Although there is not a huge body of scientific evidence available currently, there is enough to say that the method of application of irrigation water to fresh produce has an effect the microbiological risks associated with the crop. In general, and as might be expected, keeping water away from the edible parts of ready to eat crops that are consumed without cooking results in a lowered risk of an outbreak of foodborne illness. Thus, when choosing which irrigation method is most appropriate for each particular crop; consideration should be given to the location of the edible part. For those crops which grow above ground with a little clearance (e.g. strawberries, lettuce); Figure 1 below shows the most and least risky irrigation methods.
In addition, to keeping water away from the edible parts of crops, a number of publications have highlighted the importance of attachment in the contamination of fresh produce (Song et al., 2006; Critzer and Doyle, 2010). In simple terms, zoonotic agents have an ability to recognise and glue themselves to the surface of some (but not all) types of fresh produce. Different pathogens are better at attaching to different crops. Investigation of attachment and the generation of lists describing what types of fresh produce are most susceptible to which pathogens, is currently in the very early stages. There is little meaningful information available for growers currently. Hoever, it is expected that a lack of helpful data will change over the next 2-3 years. The results of those few published experiments describing the application of irrigation water to various produce is summarised below.
Flood irrigation versus subsurface irrigation
Song and colleagues (2006) compared irrigation undertaken by flooding furrows with a subsurface drip system to see which had the best microbiological safety. In their experiment, melons, lettuce and bell peppers were grown as crops. The water used for irrigation was deliberately contaminated with E. coli, Clostridium and a bacterial virus. Over a 14 day period, the numbers of these microorganisms were followed on the crops, soil surface and also in the subsoil (10 cm depth). Overall, greater contamination was recovered from the furrow-based system when compared with subsurface drip. However, the decline of the microorganisms was fastest on phylloplanes and the soil surface and the microorganisms which were delivered below the soil surface died off most slowly. For the produce tested, Song and colleagues concluded that “subsurface drip had a great potential to reduce the health risks associated with the use of contaminated irrigation water.” The implications for underground crops such as carrots were not investigated by Song, but clues to the fate of zoonotic agents in ground crops are available.
A study which investigated the fate of Salmonella in soils (NB not contamined with poor quality irrigation water) used to plant radish, turnip, and broccoli found there was a higher prevalence of contamination as compared with lettuce, tomatoes, and carrots grown in the same contaminated soils (Barak et al., 2008). Salmonella contamination of lettuce and tomatoes was exceptionally low, suggesting that the strains of Salmonella used by the researchers do not readily attach, grow, or are attracted chemotactically to the root leachates of these crops.
Contamination of above ground produce by sprinklers
Hutchison et al. (2008) determined the implications of using contaminated irrigation water applied by sprinkler onto newly planted baby spinach and lettuce on two different soil types in the UK. The fate of the zoonotic agents (E. coli O157, Campylobacter, Listeria monocytogenes and Salmonella Enteriditis) was followed on the produce and in the soil over several different growing conditions over time. The experiments showed that overhead sprinklers can contaminate produce grown above ground. In conditions of bright sunshine low rainfall, the numbers of bacteria became too low to count within 2 weeks for both crop types. Later work by Fonseca et al. (2011) confirmed that bright sunshine could rapidly reduce numbers of E. coli. In Arizona, E. coli applied to crops using overhead sprinklers was not detectable after 7 days. The Fonseca study also reported that water applied using furrows increased the soil survival of E. coli.
However, other workers have shown persistance for almost 6 months during a winter growing period (Islam et al., 2004). Oron (2002) estimated that if drip irrigation was used to apply contaminated irrigation water, the microbiological risks of a food disease outbreak were 100-1000 times lower than if the same water is applied by overhead sprinkler. In a systematic review of risk factors for produce contamination, Park et al (2012) concluded that spray irrigation using contaminated water was one of the higest risks for causing contamination of the underside of leaves.
The implications of using contaminated water in hydroponic crop rearing was investigated by Guo et al., 2004. Colonisation of tomato plants by Salmonella was used as the model of the study. After nine days of exposure of the tomato plant roots to a solution containing large numbers (4.46 to 4.65 log10 CFU of Salmonella/ml), isolation of large concentrations of bacteria (more than log10 3.0 /g plant tissue) was observed irrespective of whether the roots were intact or not at the commencement of the experiment. Although the study showed there there is the potential for a major outbreak from a hydroponic system if large scale contamination occurs, the numbers of bacteria used in the experiment were exceptionally high, and arguably too high to represent a credible real life scenario.
The orgininal studies of Guo and colleages were confirmed and extended by Koseki et al (2011) who compared Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes applied to either roots or leaves in hydroponically grown spinach. The hydroponic solution was inoculated with either a low (103) or high (106) concentrations. In parallel, the human pathogens were inoculated onto the growing leaf surface by pipetting, at concentrations of 103 and 106 CFU per leaf. In keeping with the findings of Gou et al (2004), contamination was observed only through the root system and at the the higher inoculum. For all the pathogens assessed, plant colonisation was rare when the lower inoculum was used. Compared with the roots, contamination through the leaf occurred infrequently even when the inoculum level was high. Statistical analyses of the Koseki results showed that the risk of contamination from the roots was 7 times higher than the risk of contamination from the leaves. In addition, the risk of contamination by L. monocytogenes was 30% of that for Salmonella and E. coli O157:H7.
One property of hydroponic systems not assessed by either of these studies was that hydroponic water contains nutrients which can be utilised by bacteria and the solution is held at a near constant temperature, Thus there is apotential for an uncontrolled growth of zoonotic agents to to reach the higher inoculum levels used by these studies.
Water transport pipework
Papesky et al. (2012) investigated the influence of irrigation water pipework to irrigation water quality. Small squares of stainless steel were placed inside the pipes used to convery water to a sprinkler irrigation system for a week inbetween irrigation events. Water from a supply creek, sprinkler water, residual water left in the pipes from the previous irrigation and biofilms on the steel squares were tested for numbers of E. coli. High numbers of E. coli were found in the water remaining in irrigation pipes between irrigation events, suggesting growth of E. coli in the pipes. However, the numbers of bacteria growing as biofilms on pipe walls (as assessed from the steel squares) was estimated to be higher than in water in the pipes. The report authors recommended flushing of the irrigation system prior to irrigating as a way to reduce the risk of microbial contamination of produce. Consequently, the water risk assessment tool will shortly be updated to include questions asking about flushing of pipework prior to irrigating and water quality sampling at point of application.
Water and soil splash
Although it is best practice to keep irrigation water away from the edible parts of the crop, there are a number of other factors that need to considered in conjunction with water-crop contact. The first is that pathogenic bacteria and other microrganisms survive longer on, and longer still, in soil when compared with crop phylloplanes (i.e. the surface of a crop). If there is a possibility that heavy rain can fall directly onto fields containing crops, then soil splash should be considered when assessing risks. Soil splash can transfer zoonotic agents from the soil surface onto the surfaces of crops as shown in Figure 2 below.
HDC have funded work (Monaghan and Hutchison, 2012) which has attempted to determine how far pathogens contained within contaminated soils can be transfered by raindrops. Figure 3 below shows an agar (agar is used for growing bacteria) strip. Each of the purple spots is a colony of E. coli. The E. coli was transferred to the agar stripfrom contaminated soil by a single raindrop. A large raindop of 50 microlitre volume can transfer E. coli up to 0.5 metres (20 inches) to the side of the point of impact. During the transfer, bacteria can reach a height of 0.3 metres (12 inches). The smaller the raindrops the less distance that pathogens are spread (Hutchison and Monaghan, 2012)
There were no significant differences with regard to bacterial transfer between the two loamy soil types which were compared in the Hutchison and Monaghan (2012) study. However, Park et al (2012) systematically identified that growing produce on clay soil was a risk factor for contaminated produce, although the nature of the risk was not identified.
Chemotaxis and rhizosphere growth
Habteselassie and co-workers (2010) have undertaken some novel studies using E. coli which were genetically modified to allow easier isolation from soil and crops. Introduction of E. coli to soil via manure or manure in irrigation water showed that E. coli could colonise the lettuce rhizosphere. Regardless of whether manure or water was used as the introduction method, 15 days after rhizosphere establishment, E. coli was detected on the phyllosphere of lettuce at 2.5 log CFU/g. E. coli persisted in the bulk and rhizosphere soil throughout the study duration of 41 days, but were not detected on the external portions of the phyllosphere after 27 days. The Habteselassie study is important because it establishes that E. coli can move through soil towards a nutrient source. Furthermore, the roots of plants such as lettuce can leak carbon compounds into the soil and these exudates can be used by E. coli for growth and multiplication. A review of related work by Jacobson and Bach (2012) between 2010 and 2012, has provided further confirmatory evidence that bacterial growth can occur in a number of rhizospheres including tomatoes and spinach.
A second consideration which has created increasing concern for the fresh produce industry in recent years has been the issue of internalisation of human pathogens into crops. An easy to read summary of the literature relating to internalisation has been prepared by Deering et al. (2012).
The imporant points relating to internalisation are:
1. Human pathogens such as Salmonella and E. coli O157 are present in water.
2. When seeds are soaked in contaminated water, or contaminated water is used for irrigation or pesticide application or contaminated water is used to wash harvested produce, these human pathogen can invide the plant tissues.
3. Invasion can occur through natural openings in the plant such as stomata, lenticels and lateral roots. Invasion can also occur via damaged plant tissue including cuts made to harvest crops.
4. Once inside the crop, the human pathogens are protected from harsh environmental conditions and can possibly survive for extended periods. However the insides of plants are not sterile and human pathogens attempting to establish in a plant niche must compete with the indigenous bacteria for nutrients and space before colonisation can be accomplished. In many cases human pathogens do not successfully establish inside the plant.
5. Internalisation has been described in roots, lateral roots, root hairs, stems, leaves and fruits.
Although there has been a lot of work in recent years studying internalisation, it has never been proven as the cause of an outbreak of human illness. Furthermore, the numbers of cells which are internalised into plants are always very low. In most cases studied in a laboratory, the numbers of internalised pathogens are below the infectious dose required for illness- even if the entire plant was consumed by one person. At the current time, it is not clear what role, if any, is played by internalisation in terms of fresh produce food safety.
In addition to irrigation, water is used to apply pesticides to crops. A risk assessment was undertaken by Stine and co-workers (2011). Using source water contamination data from the USA, which is broadly comparable with similar data in the UK, a 1:10,000 annual risk of infection for Salmonella and enteric viruses in surface and ground waters could easily be exceeded for some water sources. To reduce the risks associated with the consumption of fresh produce, the authors advise that if surface water is used to prepare pesticides in spray applications, it should be evaluated for its microbiological quality prior to application.
Decontamination and irradiation
Although it is legal to do so in the UK, it is considered highly unlikely that growers would irradiate fresh produce to ensure it is free from human pathogenic micro-organisms. The Food Irradiation Regulations 2009 (closely related different versions for England, Scotland, Wales and NI) set out the requirements for producing, importing and selling irradiated food in the UK. The main issue with treated fruits and vegetables is that they are required to be clearly labeled as being exposed to radiation.
Although unlikely to be of practical use, Puerta-Gomez et al (2013) have undertake mathematical modelling of the likely outcome of adopting irradiation as a critical control point. The work shows that the spinach is still a highly safe product when cross-contamination occured during post-harvest washing if the produce was harvested at 20oC, stored for at least 5h, washed with water containing chlorine at 220 ppm, and then exposed to a 1 kGy dose of radiation.
Fonseca, J.M., Fallon, S.D., Sanchez, C.A., & Nolte, K.D. (2011). Escherichia coli survival in lettuce fields following its introduction through different irrigation systems. Journal of Applied Microbiology, 110,893-902
Guo, X., van Iersel, M.W., Chen, J., Brackett, R.E. and Beuchat, L.R. (2002) Evidence of association of Salmonellae with tomato plants grown hydroponically in inoculated nutrient solution. Appl Env. Microbiol. 68,3639-3643
Habteselassie, M.Y., Bischoff, M., Applegate, B., Reuhs, B., and Turco, R.F. (2010). Understanding the role of agricultural practices in the potential colonization and contamination by Escherichia coli in the rhizospheres of fresh produce. Journal of Food Protection, 73,2001-2009.
Hutchison, M.L., Avery, S.M. and Monaghan, J.M. (2008). The air-borne distribution of zoonotic agents from livestock waste spreading and microbiological risk to fresh produce from contaminated irrigation sources. Journal of Applied Microbiology 105,848-857.
Islam, M., M. P. Doyle, S. C. Phatak, P. Millner, and Jiang, X. (2004). Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. Journal of Food Protection 67,1365– 1370.
Koseki, S., Mizuno, Y., and Yamamoto, K. (2011). Comparison of two possible routes of pathogen contamination of spinach leaves in a hydroponic cultivation system. Journal of Food Protection, 74, 1536-1542.
Monaghan, J.M. and Hutchison, M.L. (2012). Distribution and decline of human pathogenic bacteria in soil after application in irrigation water and the potential for soil-splash-mediated dispersal onto fresh produce. Journal of Applied Microbiology, 112, 1007-19.
Park, S., Szonyi, B., Gautam, R., Nightingale, K., Anciso, J. and Ivanek, R. (2012). Risk factors for microbial contamination in fruits and vegetables at the preharvest level: A systematic review. Journal of Food Protection, 75, 2055-2081
Puerta-Gomez, A., Kim, J., Moreira, R., Klutke, G. and Castell-Perez, M. (2013). Quantitative assessment of the effectiveness of intervention steps to reduce the risk of contamination of ready-to-eat baby spinach withSalmonella. Food Control, 31, 410-418
Song,I., Stine,S.W., Choi,C.Y. and Gerba,C.P. (2006) Comparison of crop contamination by microorganisms during subsurface drip and furrow irrigation. Journal of Environmental Engineering-Asce 132, 1243-1248.