text in 
English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Similars in
SciELO 
Permalink
Pathogens are found almost everywhere - in the air, in water and on all living and nonliving surfaces. Thus, the environments in which food products are produced, processed and transported can all cause cross-contamination. Some of the most common sources of microbial contamination of fresh produce are related to contaminated irrigation water (Cooley et al., 2007; Wright et al., 2013) and the use of improperly treated or contaminated manure (Fletcher et al., 2013; Olaimat and Holley, 2012).
The chances of bacterial contamination of fruits is high, where water and nutrients are readily available to support bacterial growth. Plants and fruit surfaces usually carry static charge or have a micro-rough texture, both factors increase bacterial contamination by facilitating bacterial adhesion and establishment (Rivera et al., 2009).
Escherichia coli (E. coli) represents a diverse group of bacteria. Pathogenic-E. coli strains are categorised into pathotypes based on their virulence factors. Six pathotypes are commonly associated with diarrhoea and collectively referred to as diarrhoeagenic-E. coli. For instance, Shiga toxin-producing-E. coli (STEC) is also referred as Verocytotoxin-producing-E. coli (VTEC) or enterohaemorrhagic-E. coli (EHEC). This pathotype is one of the most commonly associated with foodborne outbreaks.
The pathogenic E. coli O157:H7 is the most commonly STEC isolated in North America. It is also the most often associated serotype to bloody diarrhoea and haemolytic uremic syndrome (HUS). It normally lives in the intestinal tract systems of animals (e.g. cattle) (Ferens and Hovde, 2011). It was first reported as a gastrointestinal pathogen in 1982 (Riley et al., 1983). Since then, it has been recognized worldwide as a public health problem, causing diarrhea, haemorrhagic colitis and HUS (Wang et al., 2013). The serotypes O26:H11, O111:NM and O104:H4 have recently been linked to human infections, but pathogenic E. coli O157:H7 has been associated with most of the major outbreaks (Bavaro, 2012, Tzschoppe et al., 2012; Reineke et al., 2015).
Since E. coli O157:H7 is ranked as an important pathogen due to the number and severity of the outbreaks it has caused, this review focuses on its ability to adhere to the phylloplane, to internalise and even to multiply in the tissues of food plants. It also proposes some potentially valuable lines for future research in this area.
Characteristics and plant-bacterium communication
E. coli O157:H7 is a facultative Gram negative rod. Colourless colonies from 2-3 mm in diameter can be found when grown on sorbitol MacConkey agar (SMAC) or SMAC containing cefixime and tellurite (CT-SMAC). E. coli O157 is motile and possesses the flagellar antigen H7. VTEC O157 is probably the most highly infective E. coli; showing an infective dose of less than 15 cells (Teunis et al., 2008). The infective dose for EPEC strains is 2.3 x 106 (Donnenberg et al., 1998) and for Enterotoxigenic-E. coli (ETEC) is 109 (Freedman et al., 1998).
E. coli O157:H7 carries a number of virulence factors including a pathogenicity island, the locus of enterocyte effacement (LEE) that encodes gene regulators (Deng et al., 2004), adhesin, intimin (Lai et al., 2013), a type III secretion system (T3SS) (Kessler et al., 2015), chaperones and several secreted proteins as the translocated intimin receptor Tir (Lai et al., 2013).
It is well known that bacteria can communicate by chemical signals, to detect cell density and coordinate gene expression (Hughes and Sperandio, 2008). This process is called quorum sensing (QS). E. coli O157:H7 also may use QS signaling to communicate with plants, to regulate the expression of virulence and flagellar genes (Carey et al., 2009).
Cell-to-cell signaling between this bacterium and its host, are regulated by acylhomoserine lactones (AHLs) (Hughes and Sperandio, 2008). In addition, AHLs are involved in biofilm formation and motility (Carey et al., 2009).
Bacterial attachment, Adhesion and survival on plants
The first step of a bacterial infection in a plant tissue is phylloplane adhesion. Although, the presence and survival of pathogenic bacteria on leaves or fruit can be influenced by surface roughness and trichomes, environment conditions are also key to modulating bacterial community structures (Wan-Ying et al., 2015).
Environmental factors affecting bacterial behaviour and survival time on fresh produce are: temperature, pH, relative humidity and the presence of liquid water. These can be decisive for microbial activation and growth.
Additional factors relating to the ability of the bacteria to bind and proliferate in the plant are the motility on the phyllosphere, the pathogen's ability to leach nutrients, and its interactions with other epiphytic (Aruscavage et al., 2006) or phytopathogenic microorganisms. For example, Cooley et al. (2006) reported that Wausteria paucula can actively support and enhance the survival of E. coli O157:H7 in the rhizosphere and on the leaf-surface of lettuce. Similarly, Xanthomonas campestris pv. vitians, can also support the survival of E. coli O157:H7 on lettuce (Aruscavage et al., 2008).
The ability to attach to a plant surface is an important feature affecting bacterial establishment in plants. For instance, a study by Macarisin et al. (2012) showed that constituents of extracellular matrix such as curli and the polysaccharide cellulose can also be important for the attachment of E. coli O157:H7 to spinach leaves.
According to Franz et al. (2007) and Xicohtencatl-Cortes et al. (2009), adhesion mechanisms suggest that E. coli uses several pathways to colonise food plants. The pathogen is well adapted to the biosphere and can reach the sub-stomatal cavity and spongy mesophyll for survival. In a study by Prigent-Combaret et al. (2000), E. coli K-12 strains formed curli on coverslips. This enabled the bacteria to attach even to a clean glass surface.
To colonise host tissues, E. coli O157:H7 expresses intimin and other adhesins. Adhesins are a group of proteins involved in the attachment of E. coli to abiotic surfaces such as plastic or steel, and also the colonization of biological surfaces (McWilliams and Torres, 2014).
Moreover, these bacterial proteins located in the outer membrane, such as intimin and its translocated intimin receptor (Tir), are necessary for the adhesion between host cells and attaching and effacing (A/E) pathogens. Such A/E bacteria have an essential characteristic; the formation of an actin-rich pedestal for intimate attachment. Intimin is encoded by the eaeA gene, it is in the outer membrane, and is one of the most important virulence factor of E. coli strains. It is involved in the attachment process of E. coli O157:H7 and was considered the only colonization factor before other adhesins, such fimbrial and afimbrial were discovered (Farfan and Torres, 2012). Intimin is regulated by the locus LEE (Kendall et al., 2007). The expression of intimin can be regulated to respond to external factors such as osmolarity and pH (Torres et al., 2007). In a study carried out by Carey et al. (2009) in lettuce, the eaeA gene was down-regulated during the storage period at 15°C, as compared to at 4 °C.
The type III secretion system (T3SS) is a key virulence factor encoded on large plasmids. It is a complex nanomachine that allows bacteria to secrete effector proteins across eukaryotic membranes (Cornelis, 2006). EHEC O157:H7 colonizes the leaf surface via flagella and the T3SS (Xicohtencatl-Cortes et al., 2009). Similarly, Shaw et al. (2008) documented that E. coli O157:H7 present the gene to encode T3SS for the adhesion to spinach and lettuce leaves, and also EspA filaments that allow the surface attachment of bacteria.
Once bacteria adhere to a plant surface, it begins a recognition process on the plant surface. In a study by Cooley et al. (2003), the inoculation of E. coli into the soil, showed the microbe's ability to migrate up the stem and along the surface of Arabidopsis thaliana. At the end of this experiment, the bacterium was able to be recovered from leaves and flowers. This indicates that E. coli can easily migrate along the whole surface plant under controlled conditions. However, some researchers suggest that pathogenic bacteria such as E. coli O157:H7 can survive for short periods on the plant surface and may even move into the tissues (Erickson, 2012).
Continuous exposure to stress, may enable bacteria to survive under such conditions and may also enhance their tolerance to these conditions. Carey et al., (2009) reported that rpoS, the gene encoding for sigma S an alternative signal factor for stress response; and sodB Superoxide encoding for iron superoxide dismutase B, were up-regulated at 4 °C and down-regulated after prolonged storage at 15 °C, when bacteria were inoculated into lettuce. Many genes are under rpoS control and are involved with stress factors, such as to pH, temperature, or oxidative stresses (Dodd and Aldsworth, 2002).
Ultra violet (UV) radiation is one of the major factors limiting bacterial survival in the phyllosphere. Nevertheless, E. coli possess the gene rulAB (resistance to UV radiation) (Brandl, 2006), which has been reported to confer DNA repair capabilities and increases UV tolerance (Feil et al, 2005). Therefore, E. coli has the ability of withstand UV radiation.
On the other hand, biofilms are well-organised, structured communities of surface-associated cells enclosed in a polymer matrix that contains open water channels (Donlan and Costerton, 2002). Biofilms were first described in the late 1600s by Antonie van Leeuwenhoek (Donlan and Costerton, 2002). When Leeuwenhoek used acetic acid to destroy the dental plaque on his dentures; he noticed that only the free-swimming cells were killed. Till recently (c. 30 y ago), these early findings of microbial communities has been largely ignored, giving more attention to the bacteria itself.
Pathogens in biofilms live in a self-produced matrix. This matrix, integrated by polysaccharides, proteins, nucleic acids and lipids (Flemming and Wingender, 2010) allows bacteria to withstand long periods on plant surfaces and to be highly resistant to antimicrobial agents. Biofilm, by its nature, can be a source of secondary contamination which can shelter human pathogens, plant pathogens and symbionts. Thus, biofilm is a basic component of many plant-microbe interactions.
E. coli has several mechanisms related to adhesion that allows it to attach to surfaces, which are well documented. However, much is still unknown about the pathogen's signaling and recognition mechanisms that allow it to establish and survive in non-host organisms. Further, it remains unclear how it can attach to abiotic surfaces such steel, glass and polystyrene.
Interaction and internalization in plants
Saldaña et al. (2011) reported that E. coli O157:H7 uses a specific T3SS effector to open the stomatal guard cells of spinach leaves. This action allows internalization of the bacterium in plant tissues.
Other studies have demonstrated that E. coli O157:H7 can internalize within the seeds and roots, and then migrate to other tissues (Ávila-Quezada et al., 2010). The movement can be upwards from the roots to the lettuce foliage, as was shown in an experiment carried out by Solomon et al. (2002). In this experiment, the soil contained contaminated manure, and after two days, the bacterium was isolated from the edible lettuce leaf tissue.
The introduction of this human pathogen at early stages of plant development has been documented by Jablasone et al. (2005). They reported that E. coli O157:H7 was internalized in seedlings. They also suggested that this bacterium preferentially colonizes the root junctions. Since root junctions are sites that release exudates, they are potential gateway to start the bacteria internalization process. Lugtenberg et al. (2001) suggested that bacterial cell may have greater access to nutrients at root junctions, than those located more distantly, and therefore, when bacteria persist on plants, these facilitate internalization in developing seedlings.
Several studies have shown that E. coli O157:H7 is able to enter by natural openings in the plant surface, such as the sub-stomatal cavities in leaves (Brandl, 2008; Erickson, 2012; Kroupitski et al., 2009). Once the bacteria are inside the plant, following closure of the guard cells, they are protected from most superficial sanitisers (Gomes et al., 2009). Hence, special health risks are created when a human pathogen is able to enter the plant tissues (Deering et al., 2012; Warriner et al., 2003 a, b), since it is protected from washing and from many industrial sanitisers (Burnett and Beuchat, 2000).
Bacteria are usually found in low numbers on food plants but, under the right conditions, in a few hours a small number of bacteria cells can multiply to hundreds of thousands (Haas et al., 2014). For instance, the population of E. coli O157:H7 increased exponentially after inoculation into apple fruit tissue, where an inoculum of 2.5 x 102 CFU/wound, increased 3 log units during the first 48 h following inoculation (Janisiewicz et al., 1999).
Although E. coli O157:H7 has been found to internalize in many different plants, sometimes the bacteria do not succeed in becoming internalized. Host chemical properties, such as acidity, and oil, sugar and water contents, are relevant to the colonization and survival of such microorganisms. Nevertheless, high acidity in some fresh produce, does not necessarily affect the survival of pathogenic bacteria. Also, there are factors such as the strain or serovar of bacteria, the route of contamination, weather conditions, type of surface and the age of the plant, that together influence the probability of internalization of a human pathogenic bacterium within a plant.
Adaptation to new environments
It has been proposed that genes of Archaea species are present in the genome of E. coli O157:H7 strain EDL933 (Faguy, 2003) which may confer a range of capabilities for adaptation to new environments. Genes of the core genome such as the metabolic genes can be transferred and can improve fitness of the strain under certain environmental conditions. For instance, a few strains of E. coli can ferment glucose and the acquisition of these genes can improve its adaptation to a new environment (van-Overbeek et al., 2014).
A study by Szmolka and Nagy (2013) suggested that human pathogens may also acquire genes from other plant-associated bacterial species. This situation leads to new phenotypes of increased persistence in plants and broadens the utilization spectra of nutrients available on plants (van-Overbeek et al., 2014). The appearance of these new features, which are encrypted in the genome of the bacterium, help to improve its fitness in a food plant.
Outbreaks
Once E. coli O157:H7 has been established in fruits or vegetables, it does not lose its capacity for virulence to humans. This is proven by all the incidents linked to this pathogen and associated with the consumption of contaminated fruits and vegetables (Mukhopadhyay et al., 2014). This bacterium was first recognized as a human pathogen in 1982, it was linked with meat products. Over the next ten years, however, a range of food-produce associated outbreaks occurred (Rangel et al., 2005).
Outbreaks caused by E. coli O157:H7 are more common in the western hemisphere, but are certainly not confined to this part of the world, since one of the most notable cases was in Asia in 1996 (Michino et al., 1999). Here, the outbreak was caused by white radish consumption in Sakai City, Osaka, Japan, where more than 6000 schoolchildren were affected (Watanabe et al., 1996). Approximately 1000 of them were hospitalised with severe gastrointestinal symptoms, and about 100 had complications of HUS (haemolytic uraemic syndrome), which resulted in three deaths.
Many outbreaks have been reported in the United States (US), Canada and Great Britain. Fresh-produce associated outbreaks in the US are now one of the main vehicles linked to E. coli O157:H7, and represent 34 % of all foodborne outbreaks (Rangel et al., 2005).
A multi-state outbreak caused by packaged spinach occurred in 2006. It affected people of 26 States in the US and resulted in 183 confirmed infections and three deaths (Center for Disease Control and Prevention, 2006). The same year, epidemiological studies detected EHEC in spinach (Center for Disease Control and Prevention, 2006).
More recently, the Center for Disease Control and Prevention (2014) reported a four State outbreak of E. coli STEC O157:H7 infections linked to ready-to-eat salads, where 33 persons were infected, no deaths were reported. Therefore, in the past 25 years many outbreaks have been associated with the presence of this pathogen on the surface or internalized in food plants and fruits. Based on the above, it is reasonable to ask if this human pathogen has always possessed such an intrinsic ability to extend its life in the environment (soil, water, plant) without losing its human pathogenicity, or has this capacity been developed more recently?
Sources of contamination
Reports identify manure as the major source of contamination in the field, followed by untreated irrigation water (Aruscavage et al., 2006; Castro-Ibáñez et al., 2015; Ceuppens et al., 2015; de Quadros et al., 2014; Erickson et al., 2010; Oliveira et al., 2012). Manure used as organic fertilizer can significantly increase total coliform contamination as well as increasing the presence of E. coli (de Quadros et al., 2014; Machado et al., 2006). Pathogens such as E. coli (VTEC) O157:H7 and pathotypes of diarrhoeagenic E.coli are transmitted through water and organic fertilisers (Oliveira et al., 2012).
Several studies have pointed to flies (Talley et al., 2009), arthropods, birds, wild mammals and reptiles (Wright et al., 2013) as other major sources of pathogen transmission. It has been shown that if fresh produce is grown near livestock, these vectors can easily carry faecal contaminants to plants (Doyle and Erickson, 2012; Fletcher et al., 2013; Wasala et al., 2013). Also, flies can pick up E. coli O157:H7 from manure and deposit it onto the phylloplane of plants, where the bacteria can then colonise and multiply (Talley et al., 2009). Despite such reports, the dissemination of this bacterium by flies to fresh produce requires further investigation, for it would seem to conflict with reports suggesting that E. coli O157:H7 can survive for only two to four days on a leaf surface (Wasala et al., 2013). Other studies report that this pathogen can survive on the phylloplane of spinach leaves for 14 days (Mitra et al., 2009). These authors also claim that leaf surface bacterial colonization can increase over time. Meanwhile, Wasala et al. (2013) suggest that leaf surface colonization, by the strain ATCC4388, does not increase when exposure time increases (vegetal tissue-bacteria).
On the other hand, it has been reported that the survival of E. coli O157:H7 in soil is inversely related to the microbial community diversity already present. Hence bacteria levels can greatly vary over time in such habitats (van Elsas et al., 2011).
Many faecal contamination studies have been carried out around the world. For example, recent work in South America has recorded the microbiological quality of fruits and leaf vegetables. In Venezuela pre and post-harvest, strawberries, guavas and peaches were apparently within the permissible limits for sanitary quality (Gil et al., 2011). While in Brazil faecal coliforms were detected in lettuce, but no E. coli O157 was found in this study of 36 samples (de Quadros et al., 2014).
Studies in Mexico have shown that plant products are usually free of E. coli. The sampling included chipotle pepper (Ávila-Quezada et al., 2009), Starkimson apple, Golden Delicious apple, peach, California pepper, jalapeño pepper, Serrano pepper, saladette tomatoes, grape tomatoes and cantaloupe (Ávila-Quezada et al., 2008), and habanero pepper (Lugo-Jiménez et al., 2010). Moreover, Rivera et al. (2009) reported the presence of faecal coliforms in parsley, lettuce and radish in 36.8 % of the samples analyzed. Of these samples, 24 % were positive for E. coli. Muñoz et al. (2013) assessed faecal contamination in cabbage, lettuce and spinach, vegetables which are commonly involved in E. coli outbreaks. They found that cabbage and spinach were the most contaminated vegetables, and these often exceeded the regulatory limits for faecal coliforms.
The presence of faecal coliforms, without the identification of E. coli, suggests this indicator is not always carrying this bacterium, or not been detected due to its ability to enter in a dormant state (van Elsas et al., 2011). Strictly, vegetables not eaten raw, should not be part of this pathogen's transmission route; nevertheless, hand-to-mouth transfer during handling and food preparation is still possible. It is commonly thought, pathogens are removed by washing but it has been shown that significant numbers may remain after conventional washing and so can persist on fresh produce (Ginestrea et al., 2005). Hence it is recommended, that harvest and handling procedures be examined throughout the supply chain (Berger et al., 2010).
For food handlers, the most recurrent deficiency in sanitary practice is the lack of adequate hand-washing (Ávila-Quezada et al., 2009). Furthermore, the efficacy of many disinfectants for destroying microorganisms has come into question in recent years, so their frequent use does not necessarily constitute safe practice.
When a vegetable is exposed to bacteria, they tend to stick to the surface. However, firm attachment usually takes a few hours. Only then, does attachment becomes sufficiently strong to resist conventional washing, rendering their removal more difficult. The situation can be aggravated by persistent wetness, as this allows polymer synthesis and hence biofilm formation (Avila-Quezada et al., 2010; Flemming and Wingender, 2010).
Prevention efforts
As we have seen, the presence of E. coli O157:H7 in a food practically guarantees a disease outbreak. It is generally true that effective actions designed to prevent pathogen entry to fruits and vegetables, are more desirable than actions designed to remove or inactivate them. Thus, methods used to ensure food safety should be based on the adoption of Good Agricultural Practices (GAP), Good Manufacturing Practices (GMP), and Hazard Analysis and Critical Control Points (HACCP). Nevertheless, even these do not absolutely eliminate pathogen contamination.
Throughout the world, most cases of enteric disease are attributable to the consumption of contaminated food. However, the sources are not always properly identified, so often remain as mere, overstated assumptions. This is probably done by authorities to minimize public alarm. The fact that an outbreak linked to fruit or plant product occurs indicates that the outbreak is the result of an occasional contamination event, which can be very difficult to identify if proper records are not available for traceability.
A number of antimicrobial treatments have been tried to control E. coli O157:H7 but, significant reductions in bacterial contamination of vegetables can be very difficult to achieve.
Future research
A number of interesting questions arise from the above. These include: How are the bacterial effectors delivered to the inside of the non-host plant cells? What is the nature of the communication pathway between the microbe and the plant cell? How does protein regulation induce signaling and recognition?
Future work will address these questions through biological systems studies using postgenomic. This approach should allow understanding of gene and protein expression during recognition and internalization of E. coli in food plants. A better understanding of the interaction of this human pathogen with food plants is expected to provide important new scientific information, useful for developing new and improved strategies for minimising consumer risk.
