Population dynamics of Escherichia coli inoculated by irrigation into the phyllosphere of spinach grown under commercial production conditions

https://doi.org/10.1016/j.ijfoodmicro.2010.08.022Get rights and content

Abstract

Recent outbreaks of food-borne illnesses associated with the consumption of fresh produce have increased attention on irrigation water as a potential source of pathogen contamination. A better understanding of the behaviour of enteric pathogens introduced into agricultural systems during irrigation will aid in risk assessments and support the development of appropriate farm-level water management practices. For this reason, the survival dynamics of two nalidixic acid resistant strains of Escherichia coli after their spray inoculation into the phyllosphere and soil of field spinach were examined over two growing seasons. E. coli strains NAR, an environmental isolate, and DM3n, a non-pathogenic serotype O157:H7, were applied at rates of 104 to 107 cfu/100 ml to the fully developed spinach plants that arose subsequent to the harvesting of their upper leafy portions for commercial purposes (secondary-growth plants). After 72 h, E. coli on spinach were reduced by 3–5 logs. Culturable E. coli were recovered from plants up to 6 days post-inoculation. Survival in soil was greater than in the phyllosphere. Under ambient conditions, the mean 72 h first order decay constant computed by Chick's Law was 0.1 h−1. Although light reduction studies indicated UV irradiation negatively influenced the persistence of E. coli, a simple relationship between UV exposure and phyllosphere E. coli densities could not be established. E. coli introduced to the leafy portions of spinach via spray irrigation displayed rapid declines in their culturability under the open environmental conditions experienced during this study. A 6 day period between the last irrigation and harvest would minimize the risks of E. coli survival in the spinach phyllosphere. E. coli NAR was identified as a possible surrogate for the O157:H7 strain, DM3n.

Introduction

Fresh produce is now considered to be a significant source of food-borne illness in humans (Doyle & Erickson, 2008, Heaton & Jones, 2008, Holden et al., 2009, Little & Gillespie, 2008). Lettuce and spinach are of particular concern as they have been linked to significant outbreaks of the enteric pathogen, Escherichia coli (E. coli) O157:H7 (Delaquis et al., 2007, MMWR, 2006). Valentin-Bon et al. (2008) highlighted the risk of fresh vegetables being potential vehicles for the transmission of food-borne disease when they found that 13% and 18% of retail-level bagged spinach and lettuce samples, respectively, carried generic E. coli. Irrigation water has been identified as a potential source of bacterial contamination for fresh produce (Aruscavage et al., 2006, Heaton & Jones, 2008, Steele & Odumeru, 2004).

To minimize health risks associated with the application of irrigation water, various regulatory agencies have developed guidelines that suggest maximum acceptable levels of indicator organisms at which it is considered safe to use bacteria-containing water for such purposes. In Canada, the Council of Ministers of the Environment recommends that irrigation water containing < 100 cfu/100 ml of generic E. coli be applied to vegetables that are usually eaten uncooked (CCME, 2003). As a secondary approach, some agencies require a 14 day waiting period between the last application and crop harvesting if the water contains greater than recommended levels of indicator organisms (BC-MAFF, 2003). This is assumed to permit populations of co-existent enteric pathogens to decrease to acceptable levels via die-off in the more challenging extra-intestinal environment.

Strict adherence to such guidelines presents significant challenges for many growers. In temperate climates, peak irrigation demand generally coincides with the time at which surface water systems are at their greatest risk for microbial contamination (Jamieson et al., 2003). To mitigate the effects of soil water deficits, producers of high value crops may be forced to use water containing indicator organisms exceeding recommended levels. For the same reasons, two weeks without water prior to harvesting is often too long for many crops during the warmest growing months. Furthermore, to assure market freshness, vegetable farmers may apply water 24 h prior to harvesting (Tyrrel et al., 2006). Water-borne bacterial pathogens introduced in this way to the edible portions of these crops may survive long enough to be ingested. The mode of water delivery may influence levels of plant contamination as well (Solomon et al., 2002). Overhead sprinkler or cannon irrigation systems deposit water on both the crop canopy and soil. Thus, there is high potential for phyllosphere contamination via direct inoculation or soil splash, particularly for crops that grow close to the ground as is the case with spinach. If the phyllosphere comprises the edible portions of the plant, this is of particular significance. Sprinkler or cannon irrigation may also result in the airborne transport of pathogens, especially during windy conditions, and thus pose a potential contamination risk to adjacent fields or water systems, or provide for direct human exposure (Hutchison et al., 2008).

The phyllosphere is a largely hostile environment for bacterial pathogens (Brandl, 2006). Factors that influence pathogen survival in the phyllosphere have received extensive consideration by Heaton and Jones (2008). Ultraviolet (UV) radiation and desiccation have been identified as important factors influencing pathogen survival on the phylloplane (Dreux et al., 2007a, Heaton & Jones, 2008). However, it is unclear whether these processes influence the persistence of bacterial pathogens introduced onto vegetable crops by spray irrigation. Water-borne pathogens can be exposed to UV radiation both during and after their deposition within the phyllosphere, and are likely subject to varying degrees of desiccation. It has been reported that pathogen migration into biofilms established by native micro-flora, reduces the efficacy of these natural inactivation (disinfection) processes (Elasri & Miller, 1999, Monier & Lindow, 2005). Field-based characterization of pathogen behaviour under these conditions is necessary in order to properly assess contamination risks and predict when crops are safe for harvest.

We report here on the fate of two non-pathogenic E. coli strains following their introduction via spray irrigation into the phyllosphere of spinach grown under commercial production conditions. Spinach was selected because it is a high risk crop and has been implicated in recent E. coli O157:H7 outbreaks (MMWR, 2006). It is also a water sensitive crop that may receive little or no processing prior to consumption. Specific objectives included determining: (i) the influence of UV radiation on the persistence of culturable forms of E. coli, and (ii) an appropriate timing for harvest relative to the last irrigation.

Section snippets

Test plants and treatment plots

Studies were conducted from July through September of 2007 and 2008 in two separate locations within the Annapolis Valley of Nova Scotia, Canada. Four trials, each with 8 days of sampling, were completed per year. Spinach, Spinacia oleracea, cultivar Unipak 151 was utilized in all 2007 trials and the first 2008 trial. Cultivar Menorca was used in the remaining three 2008 trials. These two cultivars are commonly planted in Nova Scotia and it was desirable to examine E. coli survival on both so

Suitability of test system

Spinach and soil samples collected from within each treatment plot prior to the application of seeded irrigation water were consistently negative, suggesting a low incidence (i.e., < detection limit), the lack of naturally occurring nalidixic acid resistant E. coli. Thus, E. coli recovered in environmental samples after irrigation were assumed to have originated from the inoculum. Irrigation water seeded to levels ranging from 4.90 to 7.14 log cfu/100 ml resulted in plot loading rates of 6.00 to

Discussion

The numbers of culturable, leaf-associated E. coli NAR and E. coli DM3n declined rapidly subsequent to their introduction via spray irrigation into the phyllosphere of field-grown spinach. Although comparable decay kinetics were observed in Trials 1 through 7, there was some inter-trial variability. This was likely due to differences in UV exposure, wind speed, air temperature and relative humidity (Dreux et al., 2007a, Stine et al., 2005). In Trial 8, plants were inoculated early in the

Acknowledgments

Funding for this project was provided by the Canada-Nova Scotia Water Supply Expansion Program, and the Canada Research Chairs Program. The contributions of the Nova Scotia Department of Agriculture, Agriculture Agri-Food Canada Research Branch, Melvin Farms Inc. and the Nova Scotia Federation of Agriculture are greatly appreciated. The authors gratefully acknowledge the assistance provided by the Government of Canada Federal Summer Work Experience Program participants, Greg Potter and Sara

References (33)

  • B.R. Yaun et al.

    Inhibition of pathogens on fresh produce by ultraviolet energy

    International Journal of Food Microbiology

    (2004)
  • J. Abu-Ashour et al.

    Movement of bacteria in unsaturated soil columns with macropores

    Transactions of the American Society of Agricultural Engineers

    (1998)
  • D. Aruscavage et al.

    Interactions affecting the proliferation and control of human pathogens on edible plants

    Journal of Food Science

    (2006)
  • BC-MAFF. 2003. Water Quality Factsheet: Treating Irrigation and Crop Wash Water for Pathogens, BC Ministry of...
  • A. Best et al.

    Role for flagella but not intimin in the persistent infection of the gastrointestinal tissues of specific-pathogen-free chicks by shiga toxin-negative Escherichia coli O157:H7

    Infection and Immunity

    (2005)
  • M.T. Brandl

    Fitness of human enteric pathogens on plants and implications for food safety

    Annual Review of Phytopathology

    (2006)
  • Cited by (55)

    • Impact of population density and stress adaptation on the internalization of Salmonella in leafy greens

      2022, Food Microbiology
      Citation Excerpt :

      The exposure of pathogens to sublethal stresses may result in the development of tolerance against a homologous stress of higher intensity (adaptation) or against different stress (cross-adaptation) (Yousef and Courtney, 2002). Upon contact with plant surfaces, bacteria may encounter various abiotic stresses such as solar irradiation (Allende et al., 2017; Wood et al., 2010), temperature, relative humidity (Brandl et al., 2004; Heaton and Jones, 2008), desiccation (Brandl and Mandrell, 2002; López-Gálvez et al., 2018) and starvation (Van der Linden et al., 2014), as well as biotic stresses caused by the indigenous microbiota (Lima et al., 2013). Many of these stresses may cause an initial instant decline in microbial population (Bezanson et al., 2012; McKellar et al., 2014).

    • Conditions at the time of inoculation influence survival of attenuated Escherichia coli O157:H7 on field-inoculated lettuce

      2020, Food Microbiology
      Citation Excerpt :

      In the current study, favorable conditions for bacterial survival were recorded at night, when humidity was high, temperatures were low, and dew or rain events provided free water on the plant surfaces. The negative effects of solar radiation on bacterial survival have been reported for E. coli inoculated on spinach, where the populations were always higher on spinach grown under shaded screening than on spinach grown unprotected (Wood et al., 2010), and for inoculated E. coli O157:H7 on the abaxial side of lettuce leaves (Erickson et al., 2010). In the Salinas Valley, the dry season typically occurs between May to October and rainfall is rare during this period.

    View all citing articles on Scopus
    1

    Current affiliation: School of Environmental Sciences, University of Guelph, Guelph, Ontario, Canada.

    View full text