the mesocarp that increased during maturation. The development of biofilm in the greenhouse tomatoes began at 24 hours with a notable increase and the highest peak at 96h (Fig. 2b).
Wang et al. (2012), reported differences among E. coli strains under controlled growth conditions, pointing out that the ability to form biofilms was not restricted to a particular serotype. This could be due to the participation of the different elements in the adhesion of the bacteria, such as fimbriae, curli, cellulose, exopolysaccharide, and autotransporter proteins. Other investigators suggest that virulence genes constitute a key element for the formation of biofilm (Lajhar et al. 2018; Ogasawara et al. 2010; Uhlich et al. 2013). This allows us to conclude that the strains evaluated in the study due to their ability to develop biofilm and adhere to tomato fruits, makes them a potential threat of infectious intestinal diseases.
The formation of biofilms by E. coli strains is something that tells us about the survival capabilities of the pathogen, therefore the contact of the microorganism with the fruit is not favorable for the consumer.
Under controlled conditions, the formation of biofilm presented higher development at 96 h after inoculation, presenting readings with a mean value of 1.86 OD. Although the inoculation was controlled in the study, uncontrolled inoculation in the field during cultivation or at various points during the fruits’ processing and commercialization could occur through vehicles such as water, substrate, and inadequate crop management. Thus, it is feasible that in the lapse of a few days’ bacteria inoculated in this manner could colonize and form biofilm on tomato fruits meant for consumption. Erickson (2012) suggests that pathogenic bacteria like E. coli O157:H7 can survive on the surface of the plant, penetrate the epicarp and eventually establish in and colonize the mesocarp. This, in conjunction with the biofilm formation on the epidermis, makes it difficult to eliminate bacteria by traditional methods (Xicohtencatl et al. 2009). When a vegetable like tomato is exposed to bacteria, the bacteria tend to attach to the epidermis of the fruit; firm adhesion generally takes up to a few hours. At that point, the adhesion becomes strong enough to resist conventional washing, making bacteria removal more difficult. The situation can become more serious with persistent humidity, which allows the synthesis of polymers, and therefore, the formation of biofilm (Avila et al. 2018).
The strains analyzed in this study showed the capacity to produce biofilm on inert and live surfaces (tomato) within 24h of incubation. Furthermore, it has been reported that O157:H7 strains of E. coli can penetrate natural openings in the plant, such as the sub-stomatal cavities of the leaves (Brandl, 2008; Erickson, 2012; Kroupitski et al., 2009). Once the bacterial cells are found inside the plant or protected by an exopolysaccharide matrix, they are protected from most superficial disinfectants (Gomes et al., 2009). Therefore, if the pathogen possesses the ability to form biofilm and can adhere to plant tissues in a crop, there will be a latent risk to human health from the ingestion of the contaminated product (Deering et al. 2012; Warriner et al. 2003). Importantly, E. coli O157:H7 established on fruits and vegetable maintains its virulence to humans (Mukhopadhyay et al. 2014), such that the presence of E. coli O157:H7 in food practically guarantees a disease outbreak, which could have serious consequences (Figueroa 2011; Lajhar et al.2018; Torres, 2015).