SALAD DAYS (Part II) – an investigation into the microbiological safety of prepared salads

| February 28, 2009

Salad pic

Green salad leaves are attracting growing international concern as a potential source of foodborne pathogens, not only in domestic markets, but also in international trade. Vegetables such as lettuce and spinach are vulnerable to contamination at many stages in the food supply chain, but it is contamination in the field that is most worrying, mainly because the options for subsequent decontamination of green salad leaves are so limited. But recent research is beginning to reveal complex relationships between microbes and plants, which seem to play an important role in the contamination process. In the future, it may be possible to use this knowledge to devise innovative new ways to minimise, or even prevent, contamination in the field.

As we saw in Part I of this article, fresh fruit and vegetables are responsible for only a small fraction of outbreaks of foodborne disease occurring in Europe and North America. Recent estimates suggest that around 5% of outbreaks can be linked to fresh produce in general and most of these are probably a result of poor hygiene in food preparation in the home or in the catering industry. But these figures hide a trend that is the subject of growing concern amongst food safety experts worldwide. A disproportionate number of outbreaks, often large and occasionally international, have been associated with leafy salad greens, especially lettuce and spinach, where contamination seems to have occurred before harvesting. Furthermore, the number of such outbreaks has been growing in recent years.

The problem has become so acute that a joint WHO/FAO expert meeting on microbiological hazards associated with fresh produce in October 2007 concluded that leafy green vegetables should be given the highest priority in terms of fresh produce safety. The main reasons for this concern were the potential to cause large and widespread foodborne disease outbreaks and the possibilities for post-harvest processing to “amplify”, rather than reduce, contamination. Food safety experts are beginning to realise that the global trade in leafy salad greens, driven by consumer demand for healthy eating, is in fact an important vehicle for spreading human pathogens, such as Salmonella enterica, verotoxigenic E. coli (VTEC) and Shigella species around the world.

How contamination occurs

It is easy to come up with a list of possible sources of pathogen contamination for crops growing in the field: organic manure added to the soil; domestic and wild animals; birds; water contaminated with animal or human faecal material; poor hygiene by agricultural workers. What is more difficult is tracing contaminated salad greens back to an individual grower and establishing the source of contamination. For several of the large national and international outbreaks that have occurred in Europe, it proved impossible to follow highly complex supply chains back to a grower. For example, in 2000, an outbreak of Salmonella Typhimurium DT104 infection in England and Wales affected at least 361 people. The outbreak was linked with consumption of lettuce, but tracking it back to the growers proved very difficult. Nevertheless, three farms that might have supplied contaminated lettuce were audited, but no unsafe practices were identified. Similarly, the outbreak of Salmonella Senftenberg in 2007, which caused cases in the UK, the Netherlands, Denmark and the USA, was strongly linked to fresh pre-packed basil grown in Israel, but environmental investigations there did not reveal the source of the contamination. Another international outbreak, this time of Salmonella Thompson, caused illnesses in Norway, Sweden and the UK in 2004 and was linked to rucola lettuce grown in Italy. On this occasion, other Salmonella serovars and other faecal bacteria were found in samples, prompting investigators to suspect sewage contaminated irrigation water as the source, but this could not be confirmed. Investigators had more luck when looking into an outbreak of 120+ cases of E. coli O157 in Norway and Sweden in 2005. This outbreak was associated with consumption of iceberg lettuce, this time grown locally in Sweden, allowing a more rapid and thorough investigation. The source was identified as a small stream used by the grower to irrigate the crop.

Perhaps the most exhaustively investigated event was a high profile E. coli O157:H7 outbreak, which occurred in the USA in 2006. This outbreak affected more than 200 people across 26 states and was linked to pre-washed, bagged fresh spinach with a well known brand name. The spinach was traced back to farms in the Salinas Valley in California, the source of at least 10 other outbreaks since 1995. The outbreak strain of E. coli O157:H7 was isolated from a stream, cattle manure and the faeces of wild pigs in the area. This suggested that the growing environment could become contaminated from more than one source, with the pathogen being transported from one site to another by water flow and by the movement of domestic and wild animals. The findings present a more complex picture than anyone expected and show the need for more research into the sources and movement of pathogens in the field. This outbreak – and the history of previous problems in California – has been a catalyst, not only for more research into the sources and movement of pathogens in the field, but also for regulatory action to try and prevent further outbreaks. The FDA had already launched a “lettuce safety initiative” earlier in 2006, but the spinach outbreak prompted action to extend the initiative to all leafy greens.

Control and decontamination options

That brings us neatly on to how contamination of salad greens with pathogens can be prevented and/or controlled. By far the most productive approach at the growing stage is based on the established principles of Good Agricultural Practice (GAP). GAP has a very wide scope and is intended to help farmers and growers minimise environmental pollution and protect natural resources. These general principles have been used as the basis for developing more specific and detailed codes of practice for growers. A number of such codes have been published around the world, by regulatory authorities, trade associations and by bodies representing retailers, but adopt a similar approach to controlling foodborne pathogens. For example, in the USA the FDA Center for Food Safety and Applied Nutrition has published a Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables. The Guide is designed to help both domestic and foreign growers and packers prevent microbial contamination of fresh produce. It covers the following key elements:

  • Water – avoiding pathogen contamination of agricultural water (e.g. irrigation), processing water (for washing produce etc.) and cooling water
  • Manure – safe use of manure, including animal faeces and sewage sludge, as a fertiliser
  • Workers – worker health, provision of sanitary facilities and hygienic practices
  • Field sanitation – cleaning harvesting equipment and avoiding cross contamination
  • Packing facilities – cleaning and hygiene in packing areas, pest control
  • Transport – cleaning and hygienic practices in transport and proper temperature control
  • Traceability – documentation and labelling procedures to allow complete traceability from farm to retailer

Salad pic2

Other guides adopt a similar approach and provide useful and effective advice for growers and packers to prevent or minimise contamination. But such guidance can only achieve so much. Even the most diligent and responsible grower cannot hope to completely eliminate all possible sources of contamination. Any crop growing in the field is vulnerable to contamination from sources not under the growers’ control, such as wild animals and birds. To ensure absolute safety, post-harvest treatments would appear to be needed.

This is where the real problem lies. At present, completely effective methods of treating fresh produce to eliminate microbial pathogens without destroying the sensory qualities of the product are simply not generally available. Traditionally, the main decontamination method available to the produce industry has been washing with potable water, often chlorinated to provide some biocidal activity. Washing with chlorinated water is routinely applied to leafy green vegetables like lettuce, especially for pre-packed ready-to-eat salads, but its effectiveness is limited. For example, the investigation into the 2006 US E. coli O157 outbreak linked to spinach found that the spinach had been washed prior to packing and that further washing by the consumer made no difference to the chances of becoming infected. Washing rarely achieves microbial reductions of more than 100-fold even when 100 ppm or more of free chlorine is present. Other sanitisers have been tried, including chlorine dioxide and ozonated water, but all have their disadvantages. A really effective practical washing treatment, which does not cause a taint or damage the product, and which does not leave any potentially toxic residues, has yet to be developed.

This has led some food safety experts to back rather more drastic decontamination methods, notably irradiation. Researchers have shown recently that a relatively low dose of 1 kGy – enough to give a theoretical 5-log reduction in E. coli O157 – did not damage the quality, or nutritional value, of fresh cut vegetables, including lettuce. In fact, the treatment actually delayed spoilage to some extent. In the USA, the FDA announced last August that it was proposing to allow irradiation of fresh spinach and iceberg lettuce with an absorbed dose of up to 4 kGy to control pathogens. Although consumer resistance to food irradiation has been showing signs of diminishing recently, especially in North America, it is doubtful whether the process represents a practical and cost effective answer to the problem of pathogens in green salad vegetables.

Some microbiologists believe that the best solutions lie in interventions that can be applied in the field, rather than in better decontamination processes, and they point to some very recent research, which may reveal new ways to attack pathogens before they become established in fresh produce.

New insights may lead to new controls

While the sources of contamination for crops growing in the field – animals, polluted water etc. – may be clear, the mechanisms by which human pathogens attach themselves to plants are much less obvious. If contamination were a purely passive process one might expect that decontamination by washing lettuce and other crops with chlorinated water would be a very effective. The evidence suggests that this is not the case and that other factors are involved. While interactions between plant pathogens and plants have been widely studied for many years, interactions between human pathogens and plants is a relatively new field of research. Despite this, some interesting and potentially useful findings have already been published and the emerging picture seems to be surprisingly complex.

So far the evidence indicates that contamination of food plants by human pathogens is more likely to occur in the phyllosphere (the above ground plant surfaces) rather than via the root systems. This seems to be especially so for lettuce. There is also evidence that both Salmonella and E. coli can become internalised within leafy tissue, either through damaged areas, or cut surfaces produced during harvesting, or by entering the plant through the stomata. For example, E. coli O157:H7 cells have been shown to be attracted to the guard cells surrounding the stomata on leaf surfaces. Other studies have shown that pathogenic bacteria may be trapped in vesicles produced by protozoa naturally present on leaves. These trapped bacteria may also be protected from washing and from sanitising chemicals.

Human pathogens may also become components in microbial biofilms attached to leaf surfaces and this may help to attach the cells more firmly and to protect them from washing and sanitising chemicals. There is even evidence that some pathogens can multiply on leaf surfaces. A US study published in 2008 showed that, given warmth and moisture, both E. coli O157:H7 and Salmonella enterica could multiply by up to 100-fold in the phyllosphere of young lettuce plants and on the harvested leaves of older plants. This suggests that these bacteria are certainly not simply present as passive contaminants. Growth of human pathogens inside leaf tissues has not yet been conclusively demonstrated, but this too is a possibility.

Some researchers have now begun to look more closely at bacterial attachment in the phyllospere of salad greens. For example, a team from Imperial College in the UK looked at VTEC strains O157 and O26 and their attachment to salad leaves, and were able to identify how they were attached. These pathogens seem to use EspA filaments (important factors in the infection of mammalian host cells) projecting from the cell wall to attach to leaf surfaces. This is interesting because it suggests that some pathogenic bacteria are able to use the mechanisms that they would employ to attach to, and infect, cells in the guts of their normal hosts to attach to the surfaces of salad leaves, although they don’t cause disease in the plant. This also suggests that some strains of VTEC may be much better at attaching to plants than others, just as some are more virulent human pathogens than others.

The same team have also looked at the strain of Salmonella Senftenberg associated with the international outbreak linked to Israeli grown fresh basil and investigated how it was able to attach to the leaf surfaces. They found that the bacteria were able to bind to basil, lettuce, rocket and spinach leaves and they reported that flagella could be seen linking the bacterial cells to the leaf surface. Deleting the gene responsible for flagella production reduced the amount of attachment. But Salmonella Typhimurium did not behave in the same way – it was not affected by deletion of the same gene – and the study’s authors suggest that different Salmonella serovars could use different leaf attachment mechanisms. They also reported that some leaf types were much less susceptible to Salmonella contamination than others.

This research is still in its infancy, but if we can learn more about these attachment mechanisms and how they work, we may be able to gain a better understanding of the risk factors involved and perhaps devise controls to inhibit attachment in the field. Perhaps it will be possible to breed crop varieties that are resistant to pathogen attachment, or it may be feasible to spray crops with compounds that inhibit the attachment process. What seems certain is that a lot more research will be needed before we can solve the problem of contaminated salad crops. We may just have to be content with minimising the risk through GAP and good hygiene until then.

Key references

Spinach-associated Escherichia coli O157:H7 outbreak, Utah and New Mexico, 2006
Grant, J. et al
Emerging Infectious Diseases, 2008, 14(10)

Microbial contamination of fruit and vegetables and the behaviour of enteropathogens in the phyllosphere: a review
Heaton, J.C. & Jones, K.
Journal of Applied Microbiology, 2007, 104(3), 613-26

Prepared salads and public health
Little, C.L. & Gillespie, I.A.
Journal of Applied Microbiology, 2008, 105, 1729-43

Interaction of Salmonella enterica with basil and other salad leaves
Berger, C.L. et al
The ISME Journal, 2009, 3(2), 261-5

Leaf age as a risk factor in contamination of lettuce with Escherichia coli O157:H7 and Salmonella enterica
Brandl, M.T. & Amundson, R.
Applied and Environmental Microbiology, 2008, 74(8), 2298-2306

Enterohaemorrhagic Escherichia coli exploits EspA filaments for attachment to salad leaves
Shaw, R.K. et al
Applied and Environmental Microbiology, 2008, 74(9), 2908-14

Tags: , , ,

Category: Features

Comments are closed.