Escherichia coli contamination and health aspects of soil and tomatoes (Solanum lycopersicum L.) subsurface drip irrigated with on-site treated domestic wastewater

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Escherichia coli contamination and health aspects of soil and tomatoes (Solanum lycopersicum L.) subsurface drip irrigated with on-site treated domestic wastewater A. Forslund a,*, J.H.J. Ensink b, B. Markussen c, A. Battilani d, G. Psarras e, S. Gola f, L. Sandei f, T. Fletcher g, A. Dalsgaard a a

Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Groennegaardsvej 15, DK-1870 Frederiksberg C, Denmark b Department of Disease Control, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom c Department of Mathematical Sciences, Faculty of Sciences, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen Ø, Denmark d Consorzio di Bonfica di secondo grado per il Canale Emilliano Romagnolo, Area Agronomicoeambientale, Via E. Masi 8, I-40137 Bologna, Italy e Laboratory of Plant Physiology and Mineral Nutrition, Institute for Olive Tree and Subtropical Plants of Chania, National Agricultural Research Foundation, Agrokipio, 73100 Chania, Crete, Greece f Department of Tomato and Vegetables, Stazione Sperimentale Industria Conserve Alimentari, Viale F. Tanara 31/a, 43100 Parma, Italy g Public and Environmental Health Research Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom

article info

abstract

Article history:

Faecal contamination of soil and tomatoes irrigated by sprinkler as well as surface and

Received 27 October 2011

subsurface drip irrigation with treated domestic wastewater were compared in 2007 and

Received in revised form

2008 at experimental sites in Crete and Italy. Wastewater was treated by Membrane Bio

6 March 2012

Reactor (MBR) technology, gravel filtration or UV-treatment before used for irrigation.

Accepted 9 August 2012

Irrigation water, soil and tomato samples were collected during two cropping seasons and

Available online 23 August 2012

enumerated for the faecal indicator bacterium Escherichia coli and helminth eggs. The study

Keywords:

per 100 ml and Crete 488 cfu per 100 ml) and low concentrations of E. coli in soil (mean: Italy

Escherichia coli

95 cfu g
1 and Crete 33 cfu g
1). Only two out of 84 tomato samples in Crete contained E. coli

Tomato

(mean: 2700 cfu g
1) while tomatoes from Italy were free of E. coli. No helminth eggs were

Treated wastewater

found in the irrigation water or on the tomatoes from Crete. Two tomato samples out of 36

found elevated levels of E. coli in irrigation water (mean: Italy 1753 cell forming unit (cfu)

Food safety

from Italy were contaminated by helminth eggs (mean: 0.18 eggs g
1) and had been irri-

Subsurface drip irrigation

gated with treated wastewater and tap water, respectively. Pulsed Field Gel Electrophoresis

Risk assessment

DNA fingerprints of E. coli collected during 2008 showed no identical pattern between water

PFGE typing

and soil isolates which indicates contribution from other environmental sources with E. coli, e.g. wildlife. A quantitative microbial risk assessment (QMRA) model with Monte Carlo simulations adopted by the World Health Organization (WHO) found the use of tap water and treated wastewater to be associated with risks that exceed permissible limits as proposed by the WHO (1.0  10
3 disease risk per person per year) for the accidental ingestion of irrigated soil by farmers (Crete: 0.67 pppy and Italy: 1.0 pppy). The QMRA found that the consumption of tomatoes in Italy was deemed to be safe while permissible limits

* Corresponding author. Tel.: þ45 3533 2762; fax: þ45 3533 2755. E-mail address: [email protected] (A. Forslund). 0043-1354/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2012.08.011

5918

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

were exceeded in Crete (1.0 pppy). Overall the quality of tomatoes was safe for human consumption since the disease risk found on Crete was based on only two contaminated tomato samples. It is a fundamental limitation of the WHO QMRA model that it is not based on actual pathogen numbers, but rather on numbers of E. coli converted to estimated pathogen numbers, since it is widely accepted that there is poor correlation between E. coli and viral and parasite pathogens. Our findings also stress the importance of the external environment, typically wildlife, as sources of faecal contamination. ª 2012 Elsevier Ltd. All rights reserved.

1.

Introduction

Since the beginning of the 20th century the demand for water by European agriculture has more than doubled as a result of rapid population growth and urbanization (Lavalle et al., 2009). Imbalances in water supply and availability are experienced by many European Union (EU) member countries around the Mediterranean, particularly in the summer months mainly as a result of low precipitation and peak demands for irrigation water by agriculture and the tourist industry (Angelakis et al., 1999). The use of urban wastewater in agriculture has often been propagated as a way to overcome water scarcity and to protect aquatic ecosystems from contamination. In several countries around the Mediterranean treated urban wastewater has been incorporated as a resource into integrated water resource management programmes, especially in Israel, where wastewater reuse has been practiced since 1955 (Camp et al., 2000; Haruvy et al., 1999). The European Water Framework Directive (2000/60/ EC), advocates a similar approach and specifies that treated wastewater should be used in agriculture where and whenever appropriate (EU, 2000). In addition, the growing water scarcity emphasizes the need for efficient use of water for crop irrigation (Hsiao et al., 2007). Subsurface drip irrigation utilizes less water than most other types of irrigation, mainly through reduced soil evaporation, but also as the water requirements of plants can be met more precisely (Ayars et al., 1999; Battilani et al., 2009a; Shahnazari et al., 2007). An additional advantage of subsurface drip irrigation is the reduced risk of crop contamination and reduced direct exposure to farm workers when wastewater is used for irrigation (Armon et al., 2002; Song et al., 2006). Urban wastewater can contain high numbers of faecal microorganisms including disease-causing pathogens like Salmonella, Campylobacter, Shigella, enteric viruses, protozoan parasites and helminth parasites (De´portes et al., 1995; Girones, 2006; Steele and Odumeru, 2004; USEPA, 2004). Human pathogens, organic and inorganic pollutants, found in urban wastewater are a matter of concern for both farmer health and food safety (FDA, 2001; Toze, 2006). To overcome public health concerns the World Health Organization (WHO) has developed guidelines for the safe use of wastewater in agriculture. The WHO guidelines are based on health targets and the assumption that no additional cases of disease should occur as a result of exposure to wastewater or wastewater irrigated produce (WHO, 2006). The guidelines use a Quantitative Microbial Risk Assessment (QMRA) model based on a permissible annual disease risk (1.0  10
3 disease risk per

person per year) which is used to calculate a required reduction in pathogen concentrations to achieve acceptable disease risks. The guidelines promote a multiple barrier approach and the required reduction in pathogen concentration is not expected to be met only through wastewater treatment, but will also depend on the type and processing of crop grown (crop consumed uncooked vs. crops consumed cooked and industrial crops), the level of human exposure when farmers work in the field (labour intensive vs. mechanized) and the level of exposure through different irrigation methods (basin vs. bed and furrow irrigation and sprinkler vs. surface or subsurface drip irrigation) (WHO, 2006). The demand for fresh produce by European consumers has increased dramatically over the last century as a result of changing food habits and rapid population growth (Brandl, 2006) while at the same time the incidence of foodborne outbreaks caused by contaminated fresh fruit and vegetables has increased (Harris et al., 2003; Sivapalasingam et al., 2004). The pathogens most frequently linked to fruit- and vegetablerelated disease outbreaks include bacteria (Salmonella, Escherichia coli), viruses (Norwalk-like, hepatitis A), and parasites (Cryptosporidium, Cyclospora) (Tauxe et al., 1997), with Salmonella and E. coli O157:H7 being the leading causes of producerelated outbreaks in the USA (Lynch et al., 2009). Tomatoes are the second most important vegetable in the world and the annual world production is approximately 130 million tons and the cropped area worldwide is approximately 29.9 million hectare per year (FAO, 2009). The crop water consumption at global scale can be roughly estimated to 1.8  1011 m3 per year, at least half of this amount supplied as irrigation. Tomato for processing represent approximately 31% of the global production (40 million tons), cultivated on a surface of about 6.0 million hectare per year (FAO, 2011). Tomatoes have been involved in several human disease outbreaks and the faecal contamination of tomatoes has in many cases occurred postharvest e.g. dirty wash water (Hanning et al., 2009), but pond water used for irrigation has also been identified as a source of contaminated tomatoes (Greene et al., 2008). The objective of this study was to evaluate if the use of subsurface drip irrigation with domestic wastewater was associated with reduced human health risk for farm workers and consumers eating raw vegetables. At study sites in Chania, Crete and Bologna, Italy, tomatoes were irrigated with treated domestic wastewater using conventional irrigation techniques (sprinkler and surface drip irrigation) and subsurface drip irrigation. Irrigation water, soil and tomato samples were collected during two cropping seasons in the period from May 2007 to September 2008 and analysed for the

5919

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

presence of helminth eggs and the bacterial indicator organisms E. coli. Impact on human health of the different irrigation practices were assessed using the QMRA model of the WHO by comparing calculated disease risks associated with consumption of tomatoes and farmers exposures to contaminated soil with permissible disease risks.

2.

Materials and methods

During the growing seasons of 2007 and 2008, two experimental sites were selected for the cultivation of fresh and processing tomatoes. Water, soil and tomato samples were collected and analysed for the presence of the faecal indicator organisms, E. coli and helminth eggs, during this period.

2.1.

Study sites

The first experimental field site was situated in the Po Valley, northern Italy (44 340 N, 11 320 E). This area is predominantly agricultural, and the field site was located on a commercial farm. A total of 18 plots (two irrigation application types, three water qualities and three replicates) were cultivated with processing tomatoes (Perfect Peel variety). Each plot comprised of 105 m2 and 350 tomato plants were grown per plot. The distance between the tomato rows was 1.5 m and plant distance within rows was 0.2 m. The soil was characterized as a silty-clay soil (24% sand, 41% silt, 35% clay) and the shallow groundwater table was located at 0.8 m depth. Each plot received the same type of irrigation water during the two seasons to allow for potential accumulation of contaminants. The plot was fertilized with ammonium, nitrate and phosphoric acid, and no manure had been applied for the previous six years. The second experimental site was located in Greece on the Island of Crete in the peri-urban areas of Chania (35 280 N, 24 20 E) and was cultivated with fresh tomatoes (‘Verdoun’ variety). Vegetable cultivation is common on the island throughout the year. Here, 12 plots (two irrigation application types, two water qualities and three replicates), each

comprising 50 m2 (5 m  10 m) were grown with a total of 100 tomato plants per plot. The distance between tomato plant rows was 1 m and plant distance within each row was 0.5 m. The soil was a sandy clay loam (55% sand, 23% silt, 22% clay). Ammonium nitrate, potassium nitrate, monoammonium phosphate and calcium nitrate were applied to the plots through fertigation. No manure had been applied to the plots during the previous five years before the field trials, or during the two experimental seasons (2007e2008).

2.2.

Water types and irrigation practices

2.2.1.

Italian field site

Three different types of water quality were used for irrigation: i) tap water, ii) primary treated wastewater (PTWW) and iii) secondary treated wastewater (STWW) (Table 1). Tap water was ground water that had been filtered and disinfected with chlorine dioxide and was provided through the municipal water supply system. Tap water was used for irrigation of control plots. PTWW and STWW were obtained from a small wastewater treatment plant serving a nearby village (population <2000 inhabitants). PTWW was raw wastewater where large particles had been removed by screen filtration. At the wastewater treatment plant STWW had been treated by mechanical filtration, oxidation and in a sedimentation pool but without disinfection. At the study site PTWW underwent further treatment by MBR (Membrane Bio Reactor) technology (Grundfos, Bjerringbro, Denmark), referred to as MBR-water. STWW was further treated through a gravel filter (Battilani et al., 2010a). MBR-water and gravel filtrated STWW were connected to the field site by two separate pipelines. Faucets were installed on the pipelines just before they entered the experimental field to collect irrigation water. At the Italian field site, regulated deficit irrigation (RDI) was used as a watersaving irrigation strategy to reduce the amount of irrigation water (Kirda et al., 2004). The RDI method irrigates the entire root zone with less water than the potential evapotranspiration while the water stress developed by the plant has minimal effects on the yield (English and Raja, 1996). Plots irrigated by the RDI method received 71 and 89% of fully

Table 1 e Type of irrigation water, treatment, irrigation strategy and method used at study sites in Italy and Crete. Location Italy

Crete

Irrigation water type

Water treatment

Irrigation strategy c

Tap water

None

RDI

MBR watera

RDIc

STWWb

Membrane Bio Reaction Gravel filtration

Tap water

None

Full

STWWb

Gravel filtration and UV-treatment

Full

a MBR water, Primary treated waste water (PTWW) treated by the MBR technology. b STWW, Secondary treated waste water. c RDI, Regulated Deficit Irrigation.

RDIc

Irrigation method Sprinkler Subsurface drip Sprinkler Subsurface drip Sprinkler Subsurface drip Surface drip Subsurface drip Surface drip Subsurface drip

5920

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

irrigated plots as calculated by the Fertirrigere model in 2007 and 2008, respectively (Jensen et al., 2010). Plots were irrigated by mini sprinkler (Netafim Ltd, Tel Aviv, Israel) with a discharge rate up to 200 l h
1 or by subsurface drip lines (Netafim Ltd, Tel Aviv, Israel) with a discharge rate of 1.6 l h
1. Subsurface drip lines were placed in each tomato row and buried at 10-cm depth. The distance between drip emitters was 40-cm and these were positioned exactly in the middle between two tomato plants. The Fertirrigere model (Battilani, 2006) was used to estimate the amount of irrigation water needed for optimal growth of the tomato plants. This model includes equations for crop development, crop production, evapotranspiration, water and nitrogen uptake, dynamics of both inorganic and organic nitrogen in soil, and the calculation of the daily balance of water and nitrogen content in soil. Crops received water approximately every third day based on crop development, precipitation and soil water dynamics (Fig. 1). Processing tomatoes received 251 mm of irrigation water in 2007. In 2008, tomatoes irrigated by sprinkler irrigation received 193 mm and subsurface drip irrigated tomatoes received 190 mm (Table 2).

2.2.2.

Crete field site

Two different types of water quality were used for irrigation: i) tap water and ii) secondary treated wastewater (STWW) (Table 1). Tap water was groundwater from a borehole but was not treated by chlorination. STWW was obtained from the Chania wastewater treatment plant and treated by filtration, sedimentation and aeration but no disinfection was applied. STWW was transported to the experimental site and stored in four tanks, each with a volume of 5 m3. STWW was treated on site through a gravel filter and by UV light disinfection before used for irrigation. The UV lamp power was 40 W, with an UVC output at 254 nm of 16 W and an UV dose of 400 J m
2. The approximate irradiation volume was 1.5 l. During the first five weeks after planting of the tomato plants, all plots were irrigated with tap water. After this period STWW was used to irrigate the assigned plots during the remaining season (Fig. 2). The control plot continued to receive tap water for irrigation. The irrigation strategy applied in Crete was full irrigation. Plots were irrigated by surface drip lines (Netafim Ltd, Tel Aviv, Israel) or by subsurface drip lines (Netafim Ltd, Tel Aviv, Israel.). Drip lines were placed along each tomato

40

30

35

25

Irrigation, precipitation and evapotranspiration (mm/day)

30 20

25 20

15

15

10

10 5

5 0

0

30

40 35

25

30 20

25 20

15

15

10

10 5

5 0

Sprinkler

0

SSDI

Precipitation

Evapotranspiration

Temperature

Date (yyyy-mm-dd) Fig. 1 e Precipitation, evapotranspiration and irrigation water amount and daily average temperature recorded during the growing seasons at the study site in Italy. Irrigation water was applied by sprinkler and subsurface drip irrigation (SSDI).

e e 440 470 355 402 440 470 355 405 e e 104 212 0 2 e e 440 468 251 190 440 468 251 193 e e 84 87 70 78 7-Sep 4-Sep 22-Aug 22-Aug 22-May to 14-Aug 23-May to 18-Aug 12-Jun to 21-Aug 4-Jun to 21-Aug 17-May 23-May 25-Apr 22-Apr Crete

2007 2008 2007 2008 Italy

Climate conditions

The weather in Italy and Crete is characterized by mild, rainy winters and hot, dry summers. The climate is a temperate Mediterranean climate. In Italy, rainfall during the cropping season was 179 mm in 2007 and 229 mm in 2008. A single rain event at the end of August 2007 constituted with 42 mm rain and in the middle of June 2008, 95 mm of rain was measured on one day (Fig. 1). The average temperature during cropping season 2007 and 2008 was 22.8  C and 23.5  C, respectively. The precipitation amount and average temperature in Crete in 2007 during the cropping season was 50.6 mm and 22.7  C. It was not raining during the irrigation period in 2007. In 2008, there were 34.8 mm rain and the average temperature was 21.9  C during the cropping season (Fig. 2). Only 2 mm rain was measured during the irrigation period in 2008 (Table 2).

a SSDI, Subsurface Drip Irrigation. b SDI, Surface Drip Irrigation.

SDIb SSDIa Sprinkler Precipitation SDIb SSDIa Sprinkler

Irrigation water volume by irrigation method during irrigation period (mm) Length of irrigation period (days) Harvest date Irrigation period Planting date (ddemm) Year

5921

row, with surface drip lines on top of soil surface and subsurface lines buried at a depth of 15 cm. The distance between drip emitters was 50 cm and the emitter was located next to each plant. Drip emitter was able to supply 1.6 l h
1. Crops received water every day in 2007 and every second day in 2008 (Fig. 2) based on calculated evapotranspiration, using crop coefficients available for local crops (Tsanis et al., 1997). The irrigation water amount was readjusted when needed, based on additional volumetric soil water data. In 2007, 2008, tomatoes received 440 mm and 468 mm of irrigation water, respectively (Table 2).

2.3.

Location

Table 2 e Time periods and amount of precipitation and irrigation water applied by different methods in Italy and Crete.

Total water load on fields during irrigation (mm)

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

2.4.

Sample collection

2.4.1.

Collection of water samples

Irrigation water samples were collected through faucets installed on pipelines just before they entered the experimental field site. For enumeration of E. coli, a composite sample consisting of three individual 1-l samples was collected over a 4-h period. Samples were collected in 1-l sterile glass bottles. For the helminth egg analysis, a 10-l composite sample was required. Three individual 3.5-l samples was collected over a 4-h period and kept in sterile plastic containers with lids. All samples were stored in a cooling box and transported to the local laboratory for further analysis on the day of sampling. Analysis of water samples for E. coli was initiated on the day of collection. Samples for helminth egg analysis were stored at 4e5  C until further processing. Water samples were collected on 10e21 occasions during the irrigation periods. During the irrigation period in 2007 and 2008, a total of 57 and 33 water samples were analysed at the Italian field site, respectively. At the Greek site, 42 and 20 water samples were analysed through the irrigation periods, respectively.

2.4.2.

Collection of soil samples

The sampling of soil was coordinated with irrigation events as soil samples were collected on the same day of irrigation or within one to three days after an irrigation application. Soil samples were collected with an auger (diameter 2 cm) within a 25-cm radius of a drip emitter and were taken randomly in each plot. The soil core was divided into an upper soil fraction

5922

22 20 18 16 14 12 10 8 6 4 2 0

35 30 25 20 15 10 5 0

30

22 20 18 16 14 12 10 8 6 4 2 0

25

Temperature (°C)

Irrigation, evapotranspiration and precipitation (mm/day)

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

20 15 10 5 0

Irrigation Tap water

Irrigation STWW

Evapotranspiration

Precipitation

Temperature

Date (yyyy-mm-dd) Fig. 2 e Daily average temperature, precipitation, evapotranspiration and irrigation water amount recorded during the growing seasons at the Crete study site. Irrigation water was tap water or secondary treated wastewater (STWW).

(0e20 cm) and a lower soil fraction (21e40 cm) which was analysed separately. Soil samples were collected in 1-l sterile plastic bags and soil was mixed well before analysis. Soil samples were collected before irrigation started, during the irrigation period and at the time of harvest. Before irrigation start, a composite sample consisting of three soil samples was collected from each plot. During the irrigation period a composite sample of eight soil samples was collected for each soil fraction. During the study periods in 2007 and 2008, a total of 114 and 234 soil samples were analysed at the Italian field site, respectively. At the start of the growing season in 2007 before irrigation, six soil samples were collected and analysed. During the irrigation period, 36 soil samples were analysed and 72 soil samples were analysed after the irrigation was terminated. In 2008, 18 soil samples were analysed in Italy before irrigation and 180 samples during the irrigation period where as 36 soil samples were analysed at tomato harvest. At the Greek site, a total of 28 and 76 soil samples were analysed in 2007 and 2008, respectively. Four soil samples were analysed prior to irrigation in both 2007 and 2008. In 2007, 24 soil samples were collected and analysed at harvest of the tomatoes. Soil samples were analysed during the irrigation period (48 soil samples) and at harvest time (24 samples) in 2008.

2.4.3.

Collection of tomato samples

Within each plot three tomato plants were randomly selected. One to six tomatoes were picked from each tomato plant for bacteriological analysis. Tomatoes from the lower part of the plants were picked with a sterile plastic bag. For the helminth egg analysis, a composite sample containing one tomato from each of the tomato plants was analysed for the presence of helminth eggs. In Italy, 18 tomato samples were analysed for the presence of E. coli and helminth eggs on the tomato surface in both 2007 and 2008. In Crete, 24 tomato samples were analysed for E. coli and 12 tomato samples for helminth eggs each year.

2.5.

Enumeration of E. coli and helminth eggs

Irrigation water and the surface of tomatoes were analysed for the presence as well as concentration of E. coli and helminth eggs. Soil was analysed for concentration of E. coli only. Analysis for E. coli was done on Brilliance E. coli/coliform Selective Agar (Oxoid, Hampshire, UK). Samples were analysed according to the pour plate method. Briefly, a water sample was serial diluted (1:10) in Maximum Recovery Diluent (Oxoid) and 1 ml of sample dilution was transferred to an

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

empty Petri-dish and mixed with 15e20 ml melted agar (45  1  C). Plates were incubated at 37  C for 24 h. Dark purple colonies were counted as E. coli. Ten gram of soil added 90 ml of distilled water were homogenized for 30 s in a blender (Stomacher, Seward, UK) and used for preparing 10-fold dilutions which were subsequently analysed as described for the water samples. The surface of the tomatoes were washed in 200 ml sterile Maximum Recovery Diluent (Oxoid) and the washing water were thereafter diluted and mixed with melted agar and incubated as described for the water samples. Irrigation water and surface of tomatoes were analysed for the occurrence of helminth eggs according to the modified Bailenger method (Ayres and Mara, 1996) with zinc sulphate solution replaced by saturated NaCl-glucose solution.

2.6.

Isolation and characterisation of E. coli

In 2008, E. coli colonies from treated wastewater and soil were obtained from Brilliance E. coli/coliform Selective Agar plates. The E. coli isolates were collected from wastewater samples taken both before and after treatment by MBR, gravel filtration and UV treatment. Approximately 5e10 dark purple colonies were selected from each of the E. coli-positive samples and each isolate (colony) was assigned an identity number. Isolates were confirmed as E. coli using the following phenotypic tests: citrate utilization, indole production, Methyl Red/ Voges Proskauer reaction, oxidase- and catalase activity. E. coli isolates from Italy (n ¼ 137) and Crete (n ¼ 27) were analysed by Pulse Field Gel Electrophoresis (PFGE) DNA genotyping for discrimination between isolates and to determine the level of similarity of PFGE fingerprints as this may indicate to what extent the E. coli found in soil originated from the applied irrigation water. PFGE was performed according to protocol proposed by the CDC PulseNet (Ribot et al., 2006). Briefly, DNA from cell suspensions was released in a solid SeaKem Gold agarose plug (Lonza, Basel, Switzerland) and digested with the restriction endonuclease enzyme XbaI (New England Biolabs, Ipswich, USA). The separation of fragments was done on CHEF-DR III (Bio-Rad, Richmond, CA, USA) with the following conditions: 6 V cm
1 at 14  C for 19 h at a field angle of 120 with switch times of 2.2e54.4 s. Lambda phage (48.5 kb) DNA (New England Biolabs) was used as a molecular weight standard and Salmonella enterica serotype Braenderup as an internal control as well as a DNA molecular weight marker. Gels were visualized under UV light (Gel Doc, Bio-Rad) following staining with 2 mg ml
1 of ethidium bromide. Gel images of the PFGE fingerprints were analysed with GelCompar II (Applied Maths, St-Martens-Latem, Belgium) using the Dice similarity coefficient and clustering by UPGMA. Band position tolerance and the optimization coefficient were both set to 2%. Inclusion of S. enterica serotype Braenderup in all the gels showed a band pattern reproducibility of 98%. Accordingly, PFGE fingerprints were considered as distinct types when the band patterns shared less than 98% similarity.

2.7.

Health risk assessment

Health risk were calculated using the QMRA model combined with Monte Carlo simulations set out in the WHO guidelines for the safe use of wastewater in agriculture (WHO, 2006). The

5923

WHO guidelines are based on health-based targets; a maximum additional burden of disease as a result of exposure to wastewater or the consumption of wastewater irrigated produce. This burden of disease is set at a maximum permissible diarrhoeal disease risk of 1  10
3 per person per year (pppy). This disease risk is calculated for the accidental ingestion of soil during agricultural activities or the consumption of wastewater irrigated produce. In order to determine this disease risk, three indicator pathogens are used; rotavirus, Campylobacter spp. and Cryptosporidium. However, as these pathogens are rarely measured in water, soil or produce, the model uses a fixed conversion rate based on E. coli concentrations found in soil, water or produce. In this article, the result of the QMRA calculations are only presented for rotavirus as this pathogen has the lowest infection dosage and therefore is the most conservative of the three indicator pathogens. Permissible disease risks are calculated using maximum concentrations found in soil or on tomatoes over a cropping season. According to the WHO guideline wastewater is considered safe for use when less than 1 helminth egg per litre are found (WHO, 2006). In this study, the health risk assessments associated with consumption of tomatoes were based on unrestricted irrigation whereas the health risks related to ingestion of soil by farmers was assessed using the exposure scenario restricted irrigation. Further, reduction of E. coli contamination by post-treatment measures, e.g. washing of produce, disinfection or cooking, was not included in the risk assessment.

2.8.

Statistical analysis

The prevalence of E. coli in the various water types, soil fractions and on tomatoes collected from the two study sites in Italy and in Crete during 2007 and 2008 were compared using Fischer’s exact test. This test was used to test for significance since many of the 2  2 tables had expected cell counts less than 5. A p-value less than 0.05 were considered significant. Due to many samples in which E. coli could not be detected, the statistical analysis of the quantitative measurements was done as described earlier by Forslund et al. (2011). Briefly, a logarithmic regression on the binary outcomes, i.e. nondetected vs. detected, was combined with a normal regression on the logarithm of the positive measurements. More specifically, the probability p that a measurement is above the detection limit was modelled with a linear model in log( p) and when the measurement was above the detection limit the actual measurement x was modelled with a normal regression in log(x). The data analysis was done in SAS, version 9.2 (SAS Institute Inc., Cary, USA).

3.

Results

3.1.

Irrigation water quality

In 2007, 2008 at the Italian study site, 21/57 (37%) and 12/33 (37%) of irrigation water samples contained E. coli, respectively (Table 3). E. coli was not detected in any tap water samples. Overall, significantly more E. coli was found in treated wastewater during 2007 compared to 2008 ( p < 0.0001) and in STWW

5924

Table 3 e E. coli in irrigation water, soil and on tomatoes collected at study sites in Italy and Crete. Sample type

Water type

Irrigation method

Time of irrigation

Italy No. of E. coli positive samples/Total no. of samples

Crete

Geometric mean concentration of E. coli (cfu ml
1 or cfu g
1) d [maximum value] 2007

Soil

Tap water STWWa MBR-water Tap water

Surface /Subsurface Surfaceb/Subsurface Surfaceb/Subsurface Surfaceb Subsurface

STWWa

Surfaceb Subsurface

MBR-water

Surfaceb Subsurface

Tomato

Tap water STWWa MBR-water

a b c d

Surfaceb/Subsurface Surfaceb Subsurface Surfaceb/Subsurface

During/Harvest During/Harvest During/Harvest Before During/Harvest Before During/Harvest Before During/Harvest Before During/Harvest Before During/Harvest Before During/Harvest Harvest Harvest Harvest Harvest

0/30 11/30 22/30 0/4 1/54 1/4 4/54 0/4 5/54 0/4 7/54 0/4 10/54 0/4 3/54 0/12 0/6 0/6 0/12

c

ND 103 [3000] 4.84[20] NDc 30 [30] NDc 26.6 [50] NDc 63.2 [1200] NDc 11.5 [20] NDc 338 [480,000] NDc 62.1 [600] NDc NDc NDc NDc

2008 c

ND 134 [1000] 1.41 [4] NDc NDc 80 [80] NDc NDc NDc NDc 480 [23,000] NDc NDc NDc NDc NDc NDc NDc NDc

STWW, Secondary treated wastewater which was used in Italy following treatment in a gravel filter and in Crete after UV-light treatment. In Italy, surface irrigation was done by sprinkler irrigation and in Crete by surface drip irrigation. ND, Not Detected. Geometric mean calculated on values above detection limit.

2/31 4/31 e 1/2 5/24 0/2 6/24 1/2 4/24 1/2 5/24 e e e e 0/24 1/12 1/12 e

Geometric mean concentration of E. coli (cfu ml
1 or cfu g
1) d [maximum value] 2007

2008

4.00 [8] 5.96 [79] e 360 [360] 6.79 [71] NDc 3.81 [7] 5 [5] 19.0 [52] 12 [12] 6.14 [21] e e e e NDc 13,157 [13,157] 554 [554] e

NDc NDc e NDc 20 [20] NDc 24.5 [60] NDc 130 [130] NDc 2793 [9400] e e e e NDc NDc NDc e

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

Water

b

No. of E. coli positive samples/Total no. of samples

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

compared to MBR-water ( p < 0.0001). Gravel filtrated STWW contained E. coli in 37% of the water samples with 5/19 (26%) of water samples found positive for E. coli in 2007 and 6/11 (55%) samples contained E. coli in 2008. Geometric mean concentration of E. coli in STWW was 103 E. coli per ml in 2007 and 134 E. coli per ml in 2008. E. coli was present in 16/19 (84%) and 6/11 (55%) of the MBR-treated water samples in 2007 and 2008, respectively. Concentrations of 4.84 and 1.41 E. coli per ml was found in MBR-treated water in 2007 and 2008, respectively (Table 3). Only in 2007 was a higher number of E. coli positive samples found in MBR-water compared to STWW ( p < 0.0001). Helminth eggs were not detected in any water samples during the two-year study period. In 2007 at the Crete study site, E. coli were detected in 6% of tap water samples and 13% of STWW samples with a geometric mean concentration of 4.0 cfu ml
1 and 5.96 cfu ml
1 ( p < 0.0001). In 2008, no E. coli were detected in tap water or STWW (Table 3). Helminth eggs were not found in any water samples during the two-year study period.

3.2.

Hygienic quality of soil

In Italy, soil samples taken in 2007 before wastewater irrigation was initiated did not contain E. coli while 1/18 of the soil samples collected before wastewater irrigation in 2008 contained E. coli (Table 3). This single E. coli-positive soil sample was collected from a plot that the previous year had been irrigated with tap water. In 2008, E. coli was detected in 2/216 (1%) of the soil samples collected during the wastewater irrigation period. Soil samples contained E. coli in 26% (28/108) of the samples in 2007 and this was significantly higher compared to the 2008 season ( p < 0.0001). Among the 28 soil samples positive for E. coli in 2007, 18% (5/28) had been irrigated with tap water, 36% (10/28) with gravel filtrated STWW and 46% (13/28) with MBR-water. Four of the E. coli-positive soil samples that received tap water had been irrigated via subsurface drip irrigation while the remaining soil sample was irrigated by sprinklers. In plots receiving STWW, E. coli was found in 5/10 (50%) samples from plots irrigated by subsurface drip irrigation and 5/10 (50%) soil samples irrigated with sprinklers. In plots applied MBR-water by sprinklers, 10/ 13 (77%) samples was positive for E. coli. E. coli was found in 19/ 28 (68%) of the samples taken from the upper soil fraction ( p < 0.0001) and of these 10/19 (53%) had received water through subsurface drip irrigation and 9/19 (47%) received water by sprinkler irrigation. In E. coli-positive soil samples from the lower soil fraction, 2/9 (22%) samples had been irrigated by subsurface drippers and 7/9 (78%) samples had been irrigated by sprinklers. There was not a significant difference in the E. coli contamination of soil between sprinkler irrigation and subsurface drip irrigation ( p ¼ 0.53). The two soil samples positive for E. coli in 2008 had been irrigated with STWW by subsurface drip irrigation. The concentration of E. coli was 2.3  104 cfu g
1 in soil collected from the upper soil fraction (S873, Fig. 3) and 10 cfu g
1 in the second sample collected from the lower soil fraction (S876, Fig. 3). In Crete, soil samples taken in 2007 before irrigation was initiated contained E. coli in 3/4 (75%) samples (Table 3). At tomato harvest in 2007, 14/24 (58%) soil samples contained E. coli and 8/14 (57%) of these had been irrigated with tap water;

5925

4/8 soil samples by surface irrigation, and 4/8 samples by subsurface drip irrigation. Equal number of soil samples contained E. coli in surface drip irrigated (3/6) and subsurface drip irrigated (3/6) plots that received STWW treated by UVlight. A total of 57% (8/14) of the E. coli-positive soil samples were found in the upper soil fraction. In 2008, no E. coli was detected in the spring before start of wastewater irrigation as well as during the wastewater irrigation period. At harvest, it was possible to detect E. coli in 6/ 24 (25%) of the soil samples, half of them (3/6) had been irrigated with tap water and half (3/6) with UV-treated STWW. About 30% (2/6) of the E. coli-positive soil samples had been irrigated through surface drip irrigation and E. coli was detected in the upper soil fraction (S2 and S23, Fig. 4). The corresponding lower soil fraction did not contain E. coli. In subsurface drip irrigated plots (4/6), both upper and lower soil fractions contained E. coli, with the highest concentration found in the upper soil fraction (S4 and S11, Fig. 4). E. coli was found significantly more in soil samples collected in 2007 compared with the 2008 season ( p < 0.0001) and more frequent and in higher concentration in the upper soil fraction ( p ¼ 0.0267). There was a larger probability of finding E. coli in soil at the Crete study site compared to Italy ( p < 0.0001) and for both studies, a higher concentration of E. coli in irrigation water was associated with a higher concentration of E. coli in soil ( p < 0.0001).

3.3.

Hygienic quality of tomatoes

In Italy, all tomato samples were negative for E. coli (Table 3). In 2007, helminth eggs were found on the surface of tomatoes from two samples with one tomato sample originating from a plot irrigated with tap water while the other sample was from a plot that had received gravel filtrated STWW. The helminth concentration was in both cases 0.18 eggs g
1. Both of these plots received water by subsurface drip irrigation. The genus of the helminth eggs found on the tomatoes was in both cases Strongyloides while the species of the helminth eggs were not determined. In 2008, there were not detected helminth eggs on the tomatoes. In 2007 at Crete, two tomato samples (8%) contained E. coli and both had been irrigated with UV-treated STWW (Table 3). One tomato sample had an E. coli concentration of 13,157 cfu g
1 (surface drip irrigated plot) and the corresponding concentration in the soil was 12 cfu g
1 (upper soil fraction) and 52 cfu g
1 (lower soil fraction). The concentration of E. coli on the second tomato sample was 554 cfu g
1 (subsurface drip irrigated plot) and E. coli was not detected in either upper or lower soil fractions. In 2008, no E. coli were detected on any tomatoes. No helminth eggs were detected on the surface of tomatoes during the two-year study period.

3.4.

Typing of E. coli isolates

All E. coli isolates from Crete and Italy were typable by the PFGE fingerprint method and showed a wide diversity of E. coli present in treated wastewater and soil.

5926

100

98

96

94

92

90

88

86

84

82

78

80

76

74

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

W903-1

2008-08-18

W903-3

2008-08-18

W903-4

2008-08-18

S147-1

2008-05-07

W486-8

2008-07-02

W 604-8

2008-07-09

W607-3

2008-07-09

W486-6

2008-07-02

W486-1

2008-07-02

W604-10

2008-07-09

W604-7

2008-07-09

W607-2

2008-07-09

W605

2008-07-09

W604-4

2008-07-09

S873-1

2008-07-30

STWW

S873-2

2008-07-30

STWW

S873-3

2008-07-30

STWW

S873-4

2008-07-30

STWW

S873-5

2008-07-30

STWW

W903-2

2008-08-18

W607-4

2008-07-09

W898-5

2008-08-07

W604-1

2008-07-09

S876

2008-07-30

W898-3

2008-08-07

W486-2

2008-07-02

W486-5

2008-07-02

W604-9

2008-07-09

S147-2

2008-05-07

Tap water

S147-3

2008-05-07

Tap water

S147-5

2008-05-07

Tap water

S147-4

2008-05-07

Tap water

W606

2008-07-09

W486-4

2008-07-02

W898-2

2008-08-07

W898-4

2008-08-07

W604-6

2008-07-09

W604-5

2008-07-09

W607-1

2008-07-09

S147-6

2008-05-07

Tap water

S147-7

2008-05-07

Tap water

W486-7

2008-07-02

Tap water

STWW

Fig. 3 e Dendrogram showing genotypic similarities of 42 E. coli isolates based on PFGE fingerprints from Italy. Thirteen of these E. coli were isolated from soil and the type of irrigation water used is specified in the column “water source”. SD is the similarity calculated by Dice correlation coefficient.

A total of 124 E. coli isolates from treated wastewater in Italy were typed by PFGE resulting in 84 unique fingerprints. A representative selection of the different E. coli PFGE fingerprints is illustrated in Fig. 3. Both different and identical PFGE fingerprints were observed within and between sampling dates. Fourteen isolates from July, 9th (W604eW607) resulted in fourteen different PFGE fingerprint while three out of four isolates were identical on August, 18th (W903). Further,

identical PFGE fingerprints were found at different sampling dates, e.g. E. coli isolates W486 and W898 as well as isolates W604 and W898. Thirteen E. coli isolates from soil resulted in six distinctive PFGE fingerprints and showed 77e94% similarity with isolates from the treated wastewater. E. coli isolates from soil samples taken from tap water irrigated plots in May (S147) showed heterogeneity and were not identical with E. coli isolates taken at the end of July (S873 and S876). Within the

5927

100

95

90

85

80

75

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

S11-1

2008-08-27

Tap water

S11-2

2008-08-27

Tap water

S11-3

2008-08-27

Tap water

S11-4

2008-08-27

Tap water

S11-5

2008-08-27

Tap water

S11-6

2008-08-27

Tap water

W1-4

2008-08-07

W1-1

2008-08-07

W1-6

2008-08-07

W1-2

2008-08-07

W1-3

2008-08-07

W3-1

2008-08-14

W1-5

2008-08-07

W5-1

2008-08-21

W5-2

2008-08-21

W5-3

2008-08-21

W5-4

2008-08-21

W3-2

2008-08-14

S4-2

2008-08-27

STWW

S4-3

2008-08-27

STWW

S4-4

2008-08-27

STWW

S4-5

2008-08-27

STWW

S4-6

2008-08-27

STWW

S4-1

2008-08-27

STWW

S2

2008-08-27

Tap water

S23-1

2008-08-27

STWW

S23-1

2008-08-27

STWW

Fig. 4 e Dendrogram showing genotypic similarities based on PFGE fingerprints from Italy of 12 and 15 E. coli isolates from STWW and soil, respectively. SD is the similarity calculated by Dice correlation coefficient.

July sampling all isolates from the upper soil fraction (S873) showed identical PFGE profiles while the isolate from the lower soil fraction had a different profile (S876). E. coli isolates from STWW in Crete showed heterogeneity as twelve E. coli isolates resulted in eight different PFGE fingerprints (Fig. 4). E. coli obtained during the last sampling time of irrigation water showed identical PFGE fingerprints (W5). None of the PFGE fingerprint from STWW was identical with E. coli fingerprints found in soil samples. Fifteen E. coli isolates from soil resulted in four distinctive PFGE fingerprints and showed less than 91% similarity with isolates from the UV-treated wastewater. Identical PFGE fingerprint was found in soil irrigated with tap water (S2) or STWW (S4) (Fig. 4). A large variability was found in E. coli isolated from irrigation water and soil at both field sites. No similar PFGE fingerprints of E. coli were detected between irrigation water and soil indicating a possible external source of E. coli contamination.

3.5.

Health risk assessment

The results of the health risk assessment using rotavirus as a model organism showed that all the different irrigation

scenarios in Italy during 2007 were to be considered unsafe for farmers occupationally exposed to contaminated soil as the human disease risks exited the WHO guidelines of 1  10
3 per person per year (Table 4). A very high E. coli concentration found on one occasion in the lower soil fraction of sprinkler irrigated soil in Italy (480,000 cfu g
1) in 2007 translated into very high disease risk (1.0). During 2008, only subsurface drip irrigation with STWW during the irrigation period was found to be unsafe which was due to a high E. coli concentration (23,000 cfu g
1) found in one soil sample from the upper soil fraction and resulted in a high disease risk (0.94). In Crete, all irrigation scenarios with the exception of one posed an unacceptable health risk according to the WHO guidelines for farmers occupationally exposed to contaminated soil. Both in Italy and Crete several irrigation scenarios using tap water, even before irrigation had started resulted in unacceptable health risk for farmers exposed to soil. In Italy, all irrigated tomatoes were found safe for consumption. In Crete, tomatoes collected in 2007 and irrigated by STWW treated with UV were found to be unsafe for consumption as they had exceeded the guideline disease risk set by the WHO (2006).

5928

Harvest

Harvest 2008

Tomato

2007

Surfacea Subsurface Surfacea Subsurface Surfacea Subsurface Before During/Harvest 2008

Surfacea Subsurface Before During/Harvest 2007 Soil

a Surface irrigation, In Italy by sprinkler irrigation, in Crete by surface drip irrigation. b STWW, In Italy in combination with gravel filter, in Crete with additional UV treatment. Values in bold exceed the WHO guidelines of <1  10
3 pppy.

1.5 3 10L3e1.0  10
8 2.6 3 10L3e2.0  10
8 8.3 3 10L3e6.0  10
8 e 0.67e1.9  10
5 0.01e1.0  10
8 1.00e1.00 0.76e0.48 e e 0.04e3.8  10
7 9.0  10
4e2.0  10
8 8.3 3 10L3e4.0  10
8 e 7.2  10L3e1.0  10
8 2.2  10L3e1.0  10
8 e e e e e 1.0e6.1  10
6 0.06e1.0  10
6 e e e e e e e e 5.8  10L3e1.8  10
5 8.6  10L3e4.3  10
5 9.8 3 10L3e1.0  10
8 e e e e e e

STWW MBR-water Tap water

Italy

e 0.14e2.8  10
6 2.4  10L3e3.0  10
8 e e 0.94e4.1  10
5 e e e e

STWWb Tap water

Crete b

Disease risk (rotavirus) (pppy) Irrigation method Time of sampling Year Type of sample

Table 4 e Minimum and maximum disease risk (rotavirus) for farmers and crop handlers as a result of exposure to irrigated soils and the consumption of tomatoes from Italy and Crete.

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

4.

Discussion

4.1.

Irrigation methods and faecal contamination risks

In the EU, the abstraction of water for agricultural purposes, mainly irrigation, typically accounts for about 24% of the total water abstracted with up to 80% being used in agriculture in certain locations in southern Europe. In water scare areas, treated wastewater represents an alternative source of irrigation water, e.g. in Gran Canarias, Spain where 5000 ha of tomatoes are irrigated with treated wastewater (EEA, 2009). Contamination of vegetables with pathogens originating from wastewater can occur when overhead irrigation methods such as sprinkler irrigation is used because the edible parts are directly exposed to the applied water. Surface drip irrigation applies water at the soil surface and is less likely to contaminate high growing crops while subsurface drip irrigation applies the water direct to the roots with minimal transfer of pathogens to the crop surface (Pescod, 1992). In the present study, it was not possible to detect any reduced faecal contamination associated with subsurface drip irrigation, as only very few tomato samples were contaminated with E. coli. Faecal contamination levels of drupes sampled from the ground in a drip irrigated olive grove applying wastewater were usually similar or lower than drupes collected from rainfed plots (Palese et al., 2009) suggesting other environmental sources of faecal contamination than the wastewater. Cucumbers have also been found to be more faecal contaminated as a result of direct contact on the soil surface as compared to tomatoes that was grown using stakes (El Hamouri et al., 1996). In general it is expected that fruit and vegetable contamination will occur when surface irrigation is used, however, in Pakistan low levels of faecal contamination was found on vegetables irrigated through basin irrigation despite high concentrations of E. coli in the irrigation water (Ensink et al., 2007). Comparison of spray or surface irrigated lettuce grown in pots with irrigation water inoculated with E. coli O157:H7 demonstrated significantly higher contamination of lettuce by spray irrigation (Solomon et al., 2002). Studies have shown that wastewater irrigated vegetables grown above the soil surface contained higher levels of faecal contamination when surface irrigated vegetables were compared with those irrigated subsurface (Assadian et al., 2005; Sadovski et al., 1978). Further, no faecal contamination was found on spinach leafs irrigated with wastewater by subsurface drip irrigation but soil texture was found to have an effect on the vertical movement patterns of viruses added to the irrigation water (Assadian et al., 2005). Furrow irrigation resulted in generally greater microbial contamination of lettuce, bell pepper and the soil surface than did subsurface drip irrigation with the exception of cantaloupe where preferential water paths from the emitter to the soil surface occurred due to loosened soil and was associated with higher faecal contamination (Song et al., 2006). In a field study comparing sprinkler, furrow and subsurface drip irrigation of lettuce, it was found, that sprinkler irrigation resulted in substantial higher concentrations of E. coli while almost all lettuce samples from subsurface drip irrigation were free of E. coli. The irrigation water was inoculated with E. coli at

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

a concentration of 108e109 cells ml
1 (Fonseca et al., 2011). Combined these previous studies show that subsurface irrigation with low quality water are associated with less faecal contamination of crops compared with different types of surface irrigation. Irrigation of tomatoes continued until harvest due to the hot climate at the Crete field site while irrigation was terminated three weeks before harvest at the Italian field site. The period with no irrigation could have reduced the concentration of E. coli on tomatoes in Italy to below the detection level. The time interval between the last irrigation event and harvest determines the extent of contamination, as pathogens have been shown to decline with time following cessation of irrigation before harvest. It is recommended that irrigation water for tomatoes is of an appropriate microbial quality particularly when applied close to the time of harvest or during harvesting to avoid contamination of the tomatoes (FDA, 2009). Tomato and other crop growers should maximize the time between the last irrigation event and harvest to minimize food safety risks when irrigating with water of low quality is unavoidable (Barker-Reid et al., 2009), e.g. termination of overhead irrigation seven days before crop harvest has been recommended (Rangarajan et al., 2000). Based on the results from the two-year field experiments in Italy and on Crete, we are not able to either support or refute that subsurface drip irrigation as used in the present study is safer than surface irrigation as almost all tomato samples were free of E. coli and no significant difference in the level of soil contamination between irrigation methods were found.

4.2.

Faecal contamination of soil

The contamination of vegetables grown in soil irrigated with faecal contaminated water will largely depend on the survival capabilities of the pathogens in the soil and on plants. A common maximum survival time reported for bacterial pathogens in soil is two months (Gerba and Smith, 2005), but persistence of bacterial pathogens in soil may be up to five months due to moist conditions and protection against sunlight (Islam et al., 2004). A considerable longer survival time, e.g. up to two years, has been reported for helminth eggs in soil (Gerba and Smith, 2005). Soil samples in the present study were collected on the same day or within a few days after irrigation, but still E. coli were only detected in 2/216 soil samples analysed during 2008 in Italy. One of these soil samples had a concentration of 2.3  104 E. coli g
1 but irrigation water applied on that sampling day as well as the week before contained 600e1000 E. coli ml
1. Water treated with the MBR technology showed a maximum E. coli concentration of 20 cfu ml
1 but despite this low level of faecal contamination, soil samples from such irrigated plots contained up to 4.8  105 E. coli g
1 in 2007. It should also be noted that E. coli was mainly found in the upper soil fraction both during the irrigation period and at harvest time in control plots irrigated with tap water. This indicates that although the treated wastewater could have been a source of contamination, the very high concentration of E. coli found in a few soil samples suggests that an external environmental source of faecal contamination would be more likely. This is supported by our findings in Crete where soil samples from both tap water and

5929

STWW irrigated plots contained E. coli at harvest time even though no E. coli were detected in the irrigation water during the entire 2008. Falk (1949) could not detect a difference in levels of faecal coliform contamination on the surface of tomatoes grown in sewage polluted and unpolluted soils and protection of the tomatoes by waterproof paper between the plants and the soil did not lead to a lower faecal contamination of the tomatoes. This indicates that other contamination routes, e.g. dust or insect, are at least as important sources of faecal contamination as direct splashing of soil and fruits during rain events (Falk, 1949). Birds, insects, wild and domestic animals have been reported as sources of faecal contamination of the environment and fresh produce (Beuchat, 1996; Ishii and Sadowsky, 2008; Talley et al., 2009). Wild birds may harbour many pathogens including E. coli O157:H7 (Huba´lek, 2004). A study using library-dependent and library-independent microbial source tracking of E. coli showed birds and wastewater to be responsible for the faecal contamination of a beach (Edge et al., 2010). During the two seasons at the Italian field site large bird populations were observed resting on the electric cables crossing the field and droppings from these birds could have been a source of faecal contamination and explain the peak concentration levels of E. coli in soil (Table 3). In addition, hare faeces were frequently observed close to the tomato plants at the Italian field site and such wild hares have been shown to carry Campylobacter species, an important pathogen causing gastrointestinal infections in humans (Wahlstro¨m et al., 2003). Zoonotic helminths have been reported in cats, foxes and dogs (Di Cerbo et al., 2008; Habluetzel et al., 2003) and helminth eggs found in soil may therefore originate from such animals and not only from faecal contaminated irrigation water. In addition to obtaining knowledge on food safety and health aspects associated with use of wastewater for crop irrigation, there is also a need to conduct studies that determine the impact of such irrigation practices on all the soilplant system components, including the irrigation system performance due to low water quality use, e.g. clogging of drip emitters, and changes in hydraulic soil properties (Aiello et al., 2007; Battilani et al., 2009b, 2010a, 2010b).

4.3.

Faecal contamination of tomatoes

Various pathogens have been recovered from vegetables (Beuchat, 1996) and the number of documented disease outbreaks associated with the consumption of contaminated raw vegetables has increased in recent years (Sivapalasingam et al., 2004). On plant surfaces, drying and UV-light are main factors causing bacterial pathogens and helminth eggs to be inactivated typically within one month of exposure (Gerba and Smith, 2005). In this study, E. coli was found on the surfaces of only two tomato samples irrigated with UV-treated STWW at the Crete field site out of the total 84 tomato samples analysed in the entire study period. One of the contaminated tomato samples was irrigated by surface drip application and the other by subsurface drip irrigation. Generally, no or low levels of contamination of tomatoes with E. coli and pathogens have been reported at farm level, in produce during packing and distribution as well as retail sale (Mukherjee et al., 2006; Rushing et al., 1996; Sagoo et al., 2001),

5930

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

but several disease outbreaks involving tomatoes have been reported (CDC, 2007; Hedberg et al., 1999). Levels of faecal contamination of vegetables depend on the growing conditions as well as the exposure and contact with soil, manure or irrigation water. It was not possible to detect Salmonella on tomato fruits when the plants had been grown in pots irrigated on the soil surface every second day for five weeks with water containing 105 Salmonella per ml (Jablasone et al., 2004). Significantly elevated numbers of indicator organisms were found on different vegetables incl. tomatoes sprinkler irrigated with highly faecal polluted wastewater effluent compared to slightly polluted effluent (Armon et al., 1994) and pathogen contamination of crops by splashing of soil particles during heavy rain (Heaton and Jones, 2008) and via crop debris (Barak and Liang, 2008) has been reported as well. In addition, Salmonella has been shown to survive and grow on the surface of tomatoes and also taken up internally through stem scar where it was isolated inside the tomatoes (Zhuang et al., 1995). In the present study, only the surfaces of the tomatoes were analysed so information on possible internalisation of E. coli is not available. The presence and survival of pathogens, e.g. Salmonella on vegetables can in addition be dependent on the serovar (Shi et al., 2007) and the cultivar (Barak et al., 2011). Further, the ability of different bacterial pathogens to adhere to the surface of vegetables are likely to differ significantly due to variations in bacterial surface characteristics, e.g. presence of capsule and flagella (Barak et al., 2002) as well as the pathogens ability to produce biofilm including cellulose (Lapidot and Yaron, 2009; Shaw et al., 2011). We found helminth eggs on the surface of two tomato samples from Italy, where one sample originated from a plot irrigated with tap water and the other sample came from a plot irrigated with gravel filtrated STWW. Kozan et al. (2005) found no helminth eggs on 15 samples of unwashed tomatoes collected from wholesalers in Ankara, Turkey where as other vegetables, e.g. parsley contained eggs of Taenia spp. and Toxocara spp. The genus of the helminth eggs found on the tomatoes at the Italian field site was in both cases identified as Strongyloides but the species could not be determined. There have been reported more than 50 species of Strongyloides originating from a variety of hosts, e.g. domestic fowls and cats, and with the species Strongyloides stercoralis being implicated in most infections of humans and dogs (Speare, 1989). S. stercoralis seems to be endemic among people in the regions of the Po Valley, Italy (Abrescia et al., 2009), but Strongyloides spp. could also be present in wild animals or pets and through their indiscriminate defaecation contaminate the soil and tomatoes with parasite eggs. Although tomatoes may appear as low risk crops for faecal contamination due to their smooth surface, the stated reports on human disease outbreaks associated with consumption of tomatoes underlines the need for further investigations, e.g. bacteriological studies of pathogen adherence to the surface of tomatoes and pathway(s) of possible internalisation.

4.4.

Origin of E. coli contamination

PFGE is considered the “gold standard” of genetic fingerprinting methods due to its high discriminatory power and reproducibility (Stoeckel et al., 2004). PFGE is therefore

frequently used in epidemiological investigations of disease outbreaks to establish relatedness and similarities between isolates obtained from diseased individuals and isolates recovered from suspected foods or other sources suspected implicated in the outbreaks (Ackers et al., 1998; Avery et al., 2002). E. coli is prevalent in the intestine and faeces of warmblooded mammals, including wildlife, livestock and humans, and genetic diverse E. coli from the different hosts can therefore be expected present in faecal contaminated wastewater (Caugant et al., 1981; McLellan et al., 2010). E. coli diversity in faeces from a single human host over an 11-month period was 10% based on protein electrophoresis with most of the E. coli types appearing only once or within a few days (Caugant et al., 1981) and an E. coli diversity of 13% within a one-day experiment established by ribotyping (Anderson et al., 2006). Genotype analysis of E. coli isolated from children and chickens living in close contact by PFGE showed distinct diversity between the two sources (Kariuki et al., 1999). PFGE fingerprints of E. coli isolates from the wastewater in Italy and Crete exhibited considerable genetic diversity as 68% of the PFGE fingerprints shown by E. coli from water samples were unique. Such genetic heterogeneity would also be expected as the wastewater treatment plants received faecal contaminated water from a large human population over a three-month period. Interestingly PFGE fingerprints of E. coli isolated from soil showed less variability (36% unique fingerprints) and shared no identical fingerprints with E. coli from water samples. A similar variation in the E. coli PFGE fingerprints were detected in a study on genetic diversity of E. coli isolated in irrigation canals and sediments from the Rio Grande River but with similar PFGE fingerprints of E. coli isolated from water and sediment samples collected at the same sampling time (Lu et al., 2004). There may be several explanations for not finding identical PFGE fingerprints in irrigation water and soil. Even though irrigation water samples were collected frequently during a three-month period and several E. coli colonies were isolated within each sampling, there is a probability that we did not isolate the distinct E. coli genotypes found in the soil samples. However, in a parallel investigation of irrigated potatoes at the same study site in Italy it was possible to isolate E. coli with identical PFGE fingerprints in soil and wastewater samples (unpublished results). Due to the very high discriminatory power of PFGE and high genetic diversity among E. coli, it would be necessary to genotype a large number of isolates in order to determine the source of E. coli contamination in large watersheds where multiple contamination sources exists (Lu et al., 2004). Ibenyassine et al. (2006) were able to identify the same repetitive DNA sequences pattern by ERIC-PCR typing of E. coli from vegetables, incl. tomatoes, and soils irrigated with treated domestic wastewater; however this genotyping method is less discriminatory compared to PFGE (Casarez et al., 2007). Using a typing method like PFGE with a high discriminatory power can also be problematic in tracking the source of the faecal contamination as highlighted by the findings of Gordon et al. (2002) who reported a significant degree of genetic diversity among E. coli isolated from the gastrointestinal tract and a septic tank used by people for defecation. This diversity in genotypes could be due to genetic changes in E. coli occurring as a response to external

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

5931

environmental stress, e.g. reduced availability of nutrients, altered physicalechemical conditions or natural selection processes. Further, recent studies have reported that E. coli potentially can replicate in soil and eventually become “naturalised” in the environment (Ishii and Sadowsky, 2008). Gagliardi and Karns (2000) reported that E. coli O157:H7 was able to replicate in various soil types under field conditions in Maryland, USA and faecal bacteria have been shown to multiply in moist subtropical and tropical soil environments (Fujioka et al., 1999). In the present study, unique E. coli PFGE fingerprints were detected in soil even though no E. coli was found in the UV-treated wastewater in Crete in 2008. This could indicate that E. coli populations had established in the soil environment following earlier faecal contamination events like irrigation with treated wastewater or wild animals.

1987) and for Cryptosporidium on data from a waste stabilization pond in Kenya (Grimason et al., 1993). Thus, it is clear that the WHO QMRA model is very much in progress, and that additional quantitative data on E. coli e pathogen conversion rates, or even better on actual pathogens, is needed. Future guidelines in European countries on use of low quality water for irrigation should be based on the health based target approach taken by the current WHO guidelines rather than the guidelines of the United States Environmental Protection Agency (USEPA) that advocates a no risk approach, which means that water used for irrigation purposes should effectively be free of pathogens and that all wastewater should undergo tertiary treatment processes before it is allowed to be used in agriculture (USEPA, 2004).

4.5.

5.

Risk assessment

The results of the risk assessment indicated that in both Italy and Crete the different types of irrigation water used for the cultivation of tomatoes were all associated with soil samples containing E. coli concentrations that resulted in unacceptable health risks to farmers ingesting such soil. In Crete, two tomato samples contained E. coli in numbers leading to exceeding permissible health risks when consumed. As discussed in Section 4.4, there was not found any identical PFGE DNA fingerprints among E. coli isolated in irrigation water and soil samples. It is therefore unclear if the estimated unacceptable health risks occurred as a result of faecal contamination in the irrigation water used or because of the soil being contaminated from external sources, e.g. wildlife. It should here be noted that even soil irrigated with non-faecally polluted tap water yielded E. coli. Thus, it is likely that the E. coli contamination of soil and tomatoes originated primarily from the external environment rather than the irrigation water. In comparison, Bastos et al. (2008) found also using the WHO QMRA model that consumption of green pepper and salad crops manually spray-irrigated with treated wastewater (mean 12 E. coli per ml) contained 8.3e280 E. coli per gram to be safe even though the irrigation water contained higher E. coli concentrations compared to the E. coli concentrations used to irrigate tomatoes in the present study. It should be noted that the E. coli concentrations in the different water types, including the treated wastewater, used for irrigation in our study typically was similar or lower than E. coli concentrations found in streams and rivers commonly used as sources of irrigation water (An et al., 2007; Fujioka et al., 1999). The proposed maximum permissible diarrhoeal disease risk in the WHO guidelines (1.0  10
3 disease risk per person per year) has been challenged as it is considered too strict (Mara, 2011), something our results would support. It is a fundamental limitation of the WHO QMRA model that it is not based on actual pathogen numbers, but rather on numbers of E. coli converted to estimated pathogen numbers, since it is widely accepted that there is poor correlation between E. coli and viral and parasite pathogens. Further, the conversion ratios between E. coli and pathogens in the WHO QMRA model are based on inadequate documentation, i.e. the fixed ratio of rotavirus and Campylobacter are based on data from a waste stabilization pond in Brazil (Mara et al., 2007; Oragui et al.,

Conclusion

This study showed that there was a weak association between the E. coli concentrations found in water used for irrigation and those E. coli found in irrigated soil but no association with E. coli on tomatoes. Since only 2/84 tomato samples contained E. coli despite being irrigated by different irrigation methods and water qualities consumption of such tomatoes does not seem to be associated with unacceptable health risks. In addition, PFGE genotypes of E. coli isolated from irrigation water and soil showed high diversity with no identical fingerprints which suggest contribution of faecal contamination from environmental sources, e.g. wildlife. Several of the tested irrigation scenarios were found to be unsafe based on WHO guidelines, however, questions should be raised about the validity of the current guidelines as there are clearly needs for further documentation on ratios between E. coli and pathogen concentrations to improve the guidelines which should also be able to account for environmental sources of faecal contamination that do not originate from the irrigation water.

Acknowledgement We acknowledge the help of Mita Sengupta and Heidi Huus Petersen at the Faculty of Life Science, University of Copenhagen for the helminth egg analysis of water and tomato samples. The excellent technical assistance of Nina Flindt and Gitte Petersen is highly appreciated. This study was supported by EU Commission through the project “Safe and high quality food production using low quality waters and improved irrigation systems and management” (SAFIR, EU, FOOD-CT-2005023168) (www.safir4eu.org), PathOrganic (http://pathorganic. coreportal.org) and Emila Romagna Government (Italy).

references

Abrescia, F.F., Falda, A., Caramaschi, G., Scalzini, A., Gobbi, F., Angheben, A., Gobbo, M., Schiavon, R., Rovere, P., Bisoffi, Z., 2009. Reemergence of strongyloidiasis, northern Italy. Emerging Infectious Diseases 15 (9), 1531e1533.

5932

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

Ackers, M.L., Mahon, B.E., Leahy, E., Goode, B., Damrow, T., Hayes, P.S., Bibb, W.F., Rice, D.H., Barrett, T.J., Hutwagner, L., Griffin, P.M., Slutsker, L., 1998. An outbreak of Escherichia coli O157:H7 infections associated with leaf lettuce consumption. The Journal of Infectious Diseases 177 (6), 1588e1593. Aiello, R., Cirelli, G.L., Consoli, S., 2007. Effects of reclaimed wastewater irrigation on soil and tomato fruits: a case study in Sicily (Italy). Agricultural Water Management 93 (1e2), 65e72. An, Y.-J., Yoon, C.G., Jung, K.W., Ham, J.H., 2007. Estimating the microbial risk of E. coli in reclaimed wastewater irrigation on paddy field. Environmental Monitoring and Assessment 129 (1e3), 53e60. Anderson, M.A., Whitlock, J.E., Harwood, V.J., 2006. Diversity and distribution of Escherichia coli genotypes and antibiotic resistance phenotypes in feces of humans, cattle, and horses. Applied and Environmental Microbiology 72 (11), 6914e6922. Angelakis, A.N., Marecos Do Monte, M.H.F., Bontoux, L., Asano, T., 1999. The status of wastewater reuse practice in the Mediterranean basin: need for guidelines. Water Research 33 (10), 2201e2217. Armon, R., Dosoretz, C.G., Azov, Y., Shelef, G., 1994. Residual contamination of crops irrigated with effluent of different qualities: a field study. Water Science and Technology 30 (9), 239e248. Armon, R., Gold, D., Brodsky, M., Oron, G., 2002. Surface and subsurface irrigation with effluent of different qualities and the presence of Cryptosporidium oocysts in soil and on crops. Water Science and Technology 46 (3), 115e122. Assadian, N.W., Di Giovanni, G.D., Enciso, J., Iglesias, J., Lindemann, W., 2005. The transport of waterborne solutes and bacteriophage in soil subirrigated with a wastewater blend. agriculture. Ecosystems and Environment 111 (1e4), 279e291. Avery, S.M., Liebana, E., Reid, A., Woodward, M.J., Buncic, S., 2002. Combined use of two genetic fingerprinting methods, pulsedfield gel electrophoresis and ribotyping, for characterization of Escherichia coli O157 isolates from food animals, retail meats, and cases of human disease. Journal of Clinical Microbiology 40 (8), 2806e2812. Ayars, J.E., Phene, C.J., Hutmacher, R.B., Davis, K.R., Schoneman, R.A., Vail, S.S., Mead, R.M., 1999. Subsurface drip irrigation of row crops: a review of 15 years of research at the Water Management Research Laboratory. Agricultural Water Management 42 (1), 1e27. Ayres, R.M., Mara, D.D., 1996. Analysis of Wwastewater for Use in Agriculture e a Laboratory Manual of Parasitological and Bacteriological Techniques. WHO, Geneva, Switzerland. Barak, J.D., Liang, A.S., 2008. Role of soil, crop debris, and a plant pathogen in Salmonella enterica contamination of tomato plants. PLoS One 3 (2), 1e5. Barak, J.D., Whitehand, L.C., Charkowski, A.O., 2002. Difference in attachment of Salmonella enterica serovars and Escherichia coli O157:H7 to alfalfa sprouts. Applied and Environmental Microbiology 68 (10), 4758e4763. Barak, J.D., Kramer, L.C., Hao, L., 2011. Colonization of tomato plants by Salmonella is cultivar dependent, and type 1 trichomes are preferred colonization sites. Applied and Environmental Microbiology 77 (2), 498e504. Barker-Reid, F., Harapas, D., Engleitner, S., Kreidl, S., Holmes, R., Faggian, R., 2009. Persistence of Escherichia coli on injured iceberg lettuce in the field overhead irrigated with contaminated water. Journal of Food Protection 72 (3), 458e464. Bastos, R.K.X., Bevilacqua, P.D., Silva, C.A.B., Silva, C.V., 2008. Wastewater irrigation of salad crops: further evidence for evaluation of the WHO guidelines. Water Science and Technology 57 (8), 1213e1219. Battilani, A., Plauborg, F.L., Andersen, M.N., Jensen, C.R., Sandei, L., Solimando, D., 2009a. Water saving irrigation

strategies for processing tomato. Acta Horticulture 823 (1), 69e76. Battilani, A., Steiner, M., Andersen, M.N., Schweitzer, A., Lorenzen, J., Bak, S.N., Plauborg, F., 2009b. User Manual for Decentralised Functional Water Production for Irrigation Purposes. In: EU Project SAFIR Report. Available at: www. safir4eu.org (accessed 25.02.12.). Battilani, A., Steiner, M., Andersen, M., Back, S.N., Lorenzen, J., Schweitzer, A., Dalsgaard, A., Forslund, A., Gola, S., Klopmann, W., Plauborg, F., Andersen, M.N., 2010a. Decentralised water and wastewater treatment technologies to produce functional water for irrigation. Agricultural Water Management 98 (3), 385e402. Battilani, A., Bertaki, M., Blagojevic, S., Cary, L., Chartzoulakis, K., Dalsgaard, A., Forslund, A., Kloppmann, W., Li, G., Picot, G., Psarras, G., Qi, X., Sandei, L., Solimando, D., Stikic, R., Toenner-Klank, L., 2010b. Progress Report on the Behaviour of Toxic Elements and Pathogens in the Irrigation Water-soilsubsoil-crop-food System under Irrigation with Low Quality Water. In: EU project SAFIR Report. Available at: www. safir4eu.org (accessed 25.02.12.). Battilani, A., 2006. Fertirrigere V2.11: a multi-target dss to manage water and nutrients supply at macrozone level. Acta Horticulture 724 (1), 114e118. Beuchat, L.R., 1996. Pathogenic microorganisms associated with fresh produce. Journal of Food Protection 59 (2), 204e216. Brandl, M.T., 2006. Fitness of human enteric pathogens on plants and implications for food safety. The Annual Review of Phytopathology 44 (1), 367e392. Camp, C.R., Lamm, F.R., Evans, R.G., Phene, C.J., 2000. Subsurface drip irrigation: past, present and future. In: Proceedings of the Fourth Decennial Irrigation Symposium, November 14e16, Phoenix, Arizona. American Society of Agricultural Engineers, St Joseph, Mich., USA, pp. 363e372. Casarez, E.A., Pillai, S.D., Mott, J.B., Vargas, M., Dean, K.E., Di Giovanni, G.D., 2007. Direct comparison of four bacterial source tracking methods and the use of composite data sets. Journal of Applied Microbiology 103 (2), 350e364. Caugant, D.A., Levin, B.R., Selander, R.K., 1981. Genetic diversity and temporal variation in the E. coli population of a human host. Genetics 98 (3), 467e490. CDC (Centers for Disease Control and Prevention), 2007. Multistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants e United States, 2005e2006. Morbidity and Mortality Weekly Report 56 (35), 909e911. De´portes, I., Benoit-Guyed, J.-L., Zmiroub, D., 1995. Hazard to man and the environment posed by the use of urban waste compost: a review. The Science of the Total Environment 172 (2e3), 197e222. Di Cerbo, A.R., Manfredi, M.T., Bregoli, M., Ferro Milone, N., Cova, M., 2008. Wild carnivores as source of zoonotic helminths in north-eastern Italy. Helminthologia 45 (1), 13e19. Edge, T.A., Hill, S., Seto, P., Marsalek, J., 2010. Library-dependent and library-independent microbial source tracking to identify spatial variation in faecal contamination sources along a Lake Ontario beach (Ontario, Canada). Water Science and Technology 62 (3), 719e727. EEA (European Environmental Agency), 2009. Water Resources Across Europe e Confronting Water Scarcity and Drought. EEA Report No 2. El Hamouri, B., Handouf, A., Mekrane, M., Touzani, M., Khana, A., Khallayoune, K., Benchokroun, T., 1996. Use of wastewater for crop production under arid and saline conditions: yield and hygienic quality of the crop and soil contamination. Water Science and Technology 33 (10e11), 327e334. English, M., Raja, S.N., 1996. Perspectives on deficit irrigation. Agricultural Water Management 32 (1), 1e14.

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

Ensink, J.H.J., Mahmood, T., Dalsgaard, A., 2007. Wastewaterirrigated vegetables: market handling versus irrigation water quality. Tropical Medicine and International Health 12 (2), 2e7. EU, 2000. European Water Framework Directive (2000/60/EC). http://ec.europa.eu/environment/water/water-framework/ index_en.html (accessed 18.10.11.). Falk, L.L., 1949. Bacterial contamination of tomatoes grown in polluted soil. American Journal of Public Health 39 (10), 1338e1349. FAO (Food and Agriculture Organization of the United Nations), 2009. Web Site Database. Available at: http://www.fao.org/nr/ water/cropinfo_tomato.html (accessed 18.10.11.). FAO (Food and Agriculture Organization of the United Nations), 2011. Irrigation and Drainage paper no. 33 (Revised) e Tomato. In print. Personal communication of A. Battilani, Consorzio di Bonfica di secondo grado per il Canale Emilliano Romagnolo e CER, Area Agronomico e ambientale, Via E. Masi 8, Ie40137 Bologna, Italy. FDA (U.S. Food and Drug Administration), 2001. Analysis and Evaluation of Preventive Control Measures for the Control and Reduction/Elimination of Microbial Hazards on Fresh and Fresh-cut Produce. Available at: http://www.fda.gov/Food/ ScienceResearch/ResearchAreas/ SafePracticesforFoodProcesses/ucm090977.htm (accessed 18.10.11.). FDA (U.S. Food and Drug Administration), 2009. Guidance for Industry: Guide to Minimize Microbial Food Safety Hazards of Tomatoes; Draft Guidance. Available at: http://www.fda.gov/ Food/GuidanceComplianceRegulatoryInformation/ GuidanceDocuments/ProduceandPlanProducts/ucm173902. htm (accessed 18.10.11.). 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 (4), 893e902. Forslund, A., Plauborg, F., Andersen, M.N., Markussen, B., Dalsgaard, A., 2011. Leaching of human pathogens in repacked soil lysimeters and contamination of potato tubers under subsurface drip irrigation in Denmark. Water Research 45 (15), 4367e4380. Fujioka, R., Sian-Denton, C., Borja, M., Castro, J., Morphew, K., 1999. Soil: the environmental source of Escherichia coli and Enterococci in Guam’s streams. Journal of Applied Microbiology 85 (S1), 83e89. Gagliardi, J.V., Karns, J.S., 2000. Leaching of Escherichia coli O157:H7 in diverse soil under various agricultural management practices. Applied and Environmental Microbiology 66 (3), 877e883. Gerba, C.P., Smith, J.E., 2005. Sources of pathogenic microorganisms and their fate during land application of wastes. Journal of Environmental Quality 34 (1), 42e48. Girones, R., 2006. Tracking viruses that contaminate the environments. Microbe 1 (1), 19e25. Gordon, D.M., Bauer, S., Johnson, J.R., 2002. The genetic structure of Escherichia coli populations in primary and secondary habitats. Microbiology 148 (5), 1513e1522. Greene, S.K., Daly, E.R., Talbot, E.A., Demma, L.J., Holzbauer, S., Patel, N.J., Hill, T.A., Walderhaug, M.O., Hoekstra, R.M., Lynch, M.F., Painter, J.A., 2008. Recurrent multistate outbreak of Salmonella Newport associated with tomatoes from contaminated fields, 2005. Epidemiology and Infection 136 (2), 157e165. Grimason, A.M., Smith, H.V., Thitai, W.N., Smith, P.G., Jackson, M.H., Girdwood, R.W.A., 1993. Occurrence and removal of Cryptosporidium spp oocysts and Giardia spp cysts in Kenyan waste stabilization ponds. Water Science and Technology 27 (7), 97e104.

5933

Habluetzel, A., Traldi, G., Ruggieri, S., Attili, A.R., Scuppa, P., Marchetti, R., Menghini, G., Esposito, F., 2003. An estimation of Toxocara canis prevalence in dogs, environmental egg contamination and risk of human infection in the Marche region of Italy. Veterinary Parasitology 113 (3e4), 243e252. Hanning, I.B., Nutt, J.D., Ricke, S.C., 2009. Salmonellosis outbreaks in the United States due to fresh produce: sources and potential intervention measures. Foodborne Pathogens and Disease 6 (6), 635e648. Harris, L.J., Farber, J.N., Beuchat, L.R., Parish, M.E., Suslow, T.V., Garrett, E.H., Busta, F.F., 2003. Outbreaks associated with fresh produce: incidence, growth, and survival of pathogens in fresh and fresh-cut produce. Comprehensive Reviews in Food Science and Food Safety 2 (S1), 78e141. Haruvy, N., Offer, R., Hadas, A., Ravina, I., 1999. Wastewater irrigation-economic concerns regarding beneficiary and hazardous effects of nutrients. Water Resources Management 13 (5), 303e314. Heaton, J.C., Jones, K., 2008. Microbial contamination of fruit and vegetables and the behaviour of enteropathogens in the phyllosphere: a review. Journal of Applied Microbiology 104 (3), 613e626. Hedberg, C.W., Angulo, F.J., White, K.E., Langkop, C.W., Schell, W.L., Stobierski, M.G., Schuchat, A., Besser, J.M., Dietrich, S., Helsel, L., Griffin, P.M., McFarland, J.W., Osterholm, M.T., 1999. Outbreaks of salmonellosis associated with eating uncooked tomatoes: implications for public health. Epidemiology and Infection 122 (3), 385e393. Hsiao, T.C., Steduto, P., Fereres, E., 2007. A systematic and quantitative approach to improve water use efficiency in agriculture. Irrigation Science 25 (3), 209e231. Huba´lek, Z., 2004. An annotated checklist of pathogenic microorganisms associated with migratory birds. Journal of Wildlife Diseases 40 (4), 639e659. Ibenyassine, K., AitMhand, R., Karamoko, Y., Cohen, N., Ennaji, M.M., 2006. Use of repetitive DNA sequences to determine the persistence of enteropathogenic Escherichia coli in vegetables and in soil grown in fields treated with contaminated irrigation water. Letters in Applied Microbiology 43 (5), 528e533. Ishii, S., Sadowsky, M.J., 2008. Escherichia coli in the environment: implications for water quality and human health. Microbes and Environments 23 (2), 101e108. Islam, M., Doyle, M.P., Phatak, S.C., Millner, P., 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 (7), 1365e1370. Jablasone, J., Brovko, L.Y., Griffiths, M.W., 2004. A research note: the potential for transfer of Salmonella from irrigation water to tomatoes. Journal of the Science of Food and Agriculture 84 (3), 287e289. Jensen, C.R., Battilani, A., Plauborg, F., Psarras, G., Chartzoulakis, K., Janowiak, F., Stikic, R., Jovanovic, Z., Li, G., Qi, X., Liu, F., Jacobsen, S.-E., Andersen, M.N., 2010. Deficit irrigation based on drought tolerance and root signalling in potatoes and tomatoes. Agricultural Water Management 98 (3), 403e413. Kariuki, S., Gilks, C., Kimari, J., Obanda, A., Muyodi, J., Waiyaki, P., Hart, C.A., 1999. Genotype analysis of Escherichia coli strains isolated from children and chickens living in close contact. Applied and Environmental Microbiology 65 (2), 472e476. Kirda, C., Cetin, M., Dasgan, Y., Topcu, S., Kaman, H., Ekici, B., Derici, M.R., Ozguven, A.I., 2004. Yield response of greenhouse grown tomato to partial root drying and conventional deficit irrigation. Agricultural Water Management 69 (3), 191e201.

5934

w a t e r r e s e a r c h 4 6 ( 2 0 1 2 ) 5 9 1 7 e5 9 3 4

Kozan, E., Gonenc, B., Sarimehmetoglu, O., Aycicek, H., 2005. Prevalence of helminth eggs on raw vegetables used for salads. Food Control 16 (3), 239e242. Lapidot, A., Yaron, S., 2009. Transfer of Salmonella enterica serovar Typhimurium from contaminated irrigation water to parsley is dependent on curli and cellulose, the biofilm matrix components. Journal of Food Protection 72 (3), 618e623. Lavalle, C., Micale, F., Houston, T.D., Camia, A., Hiederer, R., Lazar, C., Conte, C., Amatulli, G., Genovese, G., 2009. Climate change in Europe. 3. Impact on agriculture and forestry: a review. Agronomy for Sustainable Development 29 (3), 433e446. Lu, L., Hume, M.E., Sternes, K.L., Pillai, S.D., 2004. Genetic diversity of Escherichia coli isolates in irrigation water and associated sediments: implications for source tracking. Water Research 38 (18), 3899e3908. Lynch, M.F., Tauxe, R.V., Hedberg, C.W., 2009. The growing burden of foodborne outbreaks due to contaminated fresh produce: risks and opportunities. Epidemiology and Infection 137 (3), 307e315. Mara, D.D., Sleigh, P.A., Blumenthal, U.J., Carr, R.M., 2007. Health risks in wastewater irrigation: comparing estimates from quantitative microbial risk analyses and epidemiological studies. Journal of Water and Health 5 (1), 39e50. Mara, D., 2011. Water- and wastewater-related disease and infection risk: what is an appropriate value for the maximum tolerable additional burden of disease? Journal of Water and Health 9 (2), 217e224. McLellan, S.L., Huse, S.M., Mueller-Spitz, S.R., Andreishcheva, E.N., Sogin, M.L., 2010. Diversity and population structure of sewage-derived microorganisms in wastewater treatment plant influent. Environmental Microbiology 12 (2), 378e392. Mukherjee, A., Speh, D., Jones, A.T., Buesing, K.M., DiezGonzalez, F., 2006. Longitudinal microbiological survey of fresh produce grown by farmers in the upper Midwest. Journal of Food Protection 69 (8), 1928e1936. Oragui, J.I., Curtis, T.P., Silva, S.A., Mara, D.D., 1987. The removal of excreted bacteria and viruses in deep waste stabilization ponds in northeast Brazil. Water Science and Technology 19 (3e4), 569e573. Palese, A.M., Pasquale, V., Celano, G., Fifliuolo, G., Masi, S., Xiloyannis, C., 2009. Irrigation of olive groves in Southern Italy with treated municipal wastewater: effects on microbiological quality of soil and fruits. Agriculture. Ecosystems and Environment 129 (1e2), 43e51. Pescod, M.B., 1992. Wastewater Treatment and Use in Agriculture. In: FAO Irrigation and Drainage Paper 47. Rome. Rangarajan, A., Bihn, E.A., Gravani, R.B., Scott, D.L., Pritts, M.P., 2000. Food Safety Begins on the Farm: a Grower’s Guide. Available at: http://ecommons.cornell.edu/handle/1813/2209 (accessed 18.10.11.). Ribot, E.M., Fair, M.A., Gautom, R., Cameron, D.N., Hunter, S.B., Swaminathan, B., Barrett, T.J., 2006. Standardization of pulsed-field gel electrophoresis protocols for the subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne Pathogens and Disease 3 (1), 59e67. Rushing, J.W., Angulo, F.J., Beuchat, L.R., 1996. Implementation of a HACCP program in a commercial fresh-market tomato packinghouse: a model for the industry. Dairy, Food and Environmental Sanitation 16 (9), 549e553. Sadovski, A.Y., Fattal, B., Goldberg, D., 1978. Microbial contamination of vegetables irrigated with sewage effluent by the drip method. Journal of Food Protection 41 (5), 336e340. Sagoo, S.K., Little, C.L., Mitchell, R.T., 2001. The microbiological examination of ready-to-eat organic vegetables from retail

establishments in the United Kingdom. Letters in Applied Microbiology 33 (6), 434e439. Shahnazari, A., Liu, F., Andersen, M.N., Jacobsen, S.E., Jensen, C.R., 2007. Effects of partial root-zone drying on yield, tuber size and water use efficiency in potato under field conditions. Field Crop Research 100 (1), 117e124. Shaw, R.K., Lasa, I., Garcı´a, B.M., Pallen, M.J., Hinton, J.C.D., Berger, C.N., Frankel, G., 2011. Cellulose mediates attachment of Salmonella enterica serovar Typhimurium to tomatoes. Environmental Microbiology Reports 3 (5), 569e573. Shi, X., Namvar, A., Kostrzynska, M., Hora, R., Warriner, K., 2007. Persistence and growth of different Salmonella serovars on preand postharvest tomatoes. Journal of Food Protection 70 (12), 2725e2731. Sivapalasingam, S., Friedman, C.R., Cohen, L., Tauxe, R.V., 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. Journal of Food Protection 67 (10), 2342e2353. Solomon, E.B., Potenski, C.J., Matthews, K.R., 2002. Effect of irrigation method on transmission to and persistence of Escherichia coli O157:H7 on lettuce. Journal of Food Protection 65 (4), 673e676. Song, I., Stine, S.W., Choi, C.Y., Gerba, C.P., 2006. Comparison of crop contamination by microorganisms during subsurface drip and furrow irrigation. Journal of Environmental Engineering 132 (10), 1243e1248. Speare, K., 1989. Identification of species of Strongyloides. In: Grove, D.I. (Ed.), Strongyloidiasis: a Major Roundworm Infection of Man. Taylor and Francis, London, UK, pp. 11e84. Steele, M., Odumeru, J., 2004. Irrigation water as source of foodborne pathogens on fruit and vegetables. Journal of Food Protection 67 (12), 2839e2849. Stoeckel, D.M., Mathes, M.V., Hyer, K.E., Hagedorn, C., Kator, H., Lukasik, J., O’Brien, T.L., Fenger, T.W., Samadpour, M., Strickler, K.M., Wiggins, B.A., 2004. Comparison of seven protocols to identify fecal contamination sources using Escherichia coli. Environmental Science and Technology 38 (22), 6109e6117. Talley, J.L., Wayadande, A.C., Wasala, L.P., Gerry, A.C., Fletcher, J., DeSilva, U., Gilliland, S.E., 2009. Association of Escherichia coli O157:H7 with filth flies (Muscidae and Calliphoridae) captured in leafy greens fields and experimental transmission of E. coli O157:H7 to spinach leaves by house flies (Diptera: Muscidae). Journal of Food Protection 72 (7), 1547e1552. Tauxe, R., Kruse, H., Hedberg, C., Potter, M., Madden, J., Wachsmuth, K., 1997. Microbial hazards and emerging issues associated with produce. A preliminary report to the national advisory committee on microbiological criteria for foods. Journal of Food Protection 60 (11), 1400e1408. Toze, S., 2006. Reuse of effluent water e benefits and risks. Agricultural Water Management 80 (1e3), 147e159. Tsanis, I.K., Londra, P.A., Angelakis, A.N., 1997. Assessment of water needs for irrigation in Greece. Acta Horticulturae 449 (1), 41e48. USEPA, 2004. Guidelines for Water Reuse. US Environmental Protection Agency, Washington DC, US. Wahlstro¨m, H., Tyse´n, E., Olsson Engvall, E., Bra¨ndstro¨m, B., Eriksson, E., Mo¨rner, T., Va˚gsholm, I., 2003. Survey of Campylobacter species, VTEC O157 and Salmonella species in Swedish wildlife. Veterinary Records 153 (3), 74e80. WHO, 2006. Guidelines for the Safe Use of Wastewater in Agriculture. World Health Organization, Geneva, Switzerland. Zhuang, R.Y., Beuchat, L.R., Angulo, F.J., 1995. Fate of Salmonella Montevideo on and in raw tomatoes as affected by temperature and treatment with chlorine. Applied and Environmental Microbiology 61 (6), 2127e2131.