The Origin and Control of Microorganisms Associated With

Origin and Detection of Bacterial Species Associated

with Lettuce and Salad Vegetables

Peter James Ng

B Sc. (Hons) Food Science and Technology

(The University of New South Wales, Australia)

A thesis submitted for the degree of

Doctor of Philosophy

Food Science and Technology

School of Chemical Sciences and Engineering

The University of New South Wales

2007

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………...... Date ……………………………………………...... COPYRIGHT STATEMENT

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………...... Date ……………………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………...... Date ……………………………………………......

ACKNOWLEDGEMENTS

During the duration of this study, there were many people and associations involved who need to be acknowledged.

Firstly, I would like to acknowledge my supervisor, Professor Graham Fleet, and my co- supervisor, Dr Gillian Heard, for their assistance, support and guidance during the course of this study.

Secondly, the assistance of the Australian Research Council (ARC) and Harvest

FreshCuts/Vegco Pty. Ltd., Wacol, Australia for funding of this research.

The donation of concentrated pesticides by the manufacturers, BASF Australia Ltd.,

Nufarm Australia Ltd. and Syngenta Crop Protection Pty. Ltd., Australia.

The technical staff of the Department of Food Science and Technology, especially

Camillo Taraborrelli, for their assistance and support.

I would also like to acknowledge, my fellow postgraduate research colleagues, Ai Lin,

Sungsook, Hugh, Pat and Lidia for their friendship and support during the long days and nights taken to complete this research.

Finally, thanks to my family, for their assistance and support during the course of this research.

PUBLICATIONS RELATED TO THIS THESIS

Ng, P.J., Fleet, G.H., Heard, G.M. 2005. Pesticides as a source of microbial contamination of salad vegetables. International Journal of Food Microbiology 101:

237-250

iii

TABLE OF CONTENTS

ORIGINALITY STATEMENT...... i

AUTHENTICITY STATEMENT ...... i

ACKNOWLEDGEMENTS ...... ii

PUBLICATIONS RELATED TO THIS THESIS...... iii

TABLE OF CONTENTS...... iv

ABSTRACT...... 1

CHAPTER 1: INTRODUCTION ...... 3

CHAPTER 2: LITERATURE REVIEW ...... 8

2.1 Australian fresh-cut, vegetable salad industry...... 8

2.2 Production chain...... 10

2.3 Microbiology of lettuce and fresh cut salads...... 14

2.4 Microflora of lettuce and ready-to-eat salads...... 14

2.5 Significance of microflora of lettuce and ready-to-eat salads...... 24

2.6 Sources of microbial contamination of vegetable produce ...... 28

2.6.1 Irrigation water...... 28

2.6.2 Irrigation method...... 29

2.6.3 Soil ...... 30

2.6.4 Fertilisers and Manure ...... 31

2.6.5 Pesticides...... 32

2.6.6 Pests and Farm Animals...... 33

2.6.7 Handling operations ...... 35

2.7 Methods for studying the microbial ecology of lettuce and related produce ...37

2.7.1 Culture-dependent microbiological analysis of vegetables...... 37

2.7.2 Molecular strategies for monitoring bacterial communities ...... 40

2.7.3 Factors affecting the performance of molecular methods for analysis of

bacteria...... 43

2.8 Antimicrobials and lettuce...... 45

2.8.1 Natural antimicrobials...... 45

2.8.2 Methods for the extraction of antimicrobial compounds from plant

materials ...... 46

CHAPTER 3: PESTICIDES AS A SOURCE OF MICROBIAL CONTAMINATION

OF VEGETABLE CROPS ...... 49

3.1 Introduction ...... 49

3.2 Materials and Methods ...... 51

3.2.1 Pesticides...... 51

3.2.2 Microorganisms and their cultivation ...... 51

3.2.3 Analysis of pesticide concentrates for microorganisms...... 53

3.2.4 Antimicrobial properties of pesticide solutions ...... 53

3.2.5 Growth and survival of pathogenic and spoilage organisms in pesticide

solutions ...... 54

3.2.6 Growth and survival of natural populations from agricultural water in

pesticide solutions...... 54

3.2.7 Identification of bacterial isolates ...... 55

3.3 Results ...... 57

3.3.1 Microbial content and antimicrobial activity of pesticides...... 57

3.3.2 Survival and growth of microorganisms inoculated into pesticide solutions

...... 57

3.3.3 Growth and survival of microorganisms in pesticides reconstituted in

different sources of agricultural water ...... 61

3.3.4 Identification of bacterial species that grow in pesticides ...... 69

3.4 Discussion ...... 73

CHAPTER 4: THE BACTERIAL ECOLOGY OF LETTUCE AS DETERMINED BY

CULTURAL AND PCR-DENATURING GRADIENT GEL ELECTROPHORESIS

(DGGE) ANALYSIS ...... 78

4.1 Introduction ...... 78

4.2 Materials and Methods ...... 81

4.2.1 Lettuce samples...... 81

4.2.2 Bacteriological analysis of lettuce ...... 81

4.2.3 Culture on agar media ...... 82

4.2.3 Culture on agar media ...... 83

4.2.4 DNA extraction for bacterial identification and PCR-DGGE analysis....83

4.2.5 PCR amplification of DNA for sequence identification and for PCR-

DGGE analysis...... 84

4.2.6 Denaturing gradient gel electrophoresis (DGGE)...... 86

4.2.7 Sequence analysis of rDNA for bacterial identification and DGGE bands ..

...... 87

4.2.8 Bacterial species used as reference cultures...... 87

4.2.9 Detection limit of bacterial populations by PCR-DGGE...... 89

4.2.10 Detection limit of bacterial populations inoculated onto lettuce and

analysed using PCR-DGGE ...... 89

4.2.11 Statistical analysis of data...... 90

4.3 Results ...... 91

4.3.1 Effect of maceration, rinsing and Tween 80 on recovery of bacterial

populations from lettuce samples...... 91

4.3.2 Comparison of bacterial populations obtained by maceration and rinsing ..

of lettuce samples...... 91

4.3.3 Effect of addition of Tween 80 on bacterial populations obtained from

maceration and rinses of lettuce samples...... 92

4.3.4 Bacterial species isolated from lettuce by agar plating...... 93

4.3.5 Comparison of bacterial species isolated from macerates and rinses of

lettuce samples ...... 101

4.3.6 Comparison of bacterial species isolated from macerates and rinses of

lettuce samples in the presence of Tween 80...... 102

4.3.7 Comparison of agar culture and PCR-DGGE methods for profiling the

bacterial ecology of lettuce ...... 104

4.3.8 Species isolated and detected in lettuce samples ...... 104

4.3.9 Comparison of maceration and rinsing methods, and addition of Tween 80

on detection of bacterial species in lettuce samples by PCR-DGGE...... 111

4.3.10 Overall comparison of culture and PCR-DGGE assays for detection of

bacterial species in lettuce samples...... 114

4.3.11 PCR-DGGE detection of bacterial species associated with lettuce samples

...... 116

4.3.12 The use of additives in the PCR mixture to enhance amplification of

bacterial DNA ...... 120

4.4 Discussion ...... 122

4.4.1 Populations of bacterial species associated with lettuce...... 122

4.4.2 Diversity of bacterial species associated with lettuce...... 124

vii

4.4.3 PCR-DGGE...... 130

CHAPTER 5: ELIMINATION OF CHLOROPLAST DNA FROM THE PCR-DGGE

ASSAY OF BACTERIA IN LETTUCE HOMOGENATES ...... 138

5.1 Introduction ...... 138

5.2 Materials and Methods ...... 140

5.2.1 Preparation of lettuce homogenates ...... 140

5.2.2 Effect of sample preparation on the presence of chloroplasts in lettuce

suspensions...... 140

5.2.3 Removal of chloroplast cells by differential centrifugation...... 141

5.2.4 Use of titanium hydroxide to separate bacterial and chloroplast cells....142

5.2.5 Preparative agarose gel electrophoresis ...... 143

5.2.6 Recovery of DNA from low melt agarose gels...... 144

5.2.7 Reference strains of bacteria used for the assessment of preparative gel

electrophoresis...... 144

5.2.8 DNA Extraction Methods ...... 145

5.2.9 PCR amplification of DNA for PCR-DGGE analysis ...... 147

5.2.10 Denaturing gradient gel electrophoresis (DGGE)...... 147

5.3 Results ...... 148

5.3.1 Effect of sample preparation on the release of chloroplasts from spinach

leaves...... 148

5.3.2 Removal of chloroplasts by differential centrifugation ...... 149

5.3.3 Removal of chloroplasts by titanium (III) hydroxide ...... 151

5.3.4 Removal of chloroplast by preparative gel electrophoresis...... 151

5.4 Discussion ...... 155

viii

CHAPTER 6: OCCURRENCE AND DETECTION OF ACINETOBACTER SPECIES

IN LETTUCE AND VEGETABLE SALADS; EVALUATION OF MEDIA...... 159

6.1 Introduction ...... 159

6.2 Materials and Methods ...... 161

6.2.1 Salad vegetable samples...... 161

6.2.2 Media for isolation of Acinetobacter species...... 161

6.2.3 Analysis for Acinetobacter species ...... 163

6.2.4 Evaluation of selective plating media for the isolation of Acinetobacter

species ...... 164

6.2.5 Evaluation of selective enrichment broths for growth of Acinetobacter

species ...... 165

6.3 Results ...... 167

6.3.1 Presence of Acinetobacter species in lettuce and vegetable salads as

determined by direct plating onto MLAA and MSA ...... 167

6.3.2 Comparison of MLAA and MSA media for the growth of Acinetobacter

and other species ...... 169

6.3.3 Presence of Acinetobacter species in lettuce and vegetable salads as

determined by enrichment culture...... 173

6.4 Discussion ...... 175

CHAPTER 7: THE EFFECT OF EXTRACTS FROM LETTUCE AND OTHER

VEGETABLES ON THE SURVIVAL OF BACTERIA OF SPOILAGE AND PUBLIC

HEALTH SIGNIFICANCE...... 180

7.1 Introduction ...... 180

7.2 Materials and Methods ...... 182

7.2.1 Vegetable samples...... 182

7.2.2 Vegetable juice...... 182

7.2.3 Vegetable powder ...... 182

7.2.4 Solvent extraction of vegetable powders ...... 183

7.2.5 Supercritical fluid extraction...... 184

7.2.6 Microorganisms ...... 186

7.2.7 Antimicrobial properties of vegetable extracts ...... 187

7.2.8 Hole-in-plate method ...... 188

7.2.9 Paper-disc diffusion assay...... 188

7.3 Results ...... 189

7.3.1 Vegetable juice...... 189

7.3.2 Evaluation of antimicrobial activity in solvent extracts from lettuce and

capsicum...... 189

7.3.3 Evaluation of antimicrobial activity in supercritical fluid extracts of

lettuce and capsicum ...... 194

7.4 Discussion ...... 195

CHAPTER 8: CONCLUSIONS ...... 201

CHAPTER 9: BIBLIOGRAPHY...... 206

ABSTRACT

Ready-to-eat vegetable salads containing lettuce as a main ingredient have become popular food items in recent years. Microorganisms associated with these products determine their shelf-life, sensory appeal and safety. This thesis investigates the bacterial ecology of lettuce, aspects of their pre-harvest contamination with microorganisms, and the presence of antimicrobial constituents in such produce.

Commercial pesticides (insecticides, herbicides, fungicides), used during lettuce cultivation were examined as potential sources of microbial contaminants. None of the pesticide concentrates contained viable microorganisms. After reconstitution in water, two of the pesticides supported growth of inoculated species of Pseudomonas,

Salmonella and Escherichia coli. Pesticides reconstituted in agricultural waters (bore, dam and river) supported the growth of microorganisms (e.g. Pseudomonas,

Acinetobacter, Aeromonas spp. and coliforms) naturally present in these waters. Unless properly managed, pesticide application could contribute microbial contaminants to vegetable produce, thereby affecting their quality.

Bacterial species associated with retail samples of lettuce were examined by plate culture on Tryptone Soy Agar and PCR-DGGE analysis. Macerates and rinses of lettuce sub-samples with and without addition of Tween 80 were examined to maximize bacterial recoveries. Predominant bacteria isolated by agar culture included species of

Pseudomonas, Agrobacterium, Curtobacterium and Burkholderia, at populations of 103-

106 cfu/g. PCR-DGGE was unable to recover the same incidence of species as agar culture and failed to detect bacteria in many samples. In some samples, PCR-DGGE detected species of Bacillus, Pseudomonas, Serratia and Acinetobacter, not found by culture. Failure of the PCR-DGGE analyses was attributed to interference by plant

chloroplast DNA. Preparative agarose gel electrophoresis of lettuce macerates was necessary to remove chloroplast DNA before application of PCR-DGGE analysis.

Thirty percent of lettuce samples contained Acinetobacter species at 101-104 cfu/g when examined after culture on minimal salts agar or enrichment in Baumann enrichment medium. Other Acinetobacter media failed to give reliable isolation of these species from lettuce and salad vegetables. Lettuce could be an environmental source of

Acinetobacter nosocomial infections.

Juices, solvent extracts and supercritical fluid carbon dioxide extracts of lettuce and capsicum samples did not exhibit antimicrobial action against a range of food spoilage and pathogenic bacteria.

CHAPTER 1

INTRODUCTION

Fresh cut, minimally processed vegetable products can be described as any fresh produce that has been washed, peeled, cut or shredded and packaged for sale and consumption without any further processing. Lettuces of various varieties are frequently the main component of these products, but other ingredients may include capsicum

(green pepper), carrots, cabbage and celery (Nguyen-the and Carlin 2000). Consumer demands for healthy, convenient, “life-style” foods have led to a rapid, world-wide growth in markets for these products (Jobling et al. 1998; Lucier 1998; Sloan 2000).

The increased popularity of salad vegetables has focussed greater attention on their quality and factors that affect this quality. Microorganisms have a significant role in this context. The shelf-life of ready-to-eat salads is generally about 7-14 days at 5 °C, being limited by endogenous biochemical and physiological changes in the produce, and the growth of spoilage microorganisms (Garcia-Gimeno and Zurera-Cosano 1997; Heard

1999; Heard 2002). Because these products are only minimally processed and consumed in the fresh state, they have increased risk with respect to public health. In recent years, numerous outbreaks of gastrointestinal illness have been linked to their consumption and contamination with bacterial pathogens such as Listeria monocytogenes,

Salmonella, Shigella sonnei and Escherichia coli O157 (Schlech et al. 1983; Cieslak et al. 1993; Stafford et al. 2002; Horby et al. 2003; Welinder-Olsson et al. 2003).

A comprehensive understanding of the microbial ecology of vegetable produce and the factors that lead to their contamination with microorganisms is needed for effective management of their quality and safety.

It was the original goal of this project and thesis to undertake a thorough investigation of the microbial ecology of lettuce as it evolved and changed throughout the total chain of production- cultivation at the farm, harvesting, transport, minimal processing and packaging, retailing. Lettuce was chosen as the vegetable product for study because of its prominent position as a component of ready-to-eat salad products. For the purpose of this investigation, collaborative partnerships were established with several lettuce farmers in the states of New South Wales (NSW) and Queensland (QLD), and a major processor (Harvest FreshCuts Pty. Ltd) of packaged, ready-to-eat salads in Queensland.

Unfortunately, the onset and duration of severe drought conditions in lettuce farming regions of NSW and QLD during the growing seasons of 2001 to 2005, significantly disrupted the cultivation and supply of lettuce samples needed for a systematic and meaningful investigation of their microbiology, and the factors that affected their microbial ecology. Consequently, the original goals of the project needed to be modified.

Preharvest farming practices are now recognised as being a significant source of microbial contamination of vegetable produce (Beuchat 2002). Irrigation water and application of fertilisers are known as particular microbiological hazards in this context

(Beuchat 1996a; Brackett 1999; Beuchat 2002). A diversity of pesticides is widely and frequently applied during the cultivation of vegetables, including lettuce, but little attention has been given to these agents as a potential source of pre-harvest contamination. During visits to lettuce farms and discussion with lettuce farmers, it was noted that these pesticides were reconstituted in a variety of agricultural waters and, after reconstitution, variable times could elapse before their application to the product.

Consequently, one objective of this thesis was to examine the microbiological quality of various pesticides used in the cultivation of lettuces, and to determine their potential as

a source of microbiological contamination after reconstitution in various types of agricultural waters. This investigation is presented in Chapter 3.

A review of the literature has revealed little quantitative information about the microbial ecology of lettuces. Although various bacterial species such as Pseudomonas, Serratia,

Erwinia, Enterobacter, Pantoea, Burkholderia and Stenotrophomonas are frequently isolated from lettuces, their quantitative populations are rarely reported (Riser et al.

1984; Magnuson et al. 1990; King Jr et al. 1991; Khan et al. 1992; Freire and Robbs

2000; Hamilton-Miller and Shah 2001). Population data are particularly important in determining which species are most likely to impact on product quality. Moreover, there has been no critical evaluation or discussion of the methods used to analyse the bacterial ecology of lettuce produce. Plant products are known to contain antimicrobial constituents (Beuchat & Golden 1989; Beuchat and Brackett 1990; Beuchat et al. 1994;

Cowan 1999; Rauha et al. 2000; Burnett and Beuchat 2001; Beuchat 2006; Davidson and Taylor 2007) that could impact on the recovery and isolation of bacteria, when plant macerates or homogenates are used as the starting material for bacteriological analysis.

It is now widely accepted in the field of microbial ecology that many habitats harbour viable bacterial species that are not isolated or recovered by conventional agar plate culture methods (Muyzer and Smalla 1998; Fleet 1999; Giraffa 2004; De Vero et al.

2006). Consequently, culture-independent technologies, based on the analysis of extracted DNA, are being increasingly used to obtain more accurate information about the microbial ecology of various natural habitats including foods (Ercolini et al. 2003b;

Miambi et al. 2003; Cocolin et al. 2001b, 2004; Ercolini 2004; Giraffa 2004; Handschur et al. 2005; Lee et al. 2005; De Vero et al. 2006). A widely used protocol is based on polymerase chain reaction (PCR)-denaturing gradient gel electrophoresis (DGGE).

Here, DNA extracted from the habitat is specifically amplified by PCR. Amplicons of

bacterial DNA are then separated by DGGE which resolves them into specific DNA bands on the basis of nucleic acid base sequence. Such bands are then extracted from the electrophoretic gel and further analysed for their sequence, which gives species identity on comparison with established data bases. PCR-DGGE analyses have been recently used for investigating the microbial ecology of various foods and beverages, often revealing the presence of species not detected by cultural methods (Miambi et al.

2003; De Vero et al. 2006; Nielsen et al. 2007). Such technology has not yet been used to investigate the bacterial ecology of lettuce. A second objective of this thesis was to conduct a more thorough analysis of the bacterial ecology of lettuces, using macerates and rinses of lettuces as starting material and comparing the ecological data obtained by agar plate culture and PCR-DGGE analysis. These investigations are reported in

Chapters 4 and 5.

Information obtained during the investigations reported in Chapters 3 and 4, suggested the association of Acinetobacter species with lettuce samples and with waters used for irrigation during their cultivation. Acinetobacter species have recently emerged as very significant agents in the cause of nosocomial infections (Bergogne-Bérézin and Towner

1996; Van Looveren et al. 2004; Hanlon 2005). Such infections are particularly serious because some species and strains of Acinetobacter have developed very strong resistance to antibiotics routinely used to treat the infection (Van Looveren et al. 2004;

Hanlon 2005). Vegetable salads, therefore, could be a significant environmental source of Acinetobacter species in hospitals. Chapter 6 of this thesis, provides a more detailed study of lettuce and ready-to-eat vegetable salads for the occurrence and populations of

Acinetobacter species, and gives an evaluation of cultural methods used to isolate and enumerate these species from salad produce.

During visits to lettuce farms, it was noted that harvesting and removal of roots from the head of each lettuce caused injury to the lettuce tissue that resulted in a sap which exuded or leaked out from the sections of damaged tissue. Individual lettuces were gathered and packed together in larger wooden crates for transport to the processor or retailer. The exudates permeated the lettuces in the crate. As mentioned already, vegetable extracts are known to have antimicrobial properties (Beuchat & Golden 1989) and the exudates/sap observed could impact on the bacterial ecology of the lettuce.

Chapter 7 of this thesis reports an investigation of the antimicrobial properties of lettuce extracts.

CHAPTER 2

LITERATURE REVIEW

This thesis is mainly concerned with the bacterial ecology of lettuce because of its prominent role as an ingredient in many ready-to-eat, salad vegetable products. The literature survey will provide some background information on the microbiology of salad vegetable products and lettuce, and a focus on their potential for microbial contamination at the stage of pre-harvest. Methods for studying the bacterial ecology of lettuce and other vegetable produce will be evaluated, including the recent application of culture-independent molecular methods based on the analysis of extracted DNA. A final section will consider the presence and significance of any antimicrobial constituents in lettuce and related products.

2.1 Australian fresh-cut, vegetable salad industry

A broad range of plant produce is used to prepare vegetable salads. These vegetables include carrots, red and green cabbage, several varieties of lettuce such as Iceberg, Cos, fancy green, red and oak leaf, capsicums and celery. For most salads, lettuce is the major component, with the Iceberg variety being the most commonly used. Fresh cut, minimally processed plant products include ready-to-eat salads, stir fry mixes and prepacked vegetables. These products are readily available in supermarkets and represent an easy and convenient way of preparing and supplementing a meal. The appeal and demand of these minimally processed products are due to consumer perceptions that they are healthy, tasty, convenient and fresh (Sloan 2000; Garrett et al.

2003; Buck et al. 2003; Mehrotra 2004; Pivarnik et al. 2005). 8

Internationally, the fresh-cut salad industry has grown over the last decade. In Europe, fresh ready-to-eat salads were classified as one on the fastest growing food categories in

2004, with a growth of 8% and an increase of $166 million Euros in value (Sloan 2005).

In the United Kingdom, the market for bagged (ready-to-eat) salads grew by 10% in

2002 and 2003 (Tyrrel et al. 2006). The population of the United States of America has also increased consumption of fresh fruits and vegetables (Sivapalasingam et al. 2004) and according to agricultural economists, increased in sales from $US 3 billion in 1994 to $US 12.5 billion in 2004 (Zhang 2005; Mukherjee et al. 2006). The sales of pre-cut fresh salads sold from supermarkets alone totalled $US 2.4 billion in 2004, which was up by 7.5% from the previous year (Sloan 2005). Fresh cut ready-to-eat salads are still continuing to grow in the USA, with a 10% growth being recorded in 2006 (Sloan

2007).

In Australia, preparation of fresh-cut salad products is conducted by several major companies which include OneHarvest, (formerly known as Harvest FreshCuts, Vegco

Pty Ltd and The Harvest Company), Mrs Crockett’s and Moraitis Fresh. This value added industry was worth $1.2 billion in 2004/2005 (Australian Food Statistics 2006) and had an increase in sales of 28.6% in the period of 1997 to 2001 (NFIS 2003).

However, national production of lettuce dropped from 152,000 tonnes in 2000/2001 to

132,400 tonnes in 2004/2005 due to severe drought conditions that occurred throughout much of Australia during that period, as lettuces require a significant amount of water for growth (Australian Food Statistics 2006).

2.2 Production chain

The production of ready-to-eat salads consists of growing and harvesting the produce, packing, transport, processing and retailing, and is generally referred to as the ‘farm to fork’ or ‘paddock to plate’ sequence of operations. Throughout this entire chain, it is critical to maintain optimal conditions of hygiene, temperature, humidity and time management, so that the product arrives to the consumer at the best sensory quality and in a safe microbiological condition. Aspects of this industry and the production chain have received significant attention in recent years and have been reviewed in Beuchat

(2002), IFT (2003) and Sapers et al. (2006).

The first stage of the production chain is growing of the produce on the farm. The production of lettuce requires a vast quantity of water during growth from seedling to mature vegetable, which usually takes 6-8 weeks depending on the time of year (Figure

2.1). The seedlings are usually grown from seeds by the grower and then transplanted to the field.

Seedling Pre-heart Early Hearting Full Hearting + Harvest

Figure 2.1 Key stages in the development of lettuce on the farm (adapted from McDougall et al. 2002)

Some farmers now grow baby leaf varieties of lettuce where mixed seeds of different types of lettuce are sown directly into the field. The lettuce is grown as a dense lawn until maturity (usually 2 weeks) and then harvested using a specially adapted tractor and cutter (Anon 2002).

After harvest, the lettuce is usually manually packed into boxes or containers and transported under refrigeration to the processing factory. A post-harvest rinse of the produce with potable water, just prior to entry to the processing plant is sometimes done to remove any soil or debris loosely attached to the surface of the product. An example of a typical process for the manufacture of ready-to-eat salads is shown in Figure 2.2.

After arrival in the factory, the vegetables are cut and trimmed to size. The cutting and trimming process alters the biological structure of the vegetable and also releases nutrients, which microorganisms may utilise for growth.

Harvest

Transport to factory

Postharvest rinse

Cutting & trimming of vegetables

Wash with sanitiser

Rinse with water

Remove water

Package under modified atmosphere

Storage (4°C)

Figure 2.2 Sequence of operations in the production of minimally processed salads

The cut vegetables are then washed in the presence of a sanitiser to reduce the microbial populations on the surface of the vegetable (Beuchat 1998; Beuchat et al. 1998; Beuchat et al. 2001; Sapers 2006). A variety of different sanitisers is available for washing fresh produce, with chlorine based sanitisers (hypochlorite) being the most widely used in the food industry (Cherry 1999; Sapers 2006). Other sanitisers that have been developed for the reduction of microbial load on vegetables include organic acids, hydrogen peroxide, ozone, irradiation, iodine, bromine, chlorine dioxide, trisodium phosphate, quaternary

ammonium compounds and acidified sodium chlorite (Kim et al. 1999; Lin et al. 2000;

Koseki et al. 2001; Nascimento et al. 2003; Parish et al. 2003; Koseki et al. 2004;

Sapers 2006). Usually, these treatments decrease the total microbial populations by 1-2 log cfu/g (Beuchat et al. 1998, Sapers 2006). After washing, the vegetables are usually packaged under a modified atmosphere and then maintained at refrigeration temperature

(4°C) throughout subsequent transport, storage and retailing.

Salad produce is not chemically or biologically stable, and generally has a shelf life of no longer than 7-14 days under proper refrigerated storage (Garcia-Gimeno and Zurera-

Cosano 1997; Heard 1999; Heard 2002). This means that both microbiological and physiological activities could play a role in quality degradation during storage.

Due to the use of blades during processing and harvesting, minimally processed vegetables contain much wounded or damaged tissue. These wounded areas cause a stress response in the produce itself, such as an increase in respiration rate and ethylene production, resulting in faster metabolic rates (King and Bolin 1989; Watada et al.

1996; Saltveit 1999; Surjadinata and Cisneros-Zevallos 2003; Ragaert et al. 2007).

Besides these changes in metabolic rates, damage leads to exposure to air, desiccation and exposure of enzymes to their substrates, all leading to quality degradation (Klein

1987; King and Bolin 1989; Ragaert et al. 2007).

Salad produce is not microbiologically sterile, despite the washing and sanitising operations. Prepared under good manufacturing practices, salad produce will harbour viable microbial populations of 101-103cfu/g (Heard 2002). These microorganisms will continue to grow during subsequent storage, decreasing sensory quality, limiting shelf life, and compromising product safety. The occurrence and significance of microorganisms in determining the quality and safety of lettuce and other produce are discussed in more detail in the following sections.

2.3 Microbiology of lettuce and fresh cut salads

The microorganisms associated with raw and processed vegetable products have been reviewed in recent years by many researchers (Nguyen-the and Carlin 1994, 2000;

Francis et al. 1999; Heard 2002; Harris et al. 2003; Breidt 2006). The increased focus on the microbiology of these products has arisen because of their expanding popularity and associated public health risks, as they are generally consumed without further cooking and processing. From a marketing perspective, there has been a goal to enhance their shelf-life and sensory acceptability. Because these products tend to have neutral pH and a reasonably high water activity, bacteria are of more concern than yeasts or filamentous fungi in determining their microbiological quality (Lund 1992).

Consequently, the focus of this review of background knowledge will be on bacteria.

Microbial contamination and growth can occur at both the preharvest and postharvest stages of production (Lund 1992; Beuchat 1996a; Brackett 1999; Suslow et al. 2003;

Gorny 2006), and will be discussed in the following sections.

2.4 Microflora of lettuce and ready-to-eat salads

The microbial ecology of lettuce consists of bacteria, yeasts and moulds, all coexisting in the one environment (Saddik et al. 1985; Albrecht et al. 1995; Pingulkar et al. 2001;

Viswanathan and Kaur 2001; Thunberg et al. 2002). Total populations of aerobic bacteria on the surface of lettuce vary between 104-107 cfu/g (Riser et al. 1984; Garg et al. 1990; Gras et al. 1994; Pingulkar et al. 2001), with the predominant species being reported as Pseudomonas fluorescens, Enterobacter cloacae, Erwinia carotovora and

Pantoea agglomerans (Table 2.1). Populations of yeast and moulds on the surface of

lettuce vary between 103- 106 cfu/g and <100-103 cfu/g, respectively (Tournas 2005).

The predominant moulds for these samples of lettuce were identified as species of

Alternaria, Cladosporium and Penicillium (Tournas 2005).

Table 2.1 Predominant bacterial species associated with lettuce

Bacterial Species Frequency of isolation; Reference no. samples examined Aeromonas hydrophila 2.9% (2/68) Riser et al. 1984 Citrobacter freundii 22.1% (15/68) Enterobacter aerogenes 1.5% (1/68) Enterobacter cloacae 36.8% (25/68) Klebsiella pneumoniae 8.8% (6/68) Pantoea agglomerans 14.7% (10/68) Proteus morganii 4.4% (3/68) Providencia stuartii 7.4% (5/68) Pseudomonas alcalifaciens 1.5% (1/68) Burkholderia cepacia Not reported Freire and Robbs 2000 Escherichia hermannii Klebsiella oxytocica Stenotrophomonas maltophilia Alcaligenes faecalis 2.3% (1/44) Magnuson et al. 1990 Bacillus spp. 2.3% (1/44) Curtobacterium flaccumfaciens 2.3% (1/44) Empedobacter breve 2.3% (1/44) Erwinia carotovora 27.3% (12/44) Erwinia herbicola 2.3% (1/44) Janthinobacterium lividum 2.3% (1/44) Pantoea agglomerans 2.3% (1/44) Pseudomonas cichorii 2.3% (1/44) Pseudomonas fluorescens 40.9% (18/44) Pseudomonas putida 2.3% (1/44) Serratia liquefaciens 4.6% (2/44) Serratia marcescens 2.3% (1/44) Xanthomonas campestris 4.6% (2/44) Acinetobacter spp. 5% (2/40) Hamilton-Miller and Enterobacter cloacae 2.5% (1/40) Shah 2001 Escherichia vulneris 2.5% (1/40) Pantoea agglomerans 27.5% (11/40) Pantoea dipsersa 5% (2/40) Pseudomonas fluorescens 47.5% (19/40) Pseudomonas luteola 2.5% (1/40) Pseudomonas oryzihabitans 2.5% (1/40) Rahnella aquatilis 2.5% (1/40) Sphingobacter multivorum 2.5% (1/40) Stenotrophomonas maltophilia 2% (2/40)

Pseudomonas fluorescens is a pectolytic, fluorescent pseudomonad which is responsible for a substantial proportion of soft-rot disorders in vegetables (Liao and Wells 1987;

Liao 2006). The importance of pectolytic fluorescent pseudomonads as a leading cause of spoilage of refrigerated produce is primarily due to their psychrotrophic nature, nutritional diversity, and predominant presence on the surfaces of fresh produce

(Ceponis and Friedman 1958; Brocklehurst and Lund 1981; Tekoriené 2003). The presence of pectolytic fluorescent pseudomonads at high populations on the surface of lettuce may lead to decay of fresh lettuce and vegetable salads, causing a shorter shelf- life of these products. Other species of Pseudomonas and Enterobacteriaceae¸ including the phytopathogenic bacteria such as Ps. cichorii, Ps. viridiflava, Ps. marginalis, Erw. carotovora and Erw. amylovora can also cause rot in vegetables due to the presence of pectate lysase (Lund 1983; Liao and Wells 1987; Gallois et al. 1992; Liao et al. 1993).

Other tissue degrading enzymes produced by species of Pseudomonas include cellulases, xylanases, glycoside hydrolases and lipoxygenase (Gross and Cody 1985;

Zhuang et al. 1994). Biosurfactants can be produced by 50% of strains of Ps. fluorescens A, Ps. viridiflava, Ps. mendocina and Ps. fragi (Padaga et al. 2000), which can assist in the bacterial colonisation and degradation of plant tissue.

The family Enterobacteriaceae consists of a number of Gram negative rod shaped bacteria of which the genera Erwinia, Enterobacter and Pantoea are members. The genus Pantoea consists of several species which were formerly classified in the genera

Enterobacter and Erwinia (Gavini et al. 1989; Hauben et al. 1998).

Bennik et al. (1998) described the presence of several species of Enterobacteriaceae on fresh vegetables which included Ent. cloacae, P. agglomerans, Rahnella aquatilis, Erw. carotovora, Erw. amylovora, Kl. oxytoca and Ser. odifera. Species of Pantoea, Erwinia and Enterobacter are frequently isolated from the surface of raw and processed

vegetables (Wright et al. 1976; Riser et al. 1984; Brocklehurst et al. 1987; Magnuson et al. 1990; Bennik et al. 1998; Hamilton-Miller and Shah 2001). These species are considered to cause spoilage of ready-to-eat vegetables upon storage and refrigeration due to their ability to grow as facultative anaerobes and survive in the modified atmosphere of packaged salads (Bennik et al. 1998), as well as their ability to ferment glucose to produce acids, alcohols and esters (Adams and Moss 1995). The role of these organisms in spoilage of fresh cut products is not well understood and is an area for future research (Heard 2002). Possibly, their potential to grow and spoil the produce might be increased if low storage temperatures are not properly maintained.

Overall, data on the microbial ecology of lettuce produce are not comprehensive, and more quantitative information is needed about the populations (cfu/g) of the individual species present. The total ecology not only affects sensory and shelf-life properties of the product, but the interactive behaviour of microorganisms may also affect the survival and growth of organisms of public health significance.

After processing, the total populations associated with minimally processed lettuce range between 104-106 cfu/g (Nguyen-the and Carlin 2000; Heard 2002; Delaquis

2006). The predominant microbial species associated with these products are similar to those found in whole, unprocessed vegetables (Nguyen-the and Carlin 1994; Nguyen- the and Carlin 2000). As shown in Table 2.2, species representing Pseudomonas,

Enterobacter, Erwinia and Serratia usually prevail over other genera. Among the pseudomonads, most isolates fall into the species Ps. fluorescens and less frequently Ps. putida and Ps. chlororaphis (Nguyen-the and Prunier 1989; Marchetti et al. 1992;

Jacques and Morris 1995). Psychrotrophs are an important part of the microflora (Garg et al. 1990; Barriga et al. 1991) as these species can also be pectinolytic. Populations of pectinolytic microorganisms have been determined by various authors as potential

spoilage agents, and found to represent 10% to 20% of isolates among mesophilic bacteria from shredded lettuce (Magnuson et al. 1990).

Yeast and moulds are often isolated at populations ranging from 104-105 cfu/g

(Brocklehurst et al. 1987; Garg et al. 1990; Tournas 2005), with the main species identified as Alternaria, Candida, Cladosporium, Cryptococcus laurentii, Geotrichum,

Metschnikowia pulcherrima, Penicillium, Rhodotorula, Trichosporon capitatum and

Trichosporon cutaneum (Magnuson et al. 1990; Marchetti et al. 1992, Tournas 2005).

Table 2.2 Predominant bacterial species associated with processed lettuce and ready-to-eat salads

Product Bacterial Species Reference Minimally processed Burkholderia cepacia Freire and Robbs hydroponic lettuce Enterobacter cloacae 2000 Escherichia coli Klebsiella oxytocica Pseudomonas aeruginosa Pseudomonas fluorescens Serratia marcescens Stenotrophomonas maltophilia Minimally processed Erwinia herbicola Magnuson et al. lettuce Pseudomonas fluorescens 1990 Pseudomonas putida Mixed salads Bacillus spp. Brocklehurst et al. Cytophaga spp. 1987 Erwinia carotovora Pantoea agglomerans Pseudomonas fluorescens Pseudomonas putida Vegetable salads Citrobacter diversus Wright et al. 1976 Citrobacter freundii Enterobacter aerogenes Enterobacter cloacae Escherichia coli Klebsiella aerogenes Klebsiella edwardsii Klebsiella pneumoniae Pantoea agglomerans Proteus spp. Pseudomonas aeruginosa Serratia liquefaciens Serratia marcescens Serratia rubideae Vegetable salads Brevundimonas vesicularis Marchetti et al. Burkholderia cepacia 1992 Chromobacterium violaceum Chryseomonas luteola Enterobacter cloacae Pseudomonas aeruginosa Pseudomonas aureofaciens Pseudomonas chlororaphis Pseudomonas fluorescens Pseudomonas mesophilica Pseudomonas pickettii Rahnella aquatilis Sphingomonas paucimobilis Stenotrophomonas maltophilia

Lettuce and other raw vegetables have been linked to outbreaks of foodborne microbial disease, and foodborne pathogens may form part of the microflora of these products.

Although, fruits and vegetables have been recognized as potential sources of infectious microbial agents, there has been very little evidence, until recently, documenting their risks to public health. Increased consumption of fresh produce and the growth of the fresh-cut industry have prompted interest to investigate the association of pathogens with these products.

In recent years, numerous papers have discussed the occurrence of foodborne pathogens on fresh produce (Fain 1996; Beuchat 1996a, 1996b, 1998, 2002; Francis et al. 1999;

Nguyen-the and Carlin 2000; Heard 2002). These reports along with an extensive review by the IFT (2003) have highlighted the gaps in knowledge regarding the organisms of concern, and emphasised the need for a full risk assessment on vegetable production from the farm through to the consumer. It has been accepted that the pathogens of concern in fresh-cut salads are similar to those present on raw vegetables

(Nguyen-the and Carlin 2000). Table 2.3 provides a summary of reports on the isolation of pathogenic bacteria from lettuce and other salad vegetables. There are frequent reports of the isolation of Salmonella spp. and L. monocytogenes, followed by less reports of E. coli, Campylobacter spp., Aeromonas spp. and Yersinia enterocolitica.

A recent review by Sivapalasingum et al. (2004) highlighted that, among individual fresh produce, lettuce was the product most frequently associated with foodborne disease outbreaks. Of these outbreaks, 60% were related to bacterial pathogens, 20% to viruses, 16% to parasites and the remaining 4% to chemicals and poisons.

Table 2.3 Detection of pathogenic bacteria in lettuce and ready-to-eat salads

Pathogen Product Country Incidence Reference Aeromonas spp. raw vegetables India 14/30 Puttalingamma and Manja 1998 Aeromonas spp. raw vegetables Belgium 7/27 Neyts et al. 2000 Aeromonas spp. minimally Australia 66/120 Szabo et al. 2000 processed lettuce Aeromonas hydrophila lettuce Brazil 2/30 Saad et al. 1995 Aeromonas hydrophila lettuce Italy 3/20 Villari et al. 2000 Campylobacter spp. lettuce Canada 2/165 Park and Sanders 1992 Campylobacter spp. lettuce UK 0/151 Little et al, 1999 Escherichia coli organic lettuce Norway 12/179 Loncarevic et al. 2005 Escherichia coli vegetable USA 8/63 Lin et al. 1996 salads Escherichia coli O157:H7 lettuce UK 0/151 Little et al. 1999 Listeria monocytogenes lettuce Malaysia 1/28 Tang et al. 1994 Listeria monocytogenes lettuce Sri Lanka 10/20 Francis et al. 1999 Listeria monocytogenes lettuce US 1/92 Heisick et al. 1989 Listeria monocytogenes lettuce Norway 1/200 Johannessen et al. 2002 Listeria monocytogenes leafy Malaysia 5/22 Arumugaswamy et al. vegetables 1994 Listeria monocytogenes organic lettuce Norway 2/179 Loncarevic et al. 2005 Listeria monocytogenes raw vegetables India 18/116 Pingulkar et al. 2001 Listeria monocytogenes raw vegetables Spain 8/103 de Simón et al. 1992 Listeria monocytogenes minimally Australia 3/120 Szabo et al. 2000 processed lettuce Listeria monocytogenes vegetable USA 1/63 Lin et al. 1996 salads Listeria monocytogenes vegetable Spain 21/70 Garcia-Gimeno et al. salads 1996 Listeria monocytogenes vegetables USA 22/2966 Gombas et al. 2003 salads Salmonella spp. lettuce Italy 82/120 Ercolani 1976 Salmonella spp. lettuce Netherlands 2/28 Tamminga et al. 1978 Salmonella spp. lettuce Spain 5/80 Garcia-Villanova et al. 1987 Salmonella spp. lettuce UK 0/151 Little et al. 1999 Salmonella spp. ready-to-eat UK 6/3852 Sagoo et al. 2001 vegetables Salmonella spp. various Egypt 2/71 Saddik et al. 1985 vegetables Vibrio cholerae lettuce UK 0/151 Little et al. 1999 Yersinia enterocolitica raw vegetables Brazil 1/30 Tassinari et al. 1994 Yersinia enterocolitica lettuce Norway 6/200 Johannessen et al. 2002

Species of Acinetobacter represent another group of pathogens which has been overlooked in past reviews. These bacteria occur at low populations (50-1000 cfu/g) in 17-

50% of vegetable samples, with Ac. baumannii, Ac. calcoaceticus and Ac. johnsonii

being the most frequently isolated species (Berlau et al. 1999a; Houang et al. 2001).

Lettuces, which are the main constituent of ready-to-eat salads, have a low inherent population of Acinetobacter species, leading researchers to propose that uncooked vegetables are a potential source of nosocomial infections to vulnerable patients, and that multiple antibiotic resistance is common in epiphytic bacteria as found on lettuce

(Jiwa et al. 1981; Khan et al. 1992; Hamilton-Miller and Shah 2001).

Non-bacterial pathogens may also be transmitted by fresh produce. Frequently, these originate from the irrigation water and are transmitted to the food during watering of the crops or during postharvest washing. These include viruses such as the Norwalk virus and Hepatitis A, and the parasites, Giardia, Cryptosporidium and Cyclospora.

Rotavirus and Hepatitis A have been associated with lettuce, chopped tomatoes, and strawberries (Hernández et al. 1997; Beuchat 1998; Seymour and Appleton 2001).

Consumption of food contaminated with such pathogens can give a diversity of disease symptoms, including respiratory infection, skin disorders, meningitis and gastroenteritis

(Grohmann 1997).

Viruses can survive after packaging at refrigerated temperatures (4°C), with poliovirus and Hepatitis A surviving on fresh produce for up to 14 days. Lettuce was observed to have greater potential for contamination with viruses due to the larger exposure of surface area and the protection of the organism in cracks and crevices naturally formed during lettuce growth (Kurdziel et al. 2001; Croci et al. 2002). These same concepts also apply to bacterial pathogens.

Parasites are defined as eukaryotic organisms and may be classified into two main groups, protozoa and helminths. Parasites are dependent on host organisms for survival and although their life cycles vary, they must all pass through animal or human host to survive and reproduce (Goldsmid and Speare 1997). They may infect food from

contaminated water or sewage, from food handlers, insects or the parasite may be ingested by animals and be present in animal flesh at the time of slaughter.

Cyclospora, Cryptosporidium and Giardia have been isolated from vegetables including lettuce, cucumber, carrots, sprouts and strawberries (Monge and Chinchilla 1996;

Herwaldt 2000; Robertson et al. 2000; Capuano et al. 2001; Robertson and Gjerde

2001).

2.5 Significance of microflora of lettuce and ready-to-eat salads

The implications of microbial contamination and growth on lettuce and other vegetable produce have been briefly mentioned already. Spoilage, decreased sensory appeal, and decreased shelf life lead to loss and wastage of product that have significant economic consequences. The microbiological safety of these products has also become a significant issue, as the incidence of food borne disease outbreaks associated with their consumption has increased. These spoilage and public health risks are not only determined by contamination with the appropriate species, but can also be moderated by microbial interactions that are now known to occur within each specific ecosystem.

Many microbial species which are associated with the pre-harvest microflora of lettuce and other raw vegetables cause spoilage and loss of product at both pre-harvest and post-harvest stages (Nguyen-the and Carlin 2000; Heard 2002). Spoilage of fresh cut vegetables by bacteria is characterised by brown discolouration, production of off odours, loss of texture and, to a lesser extent, soft rot (Liao and Wells 1987; Zhuang et al.1994; Padaga et al. 2000; Heard 2002). The presence of spoilage microorganism may also shorten the shelf-life of ready-to-eat vegetable products therefore leading to inferior products incapable of being sold (Nguyen-the and Carlin 1994; Brackett 1997; Heard

1999; Heard 2002). It is estimated that between 10 and 30% of fresh fruits and vegetables produced are wasted, mainly due to three factors: mechanical injuries, physiological decays and microbial spoilage (Harvey 1978). The role of microorganisms in this wastage is significant, highlighting a need to better understand the bacterial ecology of lettuce and other fresh cut vegetables.

The presence of pathogenic microorganisms on vegetables can be a cause of human diseases, with associated social and economic consequences. The ecological data presented previously in Table 2.3 are supported by increasing reports of outbreaks of food borne illness attributed to the consumption of salad produce, especially those containing lettuce (Table 2.4). Although definitive forensic investigations have been difficult to complete, it is suspected that contamination of the produce at the preharvest stage is a major causative factor in these outbreaks.

Table 2.4 Foodborne disease outbreaks linked to consumption of lettuce and other raw salad vegetables

Isolated No. of No. of Pathogen/Virus Year Location Vehicle from Reference cases deaths vehicle Escherichia coli 2006 USA Spinach 183 1 Yes CDC 2006 O157:H7 Salmonella 2001 Queensland, Lettuce 41 0 Yes Stafford et al. 2002 Bovismorbificans Australia Salmonella enterica 2000 England & Lettuce 361 0 No Horby et al. 2003 ser. Typhimurium Wales DT104 Enterohaemorrhagic 1999 Sweden Lettuce 37 0 Yes Welinder-Olsson Escherichia coli et al. 2003 Escherichia coli 1999 Texas, USA Salad 58 0 No Bergmire-Sweat et O111:H8 al. 2000 Salmonella 1999 California, Cilantro 41 0 Yes Campbell et al. Thompson USA 2001 Escherichia coli 1998 Minnesota, Parsley 77 0 Yes Naimi et al. 2003 USA Yersinia 1998 Finland Lettuce 47 1 No Nuorti et al. 2004 pseudotuberculosis Escherichia coli 1996 Connecticut & Lettuce 48 0 No Hilborn et al. 1999 O157:H7 Illinois, USA Escherichia coli 1995 Montana, Lettuce 52 0 No Ackers et al. 1998 O157:H7 USA Escherichia coli 1995 USA Lettuce 21 0 No Davidson et al. O157:H7 1996 Escherichia coli 1995 Ontario, Lettuce 21 0 No Preston et al. 1997 O157:H7 Canada Shigella sonnei 1994 Sweden Lettuce 100 0 No Frost et al. 1995 Shigella sonnei 1994 Norway, Lettuce 118 0 No Kapperud et al. Sweden, UK 1995 Escherichia coli 1993 Rhode Island Carrot 168 0 No Benoit et al. 1994 O6:NM & New Hampshire, USA Escherichia coli 1992 Maine, USA Vegetables 4 1 No Cieslak et al. 1993 O157:H7 Hepatitis A 1988 Kentucky, Lettuce 202 0 No Rosenblum et al. USA 1990 Shigella sonnei 1986 Texas, USA Lettuce 347 0 No Davis et al. 1988 Shigella sonnei 1983 Texas, USA Lettuce NR 0 No Martin et al. 1986 Listeria 1981 Maritime Vegetable 41 17 Yes Schlech et al. 1983 monocytogenes Providence, mix for Canada coleslaw Listeria 1979 Boston, USA Raw 20 5 No Ho et al. 1986 monocytogenes tomatoes, lettuce & celery Modified from Harris et al. (2003); NR- Not reported

Although still few in number, there are increasing studies showing how the total microbial ecology of plant produce can affect spoilage and public health outcomes.

The growth and survival of pathogenic bacteria on salad produce may also be moderated by the total ecology. In some studies, the growth and survival of Salmonella and L. monocytogenes were found to be enhanced by the presence of Erw. carotovora or

Ps. fluorescens (Carlin et al. 1995; Wells and Butterfield 1999). In other studies, however, populations of L. monocytogenes were decreased by the presence of postharvest rot pathogens. For example, the growth of L. monocytogenes on potato slices (Liao and Sapers 1999), spinach (Babic et al. 1996), and endive (Carlin et al.

1996) could be markedly reduced by diverse strains of fluorescent pseudomonads. The inhibition was thought to be caused by the production of iron-chelating fluorescent siderophores or antimicrobials by the pseudomonads (Liao and Fett 2001).

The application of biological control to crops while they are growing may also affect the microbial ecology of plant produce. However, field application of antagonists will expose them to adverse environmental conditions such as desiccation and solar radiation, which they will have to withstand in order to be effective against spoilage species of bacteria (Mercier and Marrone 2006).

Many factors will affect the interactive responses of bacterial populations associated with vegetable produce, highlighting the need for more comprehensive understanding of their overall microbial ecology.

2.6 Sources of microbial contamination of vegetable produce

Microbial contamination of raw vegetables may occur at any stage from when the vegetable is planted through to the harvest and processing in the factory. It has been suggested that the most important factors contributing to the microbiological quality and safety of the finished product occur at the preharvest stage (Lund 1992; Beuchat 1996;

Brackett 1999; Beuchat 2002; Suslow et al. 2003). Vegetables are grown and harvested in environments that are exposed to multiple sources of microbial contamination, with the main sources being air, irrigation water, soil, fertiliser, handlers, manure, animals and birds (Beuchat 1996a; Brackett 1999; Beuchat 2002). These sources are covered in more detail in the following sections.

2.6.1 Irrigation water

Irrigation water can be a significant source of contamination of vegetables. The impact of the water on the microbial population of vegetables will depend on (1) the source and microbial load of the water, (2) the method of irrigation and (3) the frequency of irrigation. The source of the water could be a lake, bore, river, mains, dam, a reticulated municipal supply, or treated sewage water. There is a major diversity in the species and population of microorganisms associated with these sources, with total populations ranging from 101-105 cfu/ml (Robinson and Adams 1978; Rosas et al. 1984; Falcão et al. 1993; Wang and Doyle 1998; Riordan et al. 2001). These water sources are usually used for irrigation and dilution of agrochemicals and vary according to the geographical location of the farm and supply availability. Mains or potable water is not generally used for these purposes because of cost.

Apart from contributing general microbiological flora, irrigation water can also be a source of pathogenic bacteria, viruses and parasites. Listeria monocytogenes (Watkins and Sleath 1981), Campylobacter jejuni (Bolton et al. 1987; Korhonen and Martikainen

1991), Salmonella (Watkins and Sleath 1981; Vaz da Costa-Vargas et al. 1991; Ait

Melloul and Hassani 1999; Pianetti et al. 2004), Clostridium perfringens (Watkins and

Sleath 1981) and E. coli (Robinson and Adams 1978; Watkins and Sleath 1981; Vaz da

Costa-Vargas et al. 1991; Falcão et al. 1993; Riordan et al. 2001; Steele et al. 2005) have been isolated from various sources of agricultural waters. The contamination of these water sources is usually attributed to faecal contamination by local wildlife

(Ackers et al. 1998; Hilborn et al. 1999; Pianetti et al. 2004; Steele and Odumeru 2004;

Steele et al. 2005) and possibly from human effluents if they are not properly treated.

Contamination of vegetables with E. coli O157:H7 and Salmonella spp. from water has been demonstrated by being drawn into the plant tissue via the root system, from where the organisms can migrate into the edible tissues of the vegetable. Consequently, water quality is of paramount importance for the food safety of growing crops (Solomon et al.

2002b; Wachtel et al. 2002a, 2002b; Okafo et al. 2003; Steele and Odumeru 2004;

Steele et al. 2005).

2.6.2 Irrigation method

Several methods are used to irrigate salad vegetables, depending on the geographical location of the farm and type of vegetable. These include drip, spray, flood and furrow methods. Depending upon the type of irrigation, variation in the contamination of the crops of vegetables has been observed.

Irrigation of crops results in an increase in the water activity of the vegetable surface, thereby encouraging microbial growth and survival (Perombelon et al.1979; Solomon et al. 2002a). The use of spray irrigation can lead to an increase in contamination of lettuces and capsicum with E. coli, Salmonella and Hepatitis A (Solomon et al. 2002a;

Stine et al. 2005) and disperse the microorganisms up to 350m away from the source

(Katzenelson and Teltch 1976; Katzenelson et al. 1977; Sadovski et al. 1978a, 1978b;

Bastos and Mara 1995). Increased frequency of irrigation increases the potential for microbial contamination and growth, especially when using spray irrigation (Bryan

1977; Sadovski et al. 1978a 1978b; Armon et al. 1994; Bastos and Mara 1995; Solomon et al. 2002a).

Drip and furrow irrigation methods are likely to give less microbial contamination than spray irrigation. This is due to delivery of irrigation water straight onto the soil and roots of the crops with no formation of aerosols or spraying of the upper surface area of the plants (Volcani 1969; Bastos and Mara 1995). However, greater contamination of crops of lettuce and cantaloupe were observed for furrow irrigation when compared to a sub-surface drip irrigation system, using E. coli, Cl. perfringens and coliphage as test organisms (Song et al. 2006).

2.6.3 Soil

Soil is a potential source of vegetable spoilage microorganisms as well as human pathogens such as L. monocytogenes, Clostridium, Salmonella, and E. coli O157

(Welshimer 1960; Van Renterghem et al. 1991; MacGowan et al. 1994; Dowe et al.

1997; Ogden et al. 2001; Guo et al. 2002; Jiang et al. 2002; Ogden et al. 2002). Wind blown dust is the main mechanism by which soil comes into contact with salad

vegetables (Vakili 1967; Claflin et al. 1973). Vegetables may also become contaminated with microorganisms by direct contact with soil that has splashed onto the plant surface during periods of rain or irrigation (Melick 1917; Rudolfs et al. 1951; Guo et al. 2002).

2.6.4 Fertilisers and Manure

Many different types of fertilisers and manures are used in the cultivation of vegetables.

These include chemical fertilisers (nitrogen, phosphorus and potassium (NPK)), composted fertilisers, organic fertilisers (fresh manure) and human faecal material

Fertilisers and manures have been mentioned by many researchers to be a major source of contamination of vegetables at the preharvest stage (Bryan 1977; Roberts et al. 1982;

Doyle 1990; Park and Sanders 1992; Suslow et al. 2003). The potential for contamination of crops by fertilisers and manure is dependent upon i) the type of manure or fertiliser, ii) the frequency of application and iii) the time of application. All of these factors will depend upon the farming practices used during cultivation of the vegetables. For example, if fertiliser or manure is applied to the vegetable crop close to harvest, there is a greater chance of survival of organisms, including pathogens, through to market (Cieslak et al. 1993; Natvig et al. 2002; Suslow et al. 2003; Ingham et al.

2004; Johannessen et al. 2004; Hutchinson et al. 2005).

Microorganisms have been found to survive for various periods of time in fertilisers and manure, depending upon the temperature (Strauch 1991; Wang et al. 1996:

Himathongkham et al. 1999; Jiang et al. 2002; Hutchison et al. 2004). Campylobacter

(Blaser et al. 1980), Salmonella (Watkins and Sleath 1981; Himathongkham et al.

1999), Listeria (Watkins and Sleath 1981; Skovgaard and Morgen 1988), and E. coli

O157:H7 (Wang et al. 1996; Kudva et al. 1998; Beuchat 1999; Hornitzky et al. 2000;

Duffy 2003; Johannessen et al. 2005), can be present in animal manures and contaminate vegetable crops. Application of inadequately treated animal manures is a major cause of foodborne disease outbreaks and led to formulation of new regulations in the USA and other countries about the proper treatment of animal waste before application to vegetables (Schlech et al. 1983; Cieslak et al. 1993; Suslow et al. 2003).

2.6.5 Pesticides

Pesticides are routinely applied in the cultivation of vegetables. They are used to control various spoilage bacteria and fungi, insects, weeds and other pests during the preharvest stage of production. Pesticides are commonly grouped into three categories: insecticides, fungicides and herbicides. These categories have different target populations, active ingredients and usually come in a concentrated powder or liquid form. Reconstitution with water is required before application to vegetable crops.

Procedures for reconstitution of pesticides depend upon the grower, geographical location and access to various water sources. Reconstituted pesticides can be a suitable environment for the survival and growth of microorganisms, including pathogens such as Salmonella, Shigella, E. coli O157:H7 and L. monocytogenes (Coghlan 2000; Guan et al. 2001). Consequently, application of pesticide solutions to vegetables could provide an unintentional means of contamination of vegetables with viable microorganisms. It is thought that a large outbreak of food borne illness from raspberries contaminated with the protozoan parasite, Cyclospora cayetanensis, occurred through the use of pesticides reconstituted in contaminated water (Herwaldt and Ackers 1997).

The pesticide, itself, may also affect the growth and survival of microorganisms already associated with the plant surface. Possibly, it could kill some microflora, or stimulate the growth of other microflora. It has been observed that some pesticide solutions can decrease the population of fungi, bacteria and yeast for short periods of time (15-20 days) upon application to raw foods (potato, wheat, apple leaves and fruit, and pears)

(Andrews and Kenerley 1978; Meunier and Meyer 1985; Shukla et al. 1988; Calvente et al. 1999). However, pesticides, such as maneb and carbaryl, were found to have no effect on the microflora of vegetables (Mercier and Reeleder 1986).

It has emerged from the studies of Guan et al. (2001) and Coghlan (2000) that agrochemical pesticide application to produce could be an underestimated source of contamination with spoilage and pathogenic microorganisms, and that this possibility requires more detailed investigation. Further investigations, carried out after completion of the current project, examined the survival of pathogenic bacteria in pesticide solutions and on treated tomato plants (Guan et al. 2005). Salmonella spp was the best able to survive and Listeria spp. were least able to survive in pesticide solutions. When applied to growing tomato plants, E. coli and Salmonella spp. were observed to survive for 45h and 15 days, respectively. This data demonstrates that contaminated pesticides are a source of contamination of vegetables while they are growing if contaminated water is used for reconstitution.

2.6.6 Pests and Farm Animals

Pests are another source of contamination of vegetables, and includes birds, animals

(native, farm or domestic), and insects (Lund 1992; Madden 1992; Brackett 1999;

Nguyen-the and Carlin 2000). The occurrence of various pests around vegetable crops

will depend on the type of vegetable, geographical location and whether the grower has a segregated section for any animals. Contamination of vegetable crops by pests and farm animals is usually by direct contact between the vegetable and their faeces or due to contamination of irrigation water with their faeces.

2.6.6.1 Birds

Bird faeces is a known source of bacterial contaminants, including food borne pathogens such as Salmonella spp. (Jones et al. 1978; Fenlon 1981, 1983, 1985;

Kapperud and Rosef 1983), C. jejuni (Luechtefeld et al. 1980; Kapperud and Rosef

1983), Yersinia spp. (Kapperud and Rosef 1983) and E. coli O157:H7 (Wallace et al.

1997). Bird faeces may contaminate irrigation water supplies and may be dropped onto produce. Contamination of vegetable crops by birds is plausible as seagulls have been found to contaminate drinking water with their faeces (Jones et al. 1978). Birds may also contaminate crops by attacking the vegetables while they are growing, thus damaging the surface structure and allowing microorganisms to access the internal tissues.

2.6.6.2 Native, domestic and farm animals

Native, domestic and farm animals are a potential source of preharvest contamination of vegetables (Beuchat 1996a). The probable mode of contamination of the vegetable is by direct contact of the faecal material or contamination of soil or contamination of irrigation waters. Foodborne disease outbreaks have been attributed to the presence of animals in close vicinity of the growing vegetable crops (Besser et al. 1993; Millard et al. 1994; Hilborn et al. 1999). Faeces of bovine or human origin contaminated with E. coli O157:H7 was found to be a contaminant of drinking water (Jackson et al. 1998), thus demonstrating another way by which vegetables may be contaminated. Domestic

animals such as dogs and cats have been found to be sources of Campylobacter spp., as well as all farm animals except hens (Svedham and Kaijser 1981). This finding presents a potential source of contamination of the vegetables while they are growing due to direct contact with the faeces of these animals.

2.6.6.3 Insects and worms

Insects are important in the transfer of bacteria on and among field plants (Lund 1992).

Flies are one of the most studied insects, with Shigella, Salmonella and E. coli O157:H7 capable of being carried and transmitted to humans and food crops (Bidawid et al. 1978;

Cohen et al. 1991; Iwasa et al. 1999; Janisiewicz et al. 1999; Olsen and Hammack

2000). Reduction of houseflies has been found to reduce the incidence of shigellosis

(Cohen et al. 1991). Sela et al. (2005) presented data supporting the hypothesis that the common Mediterranean fruit fly was a potential vector of human bacterial pathogens to fruit. Using a fluorescent labelled strain of E. coli, contaminated flies were found to contaminate intact apples in a cage model system. Wild flies, captured at different geographical locations, harboured coliforms, with some presumptive identification of E. coli.

Fruits and vegetables may also be contaminated by worms (Caenorhabditis elegans) that are present in the soil. Salmonella Newport, present in manure and manure compost, was transferred by worms and detected upon the surface of lettuce, strawberry and carrot samples at 3, 1, and 1 days, respectively (Kenney et al. 2006).

2.6.7 Handling operations

Vegetables can become contaminated during harvest and postharvest handling and packing. Main sources include farm personnel and contact with equipment surfaces,

such as packaging crates and pallets and from trucks during transport. Improper hygiene practices may influence the microbial safety of produce during harvest (Geldrich and

Bordner 1971; Suslow et al. 2003). Facilities should be provided by the farmers that include adequate toilet and hand washing facilities, with proper removal of sewage so as to avoid contact with crops. Cross-contamination between different crops is possible due to the same handler or polluted wash water (Lund 1992; ICMSF 1998). Control measures include the use of clean water and sanitisers for the washing of vegetables and cleaning of work surfaces, refrigeration in packing sheds and training of agricultural workers in good manufacturing and hygiene practices (Suslow et al. 2003; Greig et al.

2007).

2.7 Methods for studying the microbial ecology of lettuce and related produce

The methods for studying the microbial ecology of lettuce and related produce consist of culture-dependent and culture-independent analyses. Culture-dependent methods rely on the growth of microorganisms in liquid or agar media, whereas culture-independent methods are based on the use of molecular methods that analyse some aspect of microbial DNA. These two types of analysis will be discussed in more detail in the following sections.

2.7.1 Culture-dependent microbiological analysis of vegetables

The standard cultural approach for studying the bacterial ecology of vegetables starts with rinsing or macerating a representative sub-sample of the product, followed by plating onto an appropriate agar medium. For the detection and isolation of species likely to be present at very low populations, it is necessary to culture sub-samples in an appropriate liquid enrichment medium, followed by agar plate culture (Holbrook 2000;

Beuchat 2006). Common media used to isolate bacteria from vegetables include plate count agar (PCA), typtone soya agar (TSA), and nutrient agar (NA) (Koek et al. 1983;

Saddik et al. 1985; King et al. 1991; Albrecht et al. 1995; Thunberg et al. 2002).

Limitations of these cultural approaches that can lead to underestimations of the populations and species present include: inhibitory effects of plant extracts in macerates or rinses on bacterial survival; loss of organism viability during dilution and during the time span between dilution and plating; failure of the chosen culture medium to provide sufficient nutrients for growth of all the species present; failure of enrichment cultures

to amplify to a desired population; and poor anaerobic methodology (Fleet 1999;

Beuchat 2006). It has also been recognized that viable but non-culturable (VBNC) microorganisms may be present on vegetables due to the impacts of many environmental stresses and, therefore, not detected by conventional methods (Muyzer and Smalla 1998; Fleet 1999; Giraffa 2004; De Vero et al. 2006).

The impact of the food matrix on measurement of the bacterial ecology could be particularly significant in the case of vegetable produce because it is well recognised that many of these products contain constituents that have antimicrobial properties

(Fleet 1999; Beuchat 2006). Burnett and Beuchat (2001), compared washing, stomaching, and homogenising in 0.1% peptone for their influence on the recovery of

Salmonella inoculated onto 26 types of vegetables, fruits and herbs and found that no significant differences could be attributed to a particular preparation method. However, on examination of the tissue pH, a reduced percent recovery from herbs was attributed to the release of antimicrobials other than acids, during sample preparation. Maceration of samples may also inadvertently cause injury to some microorganism. In a study by

Han et al. (2004), strawberries that has been dip or spot inoculated with E. coli

O157:H7 and L. monocytogenes and stomached, resulted in an injury of 0.9 to 1.4 log cfu for E. coli O157:H7 and 1.4 to 1.7 log cfu for L. monocytogenes. In comparison, a simple rinse step resulted in an injury of only 0.2 to 0.6 log cfu for E. coli O157:H7 and

0.2 to 0.7 log cfu for L. monocytogenes. While blending, homogenising, or macerating may be acceptable for some types of fruits and vegetables, a simple rinsing without rupturing of plant cells may be required for other types (Han et al. 2004; Beuchat 2006).

With rinsing methods there is always the risk that a proportion of the microbial population might not be dislodged from the product surface and, therefore, escape detection (Lillard 1988b). Rinsing involves the shaking of a sample with diluent on a

rotary shaker or by some other method for varying periods of time, often 15-30 mins.

During this time, microorganisms attached to the surface of the sample are gently removed into solution, while minimising damage to the plant material (Burnett and

Beuchat 2001). The efficacy of rinsing on the removal of cells from the sample matrix has led to variable results, with rinsing being observed to recover higher or similar populations to blending or homogenising (Burnett and Beuchat 2001; Han et al. 2004) or having a lower recovery of total populations (Jay and Margitic 1979; Lillard 1988b;

Nedoluha et al. 2001). Such conclusions suggest that, for some products such as vegetable produce, it would be prudent to use both rinse and maceration methods when aiming to establish their basic microbial ecology.

Various surfactants or detergents may be added to diluents to facilitate detachment of bacterial cells by rinsing procedures. Surfactants used in this context include Tween 80, sodium dodecyl sulphate (SDS), sodium lauryl sulphate, and Triton X (Cheng-An and

Beuchat 1995; Raiden et al. 2003; Sapers 2006). These agents are selected to remove microorganisms from the surface of fruits and vegetables, and not have any antimicrobial action (Sapers 2006). The presence of surfactant in the diluent reduces the surface tension of the solution, favouring wetting of the surface and allowing better contact between the diluent and microorganisms (Bastos et al. 2005). Raiden et al.

(2003) investigated the efficacy of water, Tween 80 and sodium lauryl sulphate in removing inoculated species of Salmonella and Shigella from the surface of strawberries, tomatoes and leaf lettuce. High removal rates were obtained for all samples, and it was concluded that the detergents were no more effective than water.

However, this outcome may have been a reflection of the brief time interval between inoculation and treatment of the samples and, therefore, further research on this topic is still required. According to most authors, Tween 80 is the best choice of surfactant, due

to the enhanced removal of microorganisms from the sample matrix and least effect on microbial viability (Lillard 1988a; Hwang and Beuchat 1995; Eginton et al. 1998; Yu et al. 2001; Raiden et al. 2003; Bastos et al. 2005). However, this conclusion only applies to its use for this purpose at low concentrations (e.g. 0.1%), because at high concentrations (e.g. 1%) it could be used to decontaminate salad vegetables (Adams et al. 1989).

Given the limitations and variables associated with culturable methods, there is always a risk that they can lead to an inaccurate profile of microorganisms in the habitat, and give an oversimplified, under-estimation of the total ecology of vegetables. To overcome these limitations, culture-independent, molecular techniques have been developed.

2.7.2 Molecular strategies for monitoring bacterial communities

Molecular methods are now routinely applied in the identification of bacterial isolates to genus and species, and to differentiate strains at sub-species level, as well as being used to detect the presence of species in habitats and ecosystems. These methods are considered to be culture-independent and have the advantages of being able to detect and identify microorganisms that are present. They are also considered to be faster than conventional methods and are able to obtain a qualitative and quantitative picture of the microbial community when applied to specific habitats or ecosystems (Giraffa 2004).

Bacterial identification and typing are usually achieved by amplification of the 16S rDNA of the isolate by PCR, followed by sequencing and comparison of the sequence data to a database which contains ribosomal sequences for thousands of known bacterial species. This approach allows phylogenetic affiliation to cultured, as well as uncultured

microorganisms, and is now routinely used to identify isolates from plant habitats, including vegetable produce.

In addition to routine identification tasks, molecular methods are being increasingly used to determine and monitor microbial communities in natural habitats (Giraffa and

Neviani 2001; Rudi et al. 2002; Bhagwat 2004; Handschur et al. 2005; Johnston et al.

2005). Initially, environmental ecological studies used fluorescent-in situ-hybridisation

(FISH) where whole fixed cells could be easily identified and counted directly using genus or species-specific nucleic acid probes (Amann and Kuhl 1998; Maszenan et al.

2000). The method consists of : (i) sample preparation and cell fixation, usually by paraformaldehyde; (ii) sample immobilisation onto microscopic slides, usually Teflon coated; (iii) cell treatments to increase permeability to the probe; and (iv) in situ hybridisation with fluorescently labelled oligonucleotide probes (Giraffa and Neviani

2001). The major drawback of the FISH technique is the poor sensitivity in identifying target sequences or organisms in complex matrices (Jacobs et al. 1997), however recently the sensitivity of FISH has increased considerably, making possible the quantitation of both dominant and non-dominant microbial populations (Rudi et al.

2002; Ercolini et al. 2003a). Due to this perceived limitation, alternative techniques such as denaturing gradient gel electrophoresis (DGGE) or temperature gradient gel electrophoresis (TGGE) have been developed.

In the application of DGGE/TGGE, total DNA is extracted from the habitat and bacterial 16S rDNA is specifically amplified using PCR. The resulting amplicons are then separated by DGGE/TGGE on the basis of their sequence. Separation is based on the decreased electrophoretic mobility of a partially melted double-stranded DNA molecule in polyacrylamide gels containing a linear gradient of DNA denaturants (a mixture of urea and formamide) or a linear temperature gradient. The melting of DNA

fragments occurs in melting domains, where the base pairs of identical melting temperature are stretched. Once a domain of the lowest melting temperature or denaturant is reached in the gel, the mobility of the DNA fragment halts. Sequence variation within the DNA fragment causes different melting temperatures and fragments with difference sequences will stop migrating at different positions in the gel (Muyzer and Smalla 1998). On this basis, different bacterial species can be detected and recognized. Usually, the different DNA bands in the gel are excised, and then identified by sequencing (Ercolini 2004).

PCR-DGGE has been used to investigate the ecological profile of microorganisms in natural habitats (Muyzer and Smalla 1998) as well as various foods and beverages, including wine (Lopez et al. 2003), mineral water (Dewettinck et al. 2001), kimchi (Lee et al. 2005), fermented sausages (Cocolin et al. 2001a; Cocolin et al. 2001b; Blaiotta et al. 2003), cheese (Ercolini et al. 2003b; Cocolin et al. 2004; El-Baradei et al. 2007), cocoa (Nielsen et al. 2007) and salads (Handschur et al. 2005). During the study of salads, PCR-DGGE was used to study the changes in the bacterial communities during processing. From the resulting fingerprints, organically farmed field-grown salad was observed to have a more complex DNA band pattern on DGGE than that derived from conventionally farmed field-grown salads (Handschur et al. 2005). PCR-DGGE analysis of cocoa bean fermentations demonstrated that one species of lactic acid bacteria,

Leuconostoc pseudoficulneum, can play a more important role during fermentation than expected from culture-based findings (Nielsen et al. 2007).

2.7.3 Factors affecting the performance of molecular methods for analysis of bacteria

Although molecular methods can provide an overall picture of the microbial community of a sample, this strategy is not without limitations as the method sometimes fails to detect species found by cultural isolation. The efficacy of DNA extraction and performance of the PCR are two key steps in this molecular approach and both can be impacted by the food matrix, such as plant constituents (Lantz et al. 1994; von

Wintzingerode et al. 1997; Wilson 1997; Maukon et al. 2003; Ercolini 2004).

DNA extraction involves the lysing of cells of microorganisms by chemical or mechanical methods. Mechanical methods of DNA extraction consist of breaking open the cells to release the contents into the surrounding solution. Although recovery of the

DNA is good, the DNA is also sheared by the physical process, and may not result in good amplification by PCR (Ogram et al. 1987; Leff et al. 1995). In comparison, chemical lysis of cells uses a more gentle approach with addition of hexadecyltrimethylammonium bromide (CTAB), proteinase K or lysosyme to the cells to induce lysis along with heating of the mixture. This gentler approach allows the recovery of longer strands of DNA; however, variability may be observed between different species of microorganisms (Zhou et al. 1996). Both of these extraction methods may be inhibited by the presence of natural constituents from food matrices such as lipids, proteins, carbohydrates and salts, some of which make DNA extraction very hard (Ercolini 2004).

The PCR itself may be a source of bias in molecular studies of food samples. PCR- inherent bias, such as differential amplification of DNA due to GC content of the DNA, accessibility of templates for primer hybridisation after denaturation, efficiency of primer-template hybrid formation and concentration of templates (Wagner et al. 1994;

Heuer and Smalla 1997; von Wintzingerode et al. 1997; Polz and Cavanaugh 1998); and template reannealing with increasing PCR cycle numbers (Mathieu-Daudé et al. 1996;

Suzuki and Giovannoni 1996; Suzuki et al. 1998), may influence fingerprinting results and have to be considered in interpretation of data. PCR amplification of 16S rDNA can be biased, with the rDNA of one species out of four being preferentially amplified

(Hansen et al. 1998). The PCR bias from the study by Hansen et al. (1998) was observed to be due to the genomic DNA of some species that contain segments outside of the amplified sequence, thus inhibiting the initial PCR steps. Preferential amplification may also be due to the differences in template concentration. Chandler et al. (1997) reported that the PCR template concentration effects the composition and distribution of total community 16S rDNA clone libraries. Consequently, the quantitative and qualitative interpretation of fingerprinting data requires careful and adequate controls. Several studies by terminal restriction fragment length polymorphism

(T-RFLP) methodology analysed bias parameters relating to PCR selection, and resulted in specific recommendations for the individual PCR assays used (Clement et al. 1998;

Osborn et al. 2000). For archaeal templates, the influence of nonspecific priming and the preferential amplification of certain Korarchaeota has been reported (Reysenbach et al. 1992; Brunk and Eis 1998). Consequently, for microbial community analyses, each primer system used should be evaluated carefully for possible PCR bias (Lueders and

Friedrich 2003).

Attempts to overcome these problems have been carried out using solvents or other substances capable of enhancing denaturation of the DNA of the mixture. These additives included betaine and DMSO (dimethyl sulfoxide), both of which have been found to enhance PCR in the past by affecting the specificity of the PCR reaction and enhance PCR amplification of GC-rich DNA sequences (Filichkin and Gelvin 1992;

Rees et al. 1993; Baskaran et al. 1996; Henke et al. 1997; Chakrabarti and Schutt

2001a; 2001b). Hansen et al. (1998) attempted to overcome the bias caused by PCR using a touch-down PCR procedure, performing PCR in the presence of denaturants or cosolvents such as acetamide, DMSO, or glycerol; however, all were unsuccessful.

2.8 Antimicrobials and lettuce

Antimicrobial compounds occur in a range of different food products, including vegetables. These compounds can be released upon harvesting and processing and may affect the microflora of the vegetable. Vegetables known to contain antimicrobial compounds, along with methods for their extraction, will be discussed in the following section.

2.8.1 Natural antimicrobials

Natural antimicrobials can be obtained from many different sources such as microorganisms, vegetables, herbs, spices and fruits (Beuchat & Golden 1989; Beuchat and Brackett 1990; Beuchat et al. 1994; Walker 1994; Cowan 1999; Rauha et al. 2000;

Burnett and Beuchat 2001; Ceylon and Fung 2004; Davidson and Taylor 2007). Some vegetables have been identified to contain antimicrobial compounds which are naturally present in the edible plant tissue or are produced as a result of infection or rupture of tissues. A diversity of these is listed in Table 2.5. The possible presence of antimicrobial substances in lettuce produce has not been studied and warrants investigation.

Antimicrobial activity against L. monocytogenes has been observed for carrots (Beuchat and Brackett 1990; Nguyen-the and Lund 1991; Nguyen-the and Lund 1992; Beuchat et al. 1994), against S. enteritidis, E. coli O157:H7, non-toxigenic E. coli and L.

monocytogenes for cauliflower (Brandi et al. 2006); and against lactic acid bacteria,

Staphylococcus aureus and Candida utilis for cabbage (Pederson & Fisher 1944a,

1944b; Kyung and Fleming 1994; Han & Kyung 1995).

Table 2.5 Antimicrobial compounds that are naturally present in edible plant tissues or produced as a results of infection or rupture of tissues

Common Name Botanical Name Antimicrobial produced Alfalfa Medicago sativa Medicarpin Beet (red) Beta vulgaris Beta vulgarin Broad bean Vica faba Wyerone acid Cabbage Brassica oleracea Rapine, sinigrin Carrot Daucus carota Falcarindiol, 6- methoxymellein Chick pea Cicer aretietum Medicarpin Eggplant Solanum melongena Aubergenone French bean Phaseollus vulgaris Phaseollin Garlic Allium sativum Allyl sulfoxides Onion Allium cepa Protocatechoic acid Parsley Petroselinum spp. Begapten, graveolone, isopimpinellin, psoralen, xanthotoxin Pea (shoot) Pisum sativa Pisatin Pepper (sweet) Capsicum annuum Capsidiol Potato Solanum tuberosum Rishitin, hydroxylubimin, caffeic acid, scopoletin, α- tomatine Radish Raphanus sativus Raphanin Sweet potato Ipomoea batatas Impomeamarone Yam Discorea rotundata Hicicol, isobatasin

Adapted from Beuchat (2006)

2.8.2 Methods for the extraction of antimicrobial compounds from plant materials

As mentioned previously, antimicrobial compounds can be released from the edible tissues by rupture of the tissue during sample preparation. However, other methods have

been specifically developed for the extraction of antimicrobial compounds. These methods can consist of extraction using a chemical solvent, steam distillation or juicing.

Juicing of the plant materials consists of pressing or processing using a commercial juicer to produce a solution from the raw material (Beuchat et al. 1994; Kyung and

Fleming 1994). This method is suitable for releasing antimicrobial compounds that are water soluble and not securely bound to the plant tissue. Unfortunately, a disadvantage of this method is that excess temperature may be generated during processing, thus potentially affecting the activity of the antimicrobial compound (Pederson and Fisher

1944a; Beuchat and Brackett 1990; Nguyen-the and Lund 1991; Beuchat et al. 1994;

Kyung and Fleming 1994; Han and Kyung 1995). This method of extraction has been used to study antimicrobial activity in juices from cabbage (Pederson and Fisher 1944b;

Kyung and Fleming 1994), cauliflower (Brandi et al. 2006), lettuce (Maxcy 1982;

Gómez et al. 2002), and carrot (Beuchat et al. 1994).

Steam distillation is used to extract antimicrobial compounds that may be present in the essential oil fraction of plant materials. It involves boiling the fresh or dried plant material in distilled water, a process that results in the disruption of the gland cells and release the volatile substances contained (Deans 1991). The steam plus volatile oil passes through a cold water condenser allowing the volatile oil fraction to float on top of the water. Typically, steam distillation for three to five hours is required, while the oil yield is around 1-2% (Deans 1991).

Solvent extraction has been frequently used in the past to extract antimicrobial compounds from plant materials (Eloff 1998; Nostro et al. 2000; Rauha et al. 2000;

Palombo and Semple 2001; Alzoreky and Nakahara 2003; Rojas et al. 2003). A range of different solvents can be used for the extraction of these antimicrobial compounds and include acetone, ethanol, methanol, chloroform, methylene dichloride and water (Eloff

1998). For the solvent extraction of antimicrobial compounds, a dried, ground powder of the plant material is required. This facilitates better contact of the material with the solvent and also minimises the presence of water in the extraction (Eloff 1998). Solvent extraction of the plant material usually consists of a batch approach with the sample repetitively extracted using similar amounts of solvent. The solvent from these individual extractions is collected and concentrated using a rotary evaporator (Rauha et al. 2000; Alzoreky and Nakahara 2003). Solvent extracts have been used in the search for antimicrobial activity in medicinal plants in Australia and Peru (Nostro et al. 2000;

Palombo and Semple 2001; Rojas et al. 2003) and in commonly consumed edible plants in Asia (Alzoreky and Nakahara 2003).

Supercritical fluid extraction (SFE) is a novel method by which antimicrobial compounds could be extracted. The properties of supercritical fluids provide a good extraction of the compounds due to their high dissolving power and high mass transfer rates of solutes into the fluid (Raventós et al. 2002). Carbon dioxide (CO2) is the ideal supercritical fluid in the food industry due to its critical temperature (31.06 °C), critical pressure (73.84 bar) and critical density is (0.460 g/cm3) being relatively easy to achieve. The advantages of SFE are a product of high quality and purity, quick extraction and separation phases, an extract free of residues, a selective extraction by a specific compound and reduction in separation cost (Raventós et al. 2002). Supercritical extraction is also able to extract a wide range of compounds from non-polar to polar

(Modey et al. 1996). Although used widely for other applications, it has been used to extract antimicrobial compounds from capsicum (Cichewicz and Thorpe 1996) and essential oils from plant materials (Reverchon 1997; Marongiu et al. 2003). It has not been applied to the extraction of potential antimicrobials from lettuce produce.

CHAPTER 3

PESTICIDES AS A SOURCE OF MICROBIAL

CONTAMINATION OF VEGETABLE CROPS

3.1 Introduction

Pesticides are routinely used in the cultivation of fruit and vegetables to control insects, weeds, spoilage bacteria and fungi, and other pests. Pesticides are commonly grouped into three categories: insecticides, herbicides, and fungicides (Hislop 1976). These categories have different target populations and active ingredients, and usually come in a form of concentrated powder or liquid. Reconstitution and dilution of the pesticide with water is required before application to vegetables, and is usually done on the farm by the farmer.

As mentioned earlier (Chapter 2), there has been increased interest in the microbiological quality and safety of fresh produce such as salad vegetables, as they have been linked to outbreaks of foodborne microbial disease (Nguyen-the and Carlin

2000; Beuchat 2002; Heard 2002; Sivapalasingam et al. 2004). While postharvest contamination is often the source of the implicated microorganisms, there is increasing concern that preharvest contamination presents significant public health risks. The main sources of preharvest contamination have been identified by many researchers as fertilisers, irrigation water and soil (Lund 1992; Beuchat 1996, 2002; Brackett 1999;

Duffy et al. 2005). As mentioned already, a diversity of pesticides is regularly applied to vegetable produce, but they are rarely considered to be a source of microbial contamination. This oversight was realised when it was found that contamination of

raspberries from pesticides lead to an outbreak of cyclosporiasis food poisoning

(Herwaldt and Ackers 1997).

Some recent studies have shown that reconstituted pesticides may present a suitable environment for the survival and growth of pathogenic organisms such as Salmonella,

Shigella, Escherichia coli O157:H7, and Listeria monocytogenes (Coghlan 2000; Guan et al. 2001; Guan et al. 2005). Therefore, preharvest application of pesticide solutions onto vegetable produce could be an important additional source of microbial contamination.

There are several microbiological issues to consider in relation to the application of pesticides to vegetable produce. First, the pesticide solution itself could be a source of microbial contaminants. Second, their chemical composition might either stimulate or inhibit microbial growth, and this could be a significant property either before or after their application to the produce. Third, the microbiological quality of the water used for reconstitution and dilution of the pesticide concentrate could be an important factor, especially if the final pesticide solution can support the growth of microbial contaminants originating from the water. Finally, the time between reconstitution of the pesticide and spraying onto produce is another important variable and could range between immediate use, and use after several hours of storage. In practical situations, reconstituted pesticides might be stored for several hours at warm temperatures on the farm before application to produce. During this time, growth of microorganisms present in the water used for reconstitution could occur. In Australia, farmers use a diversity of water sources for reconstitution of pesticides. These sources include dam water, river water or underground bore water.

The aims of this chapter are to investigate if pesticides: (i) are a source of microbial contaminants; (ii) affect the survival and growth of bacteria that have spoilage and

public health significance on fresh produce; and (iii) could support microbial growth after reconstitution in several sources of agricultural water.

3.2 Materials and Methods

3.2.1 Pesticides

A total of 10 commercial pesticide concentrates were examined (Table 3.1). These included a variety of insecticides, fungicides and herbicides commonly used in the cultivation of lettuce in Australia and elsewhere (McDougall et al. 2002). These concentrates were obtained directly from the manufacturer and reconstituted in water according to recommendations of the manufacturer. When not in use, all pesticide products were stored in the dark, and fresh solutions were prepared for every experiment. Measurements of pH were performed on pesticides reconstituted with distilled water, using a pH meter.

3.2.2 Microorganisms and their cultivation

The reference microorganisms used throughout this research are listed in Table 3.2. All microbiological media were obtained from Oxoid (Melbourne, Australia), except for

Chromocult agar which was obtained from Merck (Melbourne, Australia).

Microorganisms were maintained by subculturing on Tryptone Soya Agar (TSA) at 30 or 37°C for 1-2 days, and stored at 4°C. For use in experiments, cultures were activated by transferring loop inocula into 10 ml of Tryptone Soya Broth (TSB) and incubation at

30 or 37°C for 18-24 h.

Table 3.1 Pesticide products tested, concentrations of active ingredient after reconstitution, and pH values

Conc. active ingredient Pesticide product Company Active ingredient after reconstitution pH (trade name) (g/L) Herbicides Stomp 330E BASF Pendimethalin 3.30 7.0 Fusilade Syngenta Fluazifop-p 1.10 7.2

Fungicides Kumulus DF BASF Sulfur 1.60 8.9 Champ Dry Prill WG Nufarm Copper hydroxide 0.53 10.9 Tri-Base Blue Nufarm Tri-basic Copper sulphate 0.80 7.2 Penncozeb 750 DF Nufarm Mancozeb 1.50 7.6

Insecticides Fastac Duo BASF Alpha-cypermethrin 0.05 5.4 Nudrin 225 BASF Methomyl 0.45 5.4 Ambush EC Syngenta Permethrin 0.10 6.5 Pirimor WG Syngenta Pirimicarb 1.00 8.3

BASF Australia Ltd., Noble Park, Vic, Australia Nufarm Australia Ltd., Laverton, Vic, Australia Syngenta Crop Protection Pty. Ltd., Pendle Hill, NSW, Australia

Table 3.2 Bacterial cultures used in the analysis of pesticide solutions

Culture Source Escherichia coli School of Biotechnology and Biomolecular Sciences, UNSWa (ATCCb 11775, UNSW 048200) Escherichia coli O111:NM School of Biotechnology and Biomolecular Sciences, UNSW Escherichia coli O157:H7 School of Biotechnology and Biomolecular Sciences, UNSW (ATCC 35150) Listeria monocytogenes (L2) 1768 serotype 1/2a, TECRAc, Sydney Listeria monocytogenes (L4) 1771 serotype 4c, TECRA, Sydney (ATCC 19116) Listeria monocytogenes (L5) 1773 serotype 4e, TECRA, Sydney (ATCC 19118) Pseudomonas fluorescens School of Biotechnology and Biomolecular Sciences, UNSW (UNSW 022600) Pseudomonas fluorescens School of Biotechnology and Biomolecular Sciences, UNSW (ATCC 13525, UNSW 036800) Pseudomonas fragi School of Chemical Sciences and Engineering, UNSW Salmonella enteritidis School of Biotechnology and Biomolecular Sciences, UNSW (UNSW 031901) Salmonella typhimurium School of Chemical Sciences and Engineering, UNSW a UNSW; University of New South Wales, Sydney, Australia b ATCC; American Type Culture Collection c TECRA; Tecra International, Frenchs Forest, Sydney, Australia

3.2.3 Analysis of pesticide concentrates for microorganisms

Pesticides were reconstituted in sterile distilled water and statically incubated at 30°C for 48 h. Samples were taken at 0 and 24 h intervals, serially diluted in 0.1%

Bacteriological Peptone water, and 0.1 ml spread inoculated, in duplicate, over the surface of petrie dishes of Plate Count Agar (PCA). Plates were incubated at 30°C for

48 h and examined for the presence and populations (cfu/g) of colonies.

3.2.4 Antimicrobial properties of pesticide solutions

Pesticides were reconstituted in sterile distilled water to obtain three different concentrations of the active ingredient; the concentration normally applied to vegetable

(100%), and then half (50%) and double (200%) this concentration. Paper discs (8 mm,

Whatman No. 3 filter paper) were soaked in these pesticide preparations or sterile water

(control- no pesticide) for 30 mins before use. Overnight cultures of species listed in

Table 3.2 were spread onto plates of TSA and allowed to dry. The paper discs impregnated with pesticide preparations were aseptically transferred to the surface of the inoculated agar plate. Each plate contained one control disc and three pesticide discs at the different concentrations, noted previously. Plates were incubated according to the microorganisms being analysed at either, 30°C or 37°C for 24-48 h. Plates were observed for zones of inhibited bacterial growth around the edge of the paper disc, and this was taken as an indication of antimicrobial activity of the pesticide impregnated into the disc.

3.2.5 Growth and survival of pathogenic and spoilage organisms in pesticide solutions

Pesticides were reconstituted in sterile water to the concentration recommended by the manufacturer (Table 3.1) and inoculated with cells (103-104 cfu/ml) from an overnight culture of the microbial species (Table 3.2) under test. The inoculated pesticide preparations were statically incubated for 48 h at 30°C and sampled at 12h intervals for the measurement of viable populations. Samples were serially diluted in 0.1%

Bacteriological Peptone and spread inoculated, in duplicate, onto plates of PCA. Plates were incubated for 24-48 h at 30 or 37°C, depending upon the organism being analysed, and colonies counted.

3.2.6 Growth and survival of natural populations from agricultural water in pesticide solutions

Samples of agricultural waters, used to reconstitute the pesticide concentrates on farms, were obtained from various sources and growers in the states of Queensland (QLD) and

New South Wales (NSW), Australia. These samples consisted of bore and dam water obtained from different locations near Brisbane, Queensland, and river and dam water obtained from locations near Sydney, New South Wales. Samples were stored at 4-5°C on collection, and used within 24-48 hr. Pesticides were reconstituted to the recommended concentration using agricultural water (final volume 100 ml) and statically incubated for 24-48 h at 30°C. Samples were taken at 6-12 h intervals, serially diluted in 0.1% Bacteriological Peptone water, and 0.1 ml spread inoculated, in duplicate, onto plates of PCA and Chromocult agar. Plates were incubated at 30°C for

48 h (PCA) and 37°C for 18-24 h (Chromocult), after which time colonies were

counted. Predominant colony morphologies were noted and representative isolates were purified by streaking onto plates of TSA.

3.2.7 Identification of bacterial isolates

Bacterial isolates were examined for their Gram reaction and cellular morphology, oxidase and catalase reactions and identified to species by sequencing sections of their

16S ribosomal DNA gene. DNA was extracted from disrupted cells by a modification of the method described by Kowalchuk et al. (1997). Bacterial isolates were grown overnight in Nutrient Broth. Samples (1-2 ml) of this culture were centrifuged (10-15 min at 15,000g) in a Beckman Microfuge 18 Centrifuge (Beckman Coulter Inc.,

Fullerton, CA, USA) to obtain a cell pellet. To lyse the cell pellet, 0.3 g zirconia/silica beads (diameter 0.1 mm; Daintree Scientific, Tasmania, Australia), 0.5 ml of extraction buffer (100 mM Tris-HCl [pH 8], 50 mM EDTA [pH 8], 100 mM NaCl, 1% sodium dodecyl sulphate [SDS]) and 0.5 ml phenol:chloroform:isoamyl alcohol (24:24:1;

Sigma-Aldrich, Australia) were mixed in a 2 ml screw capped microtube. The samples were shaken at 5000 rpm for 30 s in a mini-beadbeater (Biospec Products, OK, USA).

After centrifugation for 10 mins (15,000g at 4°C), 0.4 ml of the upper layer was removed and re-extracted with another 0.5 ml of phenol:chloroform:isoamyl alcohol.

DNA was precipitated by addition of 0.6 ml of isopropanol and the mix allowed to stand for 24h at -20°C. After centrifugation for 10 mins (15,000g), the pellet was washed once with 70% ethanol and allowed to air dry. The dried pellet was dissolved in 50 µl of TE buffer (10 mM Tris-HCl, 1 mM EDTA [pH 8]) and stored at -20°C.

The extracted DNA was used as template DNA in the polymerase chain reaction (PCR) to amplify sections of the 16S ribosomal RNA region (Medlin et al. 1988; Saiki et al.

1988). Two universal eubacterial primers 341F (5’- CCTACGGGAGGCAGCAG-3’) and 907R (5’-CCGTCAATTCCTTRAGTTT-3’) were use for PCR amplification, and were obtained from SigmaGenosys Australia Pty. Ltd. The PCR mixture contain 1X

PCR buffer (10 mM Tris-HCl [pH 8.3], 50 mM KCl), 200 µM of each dNTP (Roche

Diagnostics Corp., IN, USA), 0.2 µM of each primer, 1.5 mM MgCl2, 1.25 U of Taq

DNA polymerase (AmpliTaq™, Roche Molecular Systems Inc., New Jersey, USA) and

~100 ng of template DNA in 50 µl final volume. PCR amplifications were performed with a GeneAmp® PCR system 9700 (Applied Biosystems Inc., CA, USA) with the following cycling program: initial denaturation at 94°C for 5 min, 20 touchdown cycles at 94°C for 0.5 min, 60°C for 1 min and 72°C for 1.5 min, with a decrease in annealing temperature of 1°C every two cycles, followed by 10 cycles at 94°C for 0.5 min, 50°C for 1 min and 72°C for 1.5 min, and a final extension at 72°C for 10 min. The production of PCR amplicons was confirmed by 1.5% agarose gel electrophoresis, and then the amplicons were used for sequence analysis. Sequencing was carried out with the ABIPRISM® BigDye™ Terminators V3.1 Cycle Sequencing Kit (Applied

Biosystems) and assayed at the Automated DNA Analysis Facility, School of

Biotechnology and Biomolecular Science, University of New South Wales. The sequences were compared with the non-redundant nucleotide database at GenBank, using BLAST to determine bacterial identities (Altschul et al. 1990; Burks et al. 1992;

Maidak et al. 1999).

3.3 Results

3.3.1 Microbial content and antimicrobial activity of pesticides

No viable microorganisms were detected in any of the pesticide products immediately after reconstitution in sterile water or after incubation of the reconstituted product at

30°C for 48 h. As determined by the paper disc assay, none of the pesticides inhibited the growth of any of the bacterial species listed in Table 3.2.

3.3.2 Survival and growth of microorganisms inoculated into pesticide solutions

The bacterial species (Table 3.2) were examined for their ability to survive or grow in pesticides reconstituted to their recommended concentration.

Depending on the pesticide and microorganism combination, three responses were observed; (i) the bacteria died off; (ii) they remained viable without growth; or (iii) they grew (Table 3.3). In many cases, the bacteria died off very quickly after contact with the pesticide solution (D0), and no viable colonies (< 5 cfu/ml) were detected in the samples tested immediately after addition of bacterial cells to the pesticide solution. In several cases, such death was observed after 12, 24 and 36 h of microbial contact with the pesticides (D12, D24, D36). In a few examples (E. coli O157:H7 and S. enteritidis in

Kumulus DF), the inoculated cells did not grow, but remained viable at their initial population until the end of storage (48 h) (S). For the Kumulus DF and Pirimor WG preparations, there were several examples (E. coli O157:H7, E. coli O111:NM and Ps. fragi) where the inoculated species exhibited growth (G), giving final populations as high as 106-107 cfu/ml. Figures 3.1 and 3.2 show the growth and survival responses of species of Pseudomonas, Listeria, E. coli and Salmonella in Kumulus DF and Pirimor

WG preparations.

Table 3.3 Growth and survival responses of bacteria in pesticide solutions after storage at 30°C for 48 h

Pesticide solution Bacterial species Stomp Fusilade Kumulus Champ Tri-Base Penncozeb Fastac Nudrin Ambush Pirimor 330E DF Dry Prill Blue 750 DF Duo 225 EC WG Escherichia coli D0 D0 G D12 D0 D12 D0 D0 D0 G Escherichia coli D0 D0 G D12 D0 D12 D12 D0 D0 D12 O111:NM Escherichia coli D0 D0 S D12 D0 D12 D0 D0 D12 G O157:H7 Listeria monocytogenes D0 D0 D36 D12 D0 D12 D0 D0 D0 D36 (L2) Listeria monocytogenes D0 D0 D12 DD12 12 D24 D0 D12 D12 D24 (L4) Listeria monocytogenes D0 D0 D12 D12 D0 D12 D0 D12 D0 D12 (L5) Pseudomonas D12 D12 G D0 D0 D12 D0 D12 D12 G fluorescens (UNSW 022600) Pseudomonas D0 D0 G D0 D0 D0 D0 D0 D0 G fluorescens (UNSW 036800) Pseudomonas fragi D0 D0 G D0 D0 D0 D0 D12 D0 G Salmonella enteritidis D0 D0 S D12 D0 D D0 D0 D0 S Salmonella typhimurium D0 D0 G D12 D0 D12 D0 D0 D0 G

Control- inoculated sterile water with no pesticide; G- growth of bacterial population (final population > initial population); S- survival of bacterial population, but no growth; D0- bacterial population died off on contact with pesticide solution (< 5 cfu/ml); D12- bacterial population died off within 12 h; not detected after incubation of medium for 12 h; D24- bacterial population died off within 24 h not detected after incubation of medium for 24 h; D36- bacterial population died off within 36 h; not detected after incubation of medium for 36 h,

a b

c d

e f

Figure 3.1 Growth and survival of bacteria in the pesticide solutions, Kumulus DF and Pirimor WG, during incubation at 30°C for 48 h. (a) Pseudomonas fluorescens (UNSW 022600), (b) Listeria monocytogenes (L2), (c) Pseudomonas fluorescens (UNSW 036800), (d) Listeria monocytogenes (L4), (e) Pseudomonas fragi, (f) Listeria monocytogenes (L5); Control ♦ (inoculated sterile water with no pesticide); Kumulus DF □; Pirimor WG ○

a b

c d

Figure 3.2 Growth and survival of bacteria in the pesticide solutions, Kumulus DF and Pirimor WG, during incubation at 30°C for 48 h. (a) Escherichia coli, (b) Salmonella enteritidis, (c) Escherichia coli O111:NM, (d) Salmonella typhimurium, (e) Escherichia coli O157:H7; Control ♦ (inoculated sterile water with no pesticide); Kumulus DF □; Pirimor WG ○

3.3.3 Growth and survival of microorganisms in pesticides reconstituted in different sources of agricultural water

Pesticides were reconstituted in four different sources of agricultural water and monitored for the survival and growth of microorganisms naturally present in the water

(Figures 3.3-3.6). The microbiological quality of the water samples varied with the source, where initial populations ranged from 102-106 cfu/ml. Some of the water samples contained sufficient nutrients to allow growth of their indigenous microflora during incubation at 30°C (see control data Figures 3.3-3.6). The two herbicides, Stomp

330E and Fusilade supported the growth of water microflora. This was particularly evident for the dam water samples where initial populations of 103-105 cfu/ml increased to 105-107 cfu/ml by 48 h (Figures 3.3a, 3.3d). The fungicide, Penncozeb 750 DF, did not support the growth of water microflora which decreased in population (Figures 3.4c,

3.4d) or completely died off (Figures 3.4a, 3.4b) depending upon the sample. The other fungicides gave varying levels of growth. The microflora of all samples grew well in the presence of Kumulus DF, and final populations of 106-107 cfu/ml were observed (Figure

3.4). The microflora of the dam water (QLD) grew particularly well in Tri-Base Blue and Champ Dry Prill WG (Figure 3.4d), but this was not the case for the microflora of the dam water from NSW (Figure 3.4a), illustrating the influence of the species that comprise the initial microflora. Of the insecticides, best growth was observed in Pirimor

WG and Ambush EC preparations, with Fastac Duo also supporting good growth for the bore water and dam water (QLD) samples (Figure 3.5).

7.0 a 7.0 b 6.0 6.0 5.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0 1.0

Population (log cfu/ml) 1.0 Population (log cfu/ml) 0.0 0.0 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h)

8.0 8.0 d 7.0 c 7.0 6.0 6.0 5.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0

Population (logcfu/ml) 1.0

Population (log cfu/ml) (log Population 1.0 0.0 0.0 0 12243648 0 12243648 Time (h) Time (h) Figure 3.3 Survival and growth of microorganisms in herbicides reconstituted in (a) dam water NSW, (b) river water, (c) bore water, (d) dam water QLD, and incubated at 30°C for 24-48 h; Control ♦ (water without pesticide addition); Stomp 330E ○; Fusilade □.

7.0 a 8.0 b 6.0 7.0 5.0 6.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0 1.0 Population (log cfu/ml) (log Population

Population (log cfu/ml) (log Population 1.0 0.0 0.0 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h)

8.0 c 8.0 d 7.0 7.0 6.0 6.0 5.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0 Population (log cfu/ml) (log Population 1.0 Population (log cfu/ml) 1.0 0.0 0.0 0 12243648 0 12243648 Time (h) Time (h)

Figure 3.4 Survival and growth of microorganisms in fungicides reconstituted in (a) dam water NSW, (b) river water, (c) bore water, (d) dam water QLD, and incubated at 30°C for 24-48 h; Control ♦ (water without pesticide addition); Kumulus DF ○; Champ Dry Prill WG □; Penncozeb 750 DF ●; Tri-Base Blue ■.

7.0 a 8.0 b 6.0 7.0 5.0 6.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0 1.0 Population (log cfu/ml)

Population (log cfu/ml) 1.0 0.0 0.0 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h)

8.0 c 8.0 d 7.0 7.0 6.0 6.0 5.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0

Population (logcfu/ml) 1.0 Population (log cfu/ml) (log Population 1.0 0.0 0.0 0 12243648 0 12243648 Time (h) Time (h)

Figure 3.5 Survival and growth of microorganisms in insecticides reconstituted in (a) dam water NSW, (b) river water, (c) bore water, (d) dam water QLD, and incubated at 30°C for 24-48 h; Control ♦ (water without pesticide addition); Fastac Duo ○; Nudrin 225□; Ambush EC ●; Pirimor WG ■.

Chromocult agar was used to monitor the survival and growth of coliforms in pesticides reconstituted in the four samples of agricultural water. Although control samples

(Figures 3.6-3.8) showed that coliforms could survive and even grow in some water samples in the absence of pesticides, there were clear cases where pesticide addition encouraged and supported their growth. Pesticides giving such reactions were Kumulus

DF, Ambush EC, Pirimor WG and Stomp 330E (Figures 3.6-3.8). Maximum population of presumptive coliforms were observed to be as high as 103-106 cfu/ml.

Some of the organisms which grew on Chromocult agar were subsequently identified as non-coliforms. This was the case for the dam water (NSW) where the predominant organisms isolated off Chromocult agar were Aeromonas spp. (Figures 3.6a-3.8a).

Consequently, experiments with dam (NSW) and river water were repeated but samples were plated onto Chromocult agar supplemented with 5 mg/L of cefsulodin as an additional agent to inhibit non-coliform species (Alonso et al. 1996). In these cases, no coliforms were detected in the river water sample at any sample time. However, coliforms were present in the dam water and exhibited significant growth in the presence of Kumulus DF and lesser growth in the presence of Stomp 330E and Pirimor

WG (Figures 3.6e-3.8e). Significant growth was also observed in Nudrin 225 (Figure

3.8e)

5.0 a 5.0 b

4.0 4.0

3.0 3.0

2.0 2.0

1.0 1.0 Population (log cfu/ml) (log Population Population (log cfu/ml) (log Population 0.0 0.0 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h) 6.0 c 5.0 d 5.0 4.0 4.0 3.0 3.0 2.0 2.0

1.0 1.0 Population (log cfu/ml) (log Population Population (log cfu/ml) (log Population 0.0 0.0 0 122436480 12243648 Time (h) Time (h)

4.0 e

3.0

2.0

1.0 Population (log cfu/ml) (log Population 0.0 0 12243648 Time (h)

Figure 3.6 Survival and growth of coliforms in fungicides reconstituted in (a) dam water NSW, (b) river water, (c) bore water, (d) dam water QLD, (e) dam water NSW (repeat) and incubated at 30°C for 24-48 h; Control ♦ (water without pesticide addition); Kumulus DF ○; Champ Dry Prill WG □; Penncozeb 750 DF ●; Tri-Base Blue ■.

6.0 a 4.0 b 5.0 3.0 4.0

3.0 2.0

2.0 1.0 1.0 Population (log cfu/ml) (log Population Population (log cfu/ml) (log Population 0.0 0.0 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h)

4.0 c 6.0 d 5.0 3.0 4.0

2.0 3.0

2.0 1.0 1.0 Population (log cfu/ml) (log Population Population (log cfu/ml) (log Population 0.0 0.0 0 122436480 12243648 Time (h) Time (h)

3.0 e

2.0

1.0 Population (log cfu/ml) (log Population 0.0 0 12243648 Time (h)

Figure 3.7 Survival and growth of coliforms in herbicides reconstituted in (a) dam water NSW, (b) river water, (c) bore water, (d) dam water QLD, (e) dam water NSW (repeat) and incubated at 30°C for 24-48 h; Control ♦ (water without pesticide addition); Stomp 330E ○; Fusilade □.

4.0 a 4.0 b

3.0 3.0

2.0 2.0

1.0 1.0 Population (log cfu/ml) (log Population Population (log cfu/ml) (log Population 0.0 0.0 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h)

5.0 c 6.0 d 5.0 4.0 4.0 3.0 3.0 2.0 2.0

1.0 1.0 Population (log cfu/ml) Population (log cfu/ml) 0.0 0.0 0 122436480 12243648 Time (h) Time (h)

5.0 e

4.0

3.0

2.0

1.0 Population (log cfu/ml) (log Population 0.0 0 12243648 Time (h)

Figure 3.8 Survival and growth of coliforms in insecticides reconstituted in (a) dam water NSW, (b) river water, (c) bore water, (d) dam water QLD, (e) dam water NSW (repeat) and incubated at 30°C for 24-48 h; Control ♦ (water without pesticide addition); Fastac Duo ○; Nudrin 225□; Ambush EC ●; Pirimor WG ■.

3.3.4 Identification of bacterial species that grow in pesticides

The most prevalent microorganisms that grew in the reconstituted pesticide preparations

(Figures 3.3-3.8) were isolated from the plates of PCA and Chromocult agar, and identified (Tables 3.4 & 3.5). The bacterial ecology of the water samples before addition of the pesticides (control, 0 h, Table 3.4) varied with the source, with the majority of isolates being Gram negative rods. Pseudomonas species were predominant in the bore and river water samples, Acinetobacter species were prevalent in one of the dam water

(QLD) samples, and Bacillus species were prevalent in the other sample of dam water

(NSW).

Table 3.4 Predominant bacterial species isolated from pesticide solutions after reconstitution in some agricultural waters, storage for 24-48 h at 30°C and plating onto PCA

Water source & Species and population (cfu/ml) pesticide solution 0 h 12-18 h Bore Control Pseudomonas fluorescens (96%)a, Pseudomonas fluorescens (95%), 1.4 x 106 3.5 x 105 Kumulus DF Pseudomonas fluorescens or extremorientalis (88%), 2.9 x 106 Stomp 330E Pseudomonas fluorescens (93%), 8.7 x 106 Pirimor WG Pseudomonas fluorescens (93%), 9.4 x 106 Dam (QLD) Control Acinetobacter junii (99%), Acinetobacter junii (99%), 8.7 x 103; Acinetobacter johnsonii 6.7 x 103 or haemolyticus (83%), 6.2 x 103 Kumulus DF Acinetobacter junii (99%), 1.7 x 105 Stomp 330E Acinetobacter johnsonii (92%) or junii (91%), 3.8 x 106 Pirimor WG Acinetobacter junii (98%), 1.3 x 106 Dam (NSW) Control Bacillus spp. (93%), 3.5 x 102 Pseudomonas putida (85%), 1.4 x 104 Kumulus DF Aeromonas eucrenophila or encheleia (96%), 1.4 x 105 Stomp 330E Pseudomonas putida (93%), 1.3 x 104 Pirimor WG Pseudomonas putida (93%), 1.6 x 105 River Control Pseudomonas putida or migulae Afipia (94%) or Bosea spp. (93%), (90%), 7.0 x 101 2.8 x 105; Pseudomonas alcaligenes (92%) or aeruginosa (91%), 1.2 x 105 Kumulus DF Afipia or Bosea spp. (93%), 4.2 x 105 Stomp 330E Pseudomonas putida or migulae (95%), 2.9 x 105 Pirimor WG Pseudomonas putida (93%), 3.5 x 105 a Species (% match to BLAST sequence database)

Table 3.5 Predominant bacterial species isolated from pesticide solutions after reconstitution in some agricultural waters, storage for 48 h at 30°C and plating onto Chromocult agar

Water source & pesticide Species and population (cfu/ml) solution 0 h 36 h Dam (NSW) Control Aeromonas eucrenophila or Aeromonas veronii (97%), encheleia (97%)a, 5.0 x 101 3.0 x 101 Kumulus DF Aeromonas eucrenophila or encheleia (97%), 6.6 x 103 Stomp 330E Aeromonas hydrophila (96%), 5.1 x 103 Pirimor WG Pseudomonas putida (94%), 6.8 x 102 River Control Aeromonas spp. (97%), 2.1 Rahnella spp. (89%), 4.8 x 102 x 101; Klebsiella planticola (89%), 1.6 x 101 Kumulus DF Pseudomonas jessenii (94%), 4.5 x 103 Stomp 330E Serratia marcescens (87%), 3.7 x 102 Pirimor WG Aeromonas hydrophila (92%), 1.3 x 103 Dam (NSW)- repeat Control Kluyvera spp. (87%), Citrobacter freundii (89%), 2.2 x 5.6 x 101; Enterobacter 101; Enterobacter aerogenes cloacae or Pantoea (89%) or Kluyvera cryocrescens agglomerans (90%), (88%), 4.3 x 101 7.4 x 101 Kumulus DF Kluyvera georgiana (92%), 1.5 x 102; Pantoea agglomerans (87%), 1.5 x 102 Stomp 330E Citrobacter amalonaticus or Kluyvera georgiana (88%), 1.1 x 102; Enterobacter cloacae or Escherichia adecarboylata (91%), 5.6 x 101 Pirimor WG Enterobacter aerogenes or Klebsiella terrigena (89%), 1.5 x 102; Kluyvera ascorbata (92%), 1.5 x 102 a Species (% match to BLAST sequence database)

For two of the water samples (bore and dam (QLD)), the predominant species after incubation, with or without pesticide addition, were similar to those before incubation.

For the bore water, they were Ps. fluorescens, and for the dam water (QLD) they were

Ac. junii and Ac. johnsonii. For the dam water (NSW) sample, the initial population of

Bacillus spp. gave way to predominance of Ps. putida after incubation without pesticide addition, and with the addition of Stomp 330E and Pirimor WG pesticides (Table 3.4).

However, when Kumulus DF was the added pesticide, Aeromonas spp. predominated after incubation, suggesting some selective influence of this agent. For the sample of river water, Ps. putida/migulae was predominant in the initial microflora and also predominated in the Stomp 330E and Pirimor WG pesticide solutions after incubation.

However, for the control (in particular) and the Kumulus DF solution, Afipia/Bosea species predominated.

Chromocult agar was not entirely effective in selecting for the growth of coliforms, and non-coliforms sometimes even grew and predominated after plating of water samples on this medium. As explained already, the dam water (NSW) gave a predominance of

Aeromonas spp. on this agar before and after incubation with pesticides, and also Ps. putida, when Pirimor WG was the pesticide. The river water samples gave a mixture of

Klebsiella and Aeromonas spp. which gave way to either Rahnella, Pseudomonas or

Serratia spp. after incubation with pesticides (Table 3.5).

Addition of cefsulodin to Chromocult agar improved its selectivity for coliforms. Thus, only coliforms (Kluyvera, Pantoea, and Enterobacter spp.) were detected in a sample of dam water (NSW), and these species were also found after incubation of the water with various pesticides (Table 3.5). Growth of coliforms occurred in the presence of

Kumulus DF (Figure 3.6e) and the predominant species found were those of Pantoea and Kluyvera.

3.4 Discussion

Pesticides are usually chemical preparations that are widely used in the cultivation of fruits and vegetables (Andrews and Kenerley 1978; Connor 1983; Beckmann et al.

1984; Mercier and Reeleder 1986). Generally, they consist of a specific ingredient which is active against a particular group of pests, and other additives, or adjuvants often referred to as inerts, that promote efficacy and stability (Hochberg 1996). Many of the inerts function as surfactants, emulsifiers, dispersants, solubilisers, solvents, carriers/diluents, preservatives and thickeners or suspending agents (Crowdy 1971). In some circumstances, these may be added to the mixture by the farmer on reconstitution of the pesticide. Usually, the precise chemical composition of pesticide preparations is confidential or proprietary information and not publicly available. Their chemical composition can pose a risk to consumer health and, for this reason, most countries have developed legislation that governs the permissible concentration of pesticide residues in foods (EU 2007; FSANZ 2007; US EPA 2007). In contrast, however, their use in food production is not normally considered to present a microbiological risk to public health, but in light of the data reported in this chapter, this view needs to be re-considered.

Guan et al. (2001) were probably the first researchers to significantly question the microbiological risks associated with pesticide use in food production. They realised that pesticide preparations are normally reconstituted and diluted in non-potable farm or agricultural waters, and that bacterial contaminants, including pathogens, from this 73

source might then grow in the preparations prior to application to the produce. They tested this hypothesis by inoculating food-borne pathogens such as Salmonella, Listeria,

Shigella and E. coli into a diverse array of commercial pesticide preparations and found that in many, but not all preparations, species of these pathogens survived and grew.

This chapter has provided a more detailed examination of this concept and convincingly demonstrates that bacterial contaminants in agricultural waters can exhibit significant growth in some pesticide preparations.

As obtained from the manufacturer, all ten pesticide preparations examined in this study were effectively sterile. However, they were not antagonistic or inhibitory to a range of food-associated pathogenic bacteria and spoilage Pseudomonas species as determined by the paper-disc diffusion assay.

After inoculation of species of E. coli, Salmonella, Pseudomonas and Listeria into pesticide preparations, the microorganisms either died off, survived, or grew, depending on the species and pesticide preparation. Only two of the ten pesticides examined allowed the survival and growth of inoculated species of E. coli, Salmonella and

Pseudomonas, although there was some variation in response depending on strain and species. These pesticides were the fungicide, Kumulus DF, and the insecticide, Pirimor

WG. For the other pesticides, the inoculated cells died off within 12 h. Guan et al.

(2001) noted death of similar cultures in the fungicide, Dithane M-45, which has the same active ingredient, mancozeb, as Penncozeb 750 DF (Table 3.1). However, they did report growth of E. coli, S. typhimurium and L. monocytogenes in Ambush 500EC which is inconsistent with data of this chapter. Their growth studies were conducted at

22°C, whereas the present experiments were done at 30°C. Moreover, there is a lower concentration of the active ingredient (permethrin) in Ambush 500 EC than Ambush

EC, which was used in this study. Most of the data in Table 3.3 showing non-survival and death of inoculated species in liquid preparations of the pesticide are inconsistent with the observations obtained using the paper-disc diffusion assay, where no inhibition of growth of the organisms was observed. Possibly, nutrients in the agar base (TSA) may have helped to protect the organisms from the effects of the pesticide. Similar discrepancies in the behaviour of solid and liquid media for measuring antimicrobial activity were reported by Coventry et al. (1997) during studies of the sensitivity of

Listeria species to various bacteriocins. The reason for the difference was not given

(Coventry et al. 1997).

The pure culture studies reported in this chapter, in conjunction with those of Guan et al. (2001), clearly establish that some pesticide preparations support the growth of bacteria. These observations have been confirmed by additional experiments where the pesticides were reconstituted and diluted in various sources of non-potable agricultural water. Farmers access these sources of water for irrigating their crops and to reconstitute pesticides. Such waters are obtained from dams, rivers, bores, creeks and lakes. There is substantial literature showing that these waters harbour viable microbial populations ranging from 102-106 cfu/ml, depending upon the source (Robinson and Adams 1978;

Falcão et al. 1993; Wang and Doyle 1998; Riordan et al. 2001). Moreover, this population can include coliforms, faecal coliforms, E. coli and various food-associated pathogens such as Salmonella, Campylobacter jejuni, L. monocytogenes and

Clostridium perfringens (Watkins and Sleath 1981; Rosas et al. 1984; Bolton et al.

1987; Vaz da Costa-Vargas et al. 1991; Falcão et al. 1993; Arvanitidou et al. 1995;

Thurman et al. 1998; Szewzyk et al. 2000; Pianetti et al. 2004; Steele et al. 2005).

In this study, the agricultural waters used for the reconstitution of pesticides had initial bacterial populations from 103-106 cfu/ml, with the predominant species being those of

Pseudomonas, Acinetobacter, and Bacillus depending upon the source (Table 3.4). Such species are frequently isolated from potable and raw water supplies (LeChevallier et al.

1980; Olsen and Nagy 1984; Priest 1989; Guardabassi et al. 1999; Szewzyk et al. 2000;

Park et al. 2003; Steele and Odumeru 2004; Ultee et al. 2004). Apart from Bacillus spp., these organisms were also predominant in the water after storage, even with the addition of pesticides (Table 3.4). The growth of Pseudomonas and Acinetobacter species in pesticide preparations is particularly significant because they are well known as vegetable spoilage organisms (Liao and Wells 1987; Nguyen-the and Prunier 1989;

Khan et al. 1992; Nguyen-the and Carlin 2000). As will be discussed in Chapter 6,

Acinetobacter species are significant nosocomial pathogens that are attracting increasing public health interest due to their resistance to commonly used antibiotics.

Consistent with other literature (Watkins and Sleath 1981; Rosas et al. 1984; Bolton et al. 1987; Vaz da Costa-Vargas et al. 1991; Falcão et al. 1993; Thurman et al. 1998), some of the water samples contained species of coliforms and these also grew in some pesticides, after reconstitution and storage. Most likely, it is the presence of inert additives in the formulation of some pesticides that provides nutrients for microbial growth. Thus, failure in good management of pesticides at the farm level could cause them to increase the risk of vegetables becoming contaminated with microorganisms of spoilage and public health significance.

Future research should focus on the survival of microbial populations from reconstituted pesticides after application to vegetables. This has been considered by Guan et al.

(2005) for the pathogenic bacteria, Salmonella spp. and E. coli O157:H7. Salmonella

and E. coli O157:H7 were observed to survive on the surface of tomato plants for > 45h and < 15 days, respectively. Further research could determine the effect of pesticide preparations on the natural microflora of vegetables as well as the survival of non- pathogenic microbial species on the surface of vegetables.

In conclusion, this study has demonstrated that some pesticide preparations, commonly used during the cultivation of vegetables could, unintentionally, add to the microbial contamination of such produce, thereby impacting on their shelf-life and public health safety. Three strategies are proposed to minimise this increased microbial risk: (i) design pesticide formulations with “inerts” or additives that do not support the growth of microorganisms, (ii) reconstitute pesticides in potable water rather than sources of agricultural water, and (iii) educate farmers to apply pesticides immediately after reconstitution, and to minimise storage conditions that would encourage microbial growth in these preparations

CHAPTER 4

THE BACTERIAL ECOLOGY OF LETTUCE AS

DETERMINED BY CULTURAL AND PCR-DENATURING

GRADIENT GEL ELECTROPHORESIS (DGGE)

ANALYSIS

4.1 Introduction

In recent years, there has been increased consumer demand for fresh salad vegetables, leading to the development of a diverse range of ready-to-eat products (Sloan 2000;

Pivarnik et al. 2005). These products are only minimally processed (washed, chopped and packaged) and, consequently, their microbial ecology is significant in determining product quality, safety and shelf life (Nguyen-the and Carlin 2000; Heard 2002;

Johnston et al. 2006; Mukherjee et al. 2006). Lettuces are major components of many salad products, but there are few detailed studies of their general microbial ecology.

Studies on the microbiology of lettuce have mainly focussed on the occurrence and populations of major microbial groups such as total viable bacteria, yeast and moulds, lactic acid bacteria and coliforms (Saddik et al. 1985; Albrecht et al. 1995; Pingulkar et al. 2001; Viswanathan and Kaur 2001; Thunberg et al. 2002). Some of the main bacterial species associated with lettuce leaves have been mentioned in studies by Riser et al. (1984), Magnuson et al. (1990), King Jr et al. (1991), Khan et al. (1992), Freire and Robbs (2000), and Hamilton-Miller and Shah (2001), but the data are only non- quantitative reports of the isolates found. Predominant isolates included species of 78

Pseudomonas, Serratia, Erwinia, Enterobacter, Pantoea, Burkholderia and

Stenotrophomonas. Overall, the ecological data are incomplete and quantitative information is needed on the populations (cfu/g) of the individual species present.

Plant products in particular, may contain antimicrobial constituents that could impact on the reliability of their microbiological analysis, and influence conclusions about their ecology. Such constituents have been reported in cabbage (Pederson and Fisher 1944a,

1944b; Kyung and Fleming 1994), carrots (Beuchat and Brackett 1990; Nguyen-the and

Lund 1992), and cauliflower (Brandi et al. 2006). Their influences become significant when plant macerates or homogenates are used as the starting material for microbiological analysis. In these circumstances, the full impact of plant components on microorganism survival could be encountered. These effects may be minimised if surface rinses of the plant products are used as the starting material for microbiological analysis. However, in this approach there is always the risk that a proportion of the microbial population might not be dislodged from the product surface and, therefore, escape detection. The surfactant, Tween 80, can be added to the diluents to facilitate microbial detachment from the plant surface (Lillard 1988a; Adams et al. 1989; Eginton et al. 1998; Raiden et al. 2003; do Socorro Rocha Bastos et al. 2005; El-Baradei et al.

2007; Nielsen et al. 2007).

In recent years, the efficacy of culture methods for assaying the microbial ecology of many habitats including foods has been questioned. Many viable microorganisms may not be recovered with conventional culture, leading to an under estimation of the total ecology of the habitat (Muyzer and Smalla 1998; Fleet 1999; Giraffa 2004; De Vero et al. 2006). Molecular methods such as PCR-denaturing gradient gel electrophoresis

(DGGE) or PCR-temperature gradient gel electrophoresis (TGGE) that analyse

extracted DNA offer culture independent alternatives to microbiological analyses and are attracting increasing use in the field of food microbiology (Ercolini et al. 2003b;

Miambi et al. 2003; Cocolin et al. 2001a, 2004; Ercolini 2004; Giraffa 2004; Handschur et al. 2005; De Vero et al. 2006; Li et al. 2006). In this approach, PCR is used to specifically amplify microbial DNA in the food extract, and the DNA amplicons are then separated by DGGE/TGGE and identified by sequencing to give a genus or species name (Muyzer et al. 1993; Muyzer and Smalla 1998; Muyzer 1999). Using this method, various microbial species have been detected in food and beverage ecosystems, but not found by culture (Fasoli et al. 2003; Miambi et al. 2003; Temmerman et al. 2003;

Giraffa 2004; Prakitchaiwattana et al. 2004; De Vero et al. 2006). Nevertheless, this strategy is not without limitations because it sometimes fails to detect species found by cultural isolation. The efficacy of DNA extraction and performance of the PCR are two key steps in this molecular approach and both can be impacted by the food matrix, such as plant constituents (Lantz et al. 1994; von Wintzingerode et al. 1997; Wilson 1997;

Maukon et al. 2003; Ercolini 2004).

Information on the microbiology of lettuces, so far, is based on cultural analyses of macerates and homogenates. Their microbiology requires more critical investigation using both cultural and molecular methods, including consideration of any inhibitory influences of the matrix composition.

This chapter reports the major bacterial species found in lettuce as determined by the use of culture and PCR-DGGE methods, and compares the effect of maceration and rinsing of lettuce samples on the recovery of predominant bacterial species by these two methods.

4.2 Materials and Methods

4.2.1 Lettuce samples

Lettuces of the Iceberg variety were randomly purchased as unpackaged “heads” from supermarkets in Sydney. They were stored at 4°C and analysed within 24 h of purchase.

4.2.2 Bacteriological analysis of lettuce

Bacteria on lettuce were determined by (i) direct culture plating on agar media and (ii)

PCR-DGGE analysis as outlined in Figure 4.1. Samples of lettuce (25 g), comprising sections of inner and outer leaves, were aseptically taken and transferred to (i) 225 ml of

0.1% Bacteriological Peptone water (Oxoid, Adelaide, Australia) or; (ii) 225 ml of 0.1%

Bacteriological Peptone water containing 0.01% Tween 80 (Sigma-Aldrich, Australia).

These samples were subjected to either maceration or rinsing to bring associated microbial cells into suspension. For maceration, the mixture was placed in a filter-lined stomacher bag (Seward) and homogenised in a Stomacher (Model BA6021, Seward,

Norfolk, UK) for one minute. For rinsing, the mixture was transferred to a 500 ml conical flask and shaken on an orbital shaker (150 rpm, Model P03422, Paton

Industries) for 30 min. The mixture was then poured through a filter stomacher bag to remove the solid material from the rinse, which was used for analysis.

Lettuce (25 g)

Diluent 225 ml (0.1% Bacteriological peptone with/without Tween 80)

Maceration Rinse Maceration Rinse with with Tween 80 Tween 80

Supernatant

Culture analysis PCR-DGGE

Serial dilution Sediment microbial cells

Plating onto Tryptone Extraction of total DNA Soya Agar (TSA)

Specific amplification of Incubation (30°C/48h) bacterial rDNA (PCR)

Colony counts and Separation of DNA identification of colonies amplicons by DGGE by DNA sequencing

Identification of bacterial species by sequencing of amplicons

Figure 4.1 Outline of protocol for the analysis of bacteria on lettuce

4.2.3 Culture on agar media

Homogenates and rinses of lettuce samples were serially diluted in 0.1% Bacteriological

Peptone and duplicate samples (0.1 ml) of each dilution were spread inoculated onto the surface of plates of Tryptone Soya Agar (TSA, Oxoid). The plates were incubated for

48 hrs at 30°C and the different colony types noted and counted. Representatives of the different colony types were selected and purified by streaking onto Nutrient Agar (NA,

Oxoid) and maintained at 4°C for identification. They were identified by observation of cellular morphology, Gram stain, oxidase and catalase reactions, Hugh and Leifson’s oxidation/reduction, motility tests, plus the ability to grow aerobically and anaerobically at 30°C. Predominant species were further identified by sequencing sections of the 16S rDNA gene.

4.2.4 DNA extraction for bacterial identification and PCR-DGGE analysis

Pellets of microbial cells for DNA extraction were prepared from pure culture isolates and from homogenates and rinses of lettuce samples. Isolates were grown up in

Tryptone Soya Broth (TSB, Oxoid) for 24 h at 30°C, from which 1 ml was taken and centrifuged at 12,000g (10 min, 4°C) to give the pellet of cells for DNA extraction.

Homogenates (225 ml) and rinses (225 ml) of lettuce samples were centrifuged (Avanti

J-E, Beckman Coulter Refrigerated Centrifuge, Beckman-Coulter Inc., California USA) at 12,000g (15 min, 4°C) to produce a cell pellet which was used for DNA extraction.

All microbial pellets were suspended in 0.5 ml of extraction buffer (100 mM Tris-HCl

[pH 8], 50 mM EDTA [pH 8], 100 mM NaCl, 1% sodium dodecyl sulphate) and 0.5 ml of phenol:chloroform:isoamyl alcohol (24:24:1; Sigma-Aldrich) containing 0.3 g of

zirconia/silica beads (diameter 0.1 mm; Daintree Scientific, Tasmania, Australia). The cells were shaken at 5000 rpm for 30 s in a mini-beadbeater (Biospec Products, OK,

USA). The homogenate was centrifuged for 10 min (15,000g at 4°C) and 0.5 ml of the upper layer of the supernatant, which contains the DNA, was transferred to a sterile

Eppendorf tube and 0.5 ml phenol:chloroform:isoamyl alcohol added. The solutions were mixed thoroughly and the tube centrifuged for 10 min (15,000g at 4°C). An aliquot (0.4 ml) of the upper layer of the supernatant (containing the DNA), was transferred to a sterile Eppendorf tube and the DNA precipitated by adding 0.6 ml of isopropanol (Sigma-Aldrich) and placing the tube in a freezer (-20°C) for 24 h. After storage and thawing, the tube was centrifuged for 10 min (15,000g) and the supernatant discarded. The pellet (DNA) was washed by addition of 1 ml of 70% ethanol and mixing of the tube. The mixture was centrifuged (10,000g for 10 min), the supernatant discarded and the pellet allowed to dry (room temperature, 10-15 min). The dried pellet was dissolved in 50 µl of TE buffer (10 mM Tris-HCl, 1 mM EDTA [pH 8]) and stored in a freezer at -20°C until required.

4.2.5 PCR amplification of DNA for sequence identification and for PCR-DGGE analysis

Purified DNA from culture isolates was amplified with two universal eubacterial primers, 341F (5’-CCTACGGGAGGCAGCAG-3’) and 907R (5’-CCGTCAATTCCTT

TRAGTTT-3’) which were obtained from Sigma-Genosys Australia Pty Ltd. The PCR mix contained 1X PCR buffer (10 mM Tris-HCl [pH 8.3], 50 mM KCl), 200 µM of each dNTP (Roche Diagnostics Corp., IN, USA), 0.2 µM of each primer, 1.5 mM

MgCl2, 1.25 U of Taq DNA polymerase (AmpliTaq™, Roche Molecular Systems, New

Jersey, USA) and ~100 ng of template DNA in 50 µl final volume.

Amplification was carried out with a GeneAmp® PCR system 9700 (Applied

Biosystems, CA, USA) with the following cycling program: initial denaturation at 94°C for 5 min, 20 touchdown cycles at 94°C for 0.5 min, 60°C for 1 min and 72°C for 1.5 min, with a decrease in annealing temperature of 0.5°C every cycle, followed by 10 cycles at 94°C for 0.5 min, 50°C for 1 min and 72°C for 1.5 min, and a final extension at 72°C for 10 min.

To investigate the bacterial communities by PCR-DGGE, purified DNA extracted from homogenates and rinses of lettuces was amplified with primers 341F-GC/518R (Muyzer et al. 1993). The 40 nucleotide GC rich sequence (5’-CGCCCGCCGCGCCCC

GCGCCCGTCCCGCCGCCCCGCCCG -3’) was attached to the 5’ end of the forward primer, 341F to facilitate the detection of sequence variations of amplified DNA fragments by subsequent DGGE (Myers et al. 1985; Muyzer et al. 1993). The primers,

341F-GC (5’-GC-clamp-CCTACGGGAGGCAGCAG-3’) and 518R (5’-

ATAACCGCGGCTGCTGG-3’) target the V3 region of the 16S ribosomal RNA gene

(Muyzer et al. 1993). The PCR mix contained 1X PCR buffer (10 mM Tris-HCl [pH

8.3], 50 mM KCl), 200 µM of each dNTP (Roche Diagnostics Corp.), 0.2 µM of each primer, 1.5 mM MgCl2, 1.25 U of Taq DNA polymerase (AmpliTaq™, Roche

Molecular Systems) and ~100 ng of template DNA in 50 µl final volume. Amplification was performed on a GeneAmp® PCR system 9700 (Applied Biosystems) using the following program: initial denaturation at 94°C for 5 min, 20 touchdown cycles at 94°C for 0.5 min, 65°C for 1 min and 72°C for 1 min, with a decrease in annealing

temperature of 0.5°C every cycle, followed by 10 cycles at 94°C for 0.5 min, 55°C for 1 min and 72°C for 1 min, and a final extension at 72°C for 10 min.

Presence of PCR products was confirmed by agarose gel electrophoresis of 2-10 µl of sample loaded into a 1.5% agarose gel and run at 120V for 40 min. The gel was then stained in ethidium bromide (1.0 µg/ml) for 10-15 min and visualised on a UV illuminator (300 nm).

4.2.6 Denaturing gradient gel electrophoresis (DGGE)

The PCR amplicons were separated by DGGE using a D-Code Universal Mutation

Detection System (Bio-Rad, Hercules, CA, USA). Essentially, DGGE was performed as described by Muyzer et al. (1993). Eight percent polyacrylamide gel (19:1 acrylamide:bisacrylamide, Bio-Rad) with 30-60% linear denaturing gradient (100% denaturant corresponding to 7M urea and 40% formamide) were used for analysing amplicons produced with the 341F-GC/518R primers. Electrophoresis was performed for 10 min at 20V and for 6 h at 120V with a constant temperature of 60°C in 1X TAE buffer (20 mM Tris-acetate, pH 7.4, 10 mM sodium acetate, 0.5 mM EDTA).

Subsequently, gels were stained for 15-20 mins in 1X TAE buffer containing SYBR green dye (10 µl/100 ml TAE), and then visualised on a UV illuminator (300 nm) and photographed with a Polaroid DS-34 camera.

DNA bands in DGGE gels were excised and incubated overnight in 100 µl of sterile

Milli-Q water at 4°C. Eluted DNA was reamplified with corresponding primers without the GC-Clamp. Amplified products were purified using a QIAquick PCR purification kit (Qiagen Pty Ltd, Clifton Hill, Australia), and subjected to sequence analysis.

4.2.7 Sequence analysis of rDNA for bacterial identification and DGGE bands

PCR amplicons of DNA from bacterial isolates and from band eluates of DGGE gels were identified by sequence analysis of the 16S ribosomal DNA using the ABIPRISM®

BigDye™ Terminators V3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City,

CA, USA). The products were analysed for their sequence at the Automated DNA

Analysis Facility, School of Biotechnology and Biomolecular Science, University of

New South Wales. The sequences were aligned to determine the closest known relatives of partial rDNA gene fragments available from BLAST in the GENBANK database

(Karlin and Altschul 1990).

4.2.8 Bacterial species used as reference cultures

Reference strains of bacteria used in this study are listed in Table 4.1 and were cultured and maintained on NA. Their identities were confirmed by sequencing of the V3-V6 region of 16S rDNA as described already. Bacteria were grown in 5 ml of Nutrient

Broth at 30°C for 24 h. Culture (1.0 ml) was transferred to a 1.5 ml cryogenic tube and centrifuged at 16,000g for 10 min at 4°C to sediment the bacterial cells. Cell pellets were stored at -20°C until extraction of DNA for determination of rDNA sequence or use in PCR-DGGE analyses.

In some experiments, cell suspensions containing bacterial species at different concentrations were prepared. Bacteria were grown in Nutrient Broth as described and then serially diluted with 0.1% Bacteriological Peptone water. The populations (cfu/ml)

of bacterial species in the dilutions were determined by spread plate culture on Plate

Count Agar (PCA, Oxoid).

Table 4.1 Bacterial cultures used in the evaluation of PCR-DGGE and confirmed identity by sequencing

Bacterial species Source Confirmed identity a Acinetobacter johnsonii* School of Chemical Sciences and 97% Engineering, UNSW Agrobacterium School of Chemical Sciences and 97% larrymoorei* Engineering, UNSW Burkholderia cepacia* School of Chemical Sciences and 98% Engineering, UNSW Curtobacterium School of Chemical Sciences and 98% flaccumfaciens* Engineering, UNSW Escherichia coli (ATCC School of Biotechnology and E. coli (98%) 11775) Biomolecular Sciences, UNSW Escherichia coli School of Biotechnology and E. coli (98%) O111:NM Biomolecular Sciences, UNSW Escherichia coli O157:H7 School of Biotechnology and E. coli (98%) (ATCC 35150) Biomolecular Sciences, UNSW Listeria monocytogenes 1768 serotype 1/2a, TECRA L. monocytogenes (98%) (L2) Diagnostics, Sydney Listeria monocytogenes 1771 serotype 4c, TECRA L. monocytogenes (97%) (L4) (ATCC 19116) Diagnostics, Sydney Listeria monocytogenes 1773 serotype 4e, TECRA L. monocytogenes (97%) (L5) (ATCC 19118) Diagnostics, Sydney Pantoea stewartii* School of Chemical Sciences and 97% Engineering, UNSW Pseudomonas fluorescens School of Biotechnology and Ps. fluorescens (99%) (UNSW 022600) Biomolecular Sciences, UNSW Pseudomonas fluorescens School of Biotechnology and Ps. fluorescens (98%) (ATCC 13525) Biomolecular Sciences, UNSW Pseudomonas fragi School of Chemical Sciences and Pseudomonas sp. (97%) Engineering, UNSW Pseudomonas School of Chemical Sciences and 97% pavonaceae* Engineering, UNSW Salmonella enteritidis School of Biotechnology and S. enteritidis (97%) (UNSW 031901) Biomolecular Sciences, UNSW Salmonella typhimurium School of Chemical Sciences and S. typhimurium (97%) Engineering, UNSW

* These bacterial species were isolated during this study a confirmed identity by sequencing of the 16S rDNA, (% match according to the BLAST database) 88

4.2.9 Detection limit of bacterial populations by PCR-DGGE

Reference strains of bacteria (Table 4.1) were individually cultured to late exponential phase (approx. 108 cfu/ml) and diluted in 0.1% Bacteriological Peptone water to give final populations of approximately 100, 101, 102, 103, 104, 105, 106, 107, and 108 cfu/ml.

Samples (1 ml) of the different dilutions were processed for DNA extraction, PCR and

DGGE as described already.

4.2.10 Detection limit of bacterial populations inoculated onto lettuce and analysed using PCR-DGGE

Samples of lettuce (25 g), comprising sections of inner and outer leaves, were aseptically taken and transferred to 225 ml of 0.1% Bacteriological Peptone water.

Reference strains of bacteria (Table 4.1) were prepared and added to the mixtures to give final populations of 102, 104 and 106 cfu/g. The samples were macerated or rinsed as described previously, and the supernatant was centrifuged at 10,000 rpm for 10-15 min at 4°C. The supernatant was discarded and the cell pellet transferred to a sterile cryo-vial. The DNA was extracted and the samples analysed by nested and one-step

PCR. For the nested PCR, the primers 27F (5’-GAGTTTGATCCTGGCTCAG-3’) and

907R (5’-CCGTCAATTCCTTRAGTTT-3’) were used (Schabereiter-Gurtner et al.

2001; Egert and Friedrich 2003). The PCR mix contained 1X PCR buffer (10 mM Tris-

HCl [pH 8.3], 50 mM KCl), 200 µM of each dNTP (Roche Diagnostics Corp.), 0.2 µM of each primer, 1.5 mM MgCl2, 1.25 U of Taq DNA polymerase (AmpliTaq™, Roche

Molecular Systems) and ~100 ng of template DNA in 50 µl final volume. Amplification was performed using the following program: initial denaturation at 94°C for 5 min, 20 89

touchdown cycles at 94°C for 0.5 min, 60°C for 1 min and 72°C for 2 min, with a decrease in annealing temperature of 0.5°C every cycle, followed by 10 cycles at 94°C for 0.5 min, 50°C for 1 min and 72°C for 2 min, and a final extension at 72°C for 10 min. After checking for the presence of PCR products, 1-2 µl of product was taken and amplified using the primers 341F-GC and 518R as described for the one-step PCR. The

PCR amplicons for the one-step and nested procedures were then analysed by DGGE.

4.2.11 Statistical analysis of data

Statistical analyses were performed using Microsoft Excel. One-way ANOVA tests and t-tests were done to derive the statistical differences (p<0.05) of microbial populations between maceration and rinse, maceration with Tween 80 and rinse with Tween 80, maceration and maceration with Tween 80 and rinse and rinse with Tween 80.

4.3 Results

4.3.1 Effect of maceration, rinsing and Tween 80 on recovery of bacterial populations from lettuce samples

Sub-samples of the same head of lettuce were analysed for total viable counts in rinses and macerates, prepared with or without the addition of Tween 80. Counts were performed by plating onto TSA (Table 4.2).

4.3.2 Comparison of bacterial populations obtained by maceration and rinsing of lettuce samples

A total of 26 lettuce sub-samples were compared for their bacterial populations after preparation by maceration and by rinsing. Thirteen of these sub-samples were prepared in the presence of Tween 80 and another 13 were prepared in the absence of Tween 80.

The influence of Tween 80 on the population data will be discussed in the next section.

Populations recovered from the sub-samples ranged from not detectable (<100 cfu/g) to

2.4 x 106 cfu/g (Table 4.2). For 13 of the sub-samples, populations found after maceration and rinses were not significantly different (p<0.05). For seven of the 26 lettuce sub-samples, rinsing gave higher populations than maceration, and for six of the sub-samples, maceration gave higher populations than rinsing. In several of these samples (e.g. samples 3, 4, 5, 6 & 12), there were five to 10-fold differences between the two populations.

Table 4.2 Bacterial populations recovered from samples of Iceberg lettuce by maceration and rinse techniques with or without the addition of Tween 80

Populations (cfu/g) Sample Maceration Maceration + Rinse Rinse + Tween Tween 80 80 1 2.4 x 106 aA* 1.3 x 106 aA** 2.0 x 106 aA* 1.3 x 106 aA** 2 9.8 x 105 aA 1.1 x 106 aB 9.6 x 105 aA 8.8 x 105 aA 3 2.5 x 105 aA 3.1 x 106 aB 6.4 x 105 bA 1.1 x 106 bB 4 1.9 x 103 aA 6.1 x 103 aB 1.6 x 104 bA 3.3 x 104 bB 5 2.4 x 103 aA 1.4 x 103 aA ND bA 1.0 x 102 bA 6 1.0 x 102 aA 8.2 x 103 aB ND aA 1.4 x 103 bB 7 ND aA 1.5 x 102 aB 2.5 x 102 aA 2.2 x 103 bB 8 5.2 x 104 aA 1.8 x 104 aA 5.9 x 104 aA 2.1 x 105 bB 9 5.0 x 103 aA 9.4 x 103 aB 6.6 x 105 bA 1.0 x 105 bB 10 2.5 x 104 aA 9.0 x 104 aB 1.8 x 104 aA 5.7 x 104 aB 11 9.0 x 104 aA 1.7 x 105 aB 7.6 x 104 aA 1.4 x 105 aB 12 2.7 x 105 aA 2.8 x 105 aA 7.4 x 104 bA 5.1 x 105 aB 13 4.7 x 105 aA 4.7 x 105 aA 2.3 x 105 aA 1.0 x 105 bB

ND, No bacterial species detected in lettuce (<100 cfu/g) * Different lower case superscripts denote statistically significant differences (p<0.05) between maceration and rinse **Different upper case superscripts denote statistically significant differences (p<0.05) between no Tween 80 and the addition of Tween 80

4.3.3 Effect of addition of Tween 80 on bacterial populations obtained from maceration and rinses of lettuce samples

Thirteen lettuce samples were examined for their viable bacterial population by maceration with peptone water and by maceration with peptone water to which Tween

80 had been added (Table 4.2). Eight of the 13 samples gave higher populations when

Tween 80 was added to the maceration solution. In one sample (sample 7), no viable populations (<100 cfu/g) were detected by maceration with peptone water with no added Tween 80. For two samples (samples 3 and 6), greater then 10-fold higher

populations were observed when Tween 80 was added to the solution used for maceration.

Thirteen lettuce samples were examined for their viable bacterial populations by rinsing with peptone water and by rinsing with peptone water to which Tween 80 had been added (Table 4.2). Nine of the 13 samples gave higher counts when Tween 80 was added to the rinse solution. For two samples (samples 5 and 6), no viable population

(<100 cfu/g) were detected by rinsing with peptone water without Tween 80 but populations of 102-103 cfu/g were recovered when Tween 80 was added to the rinse solution. In another sample (sample 7), addition of Tween 80 to the rinse resulted in recovery of a 10-fold higher population. It could be considered that the addition of

Tween 80 surfactant to the rinse facilitated the removal of microorganisms from the surface of lettuce.

In total, 17 samples out of 26 samples (65%) analysed by maceration or rinse with and without the addition of Tween 80 gave significantly different total populations (P<0.05) which were higher in the presence of Tween 80. This result shows that the addition of

Tween 80 to the solution for macerates and rinses has a significant impact on the total bacterial populations recovered from samples of Iceberg lettuce.

4.3.4 Bacterial species isolated from lettuce by agar plating

Bacterial colonies on spread plates of TSA were carefully examined for colony morphology using a stereomicroscope and for cell morphology using phase contrast microscopy. The most predominant colony morphologies were recorded and representatives of these were isolated and purified by streaking on TSA. These isolates

were further screened and grouped on the basis of several biochemical and physiological phenotypes (Gram stain, oxidase, catalase, Hugh and Leifson’s oxidative/fermentative tests) (Table 4.3). Representatives of each group were then selected for identification by sequencing of regions of the rDNA gene (Table 4.3).

Isolates and identifications were made for lettuce samples analysed by rinsing and maceration methods, with and without the addition of Tween 80 to the maceration and rinsing medium (Tables 4.4, 4.5, 4.6, and 4.7)

A diversity of 38 bacterial species was isolated from the 13 different samples of lettuce.

None of the samples gave the same profile of isolates although a few species (e.g.

Burkholderia cepacia, Curtobacterium flaccumfaciens) were isolated from three or more of the samples.

The most frequently isolated bacterial species from lettuce samples, independent of the sampling method, were Pseudomonas synxantha/fluorescens, Curt. flaccumfaciens,

Burk. cepacia, Agrobacterium larrymoorei and Stenotrophomonas maltophilia (Table

4.8).

Table 4.3 Properties used to identify predominant bacterial species isolated from Iceberg lettuce

rDNA sequence Oxidative/ % match to Bacterial species Gram stain Motility Catalase Oxidase fermentative BLAST database Acinetobacter +/- cocci No Oxidative + - 99% baumannii Acinetobacter +ve cocci No Non-ferment + - 98% calcoaceticus Acinetobacter -ve cocci/ No Non-ferment + - 99% johnsonii short rod Acinetobacter -ve No Oxidative + - 98% rhizosphaerae cocci/rod Agrobacterium -ve short No Oxidative + - 98% larrymoorei rod Burkholderia -ve rod No Oxidative + + 98% cepacia Curtobacterium +ve rod No Oxidative + - 98% flaccumfaciens Pantoea -ve rod Yes Fermentative + - 98% agglomerans Pantoea ananatis -ve rod Yes Fermentative + - 99% Pantoea ananatis/ -ve rod Yes Fermentative + - 98% agglomerans Pantoea stewartii -ve rod Yes Fermentative + - 98% Pseudomonas -ve rod Yes Oxidative + + 98% cichorii Pseudomonas -ve rod Yes Oxidative + + 98% flavescens Pseudomonas -ve rod Yes Oxidative + + 99% fluorescens Pseudomonas -ve rod Yes Oxidative + + 98% migulae Pseudomonas -ve rod Yes Oxidative + + 98% pavonaceae Pseudomonas putida -ve rod Yes Oxidative + - 98% Pseudomonas -ve rod Yes Oxidative + - 98% straminea Pseudomonas -ve rod Yes Oxidative + + 98% synxantha/ fluorescens Pseudomonas -ve rod Yes Oxidative + + 98% syringae Pseudomonas -ve rod Yes Oxidative + - 98% viridiflava Serratia marcescens -ve rod Yes Fermentative + - 98% Serratia -ve rod Yes Fermentative + - 98% proteamaculans Stenotrophomonas -ve rod No Non-ferment + + 99% maltophilia Zoogloea ramigera -ve rod No Non-ferment + - 98%

Non-ferment, isolate did not ferment glucose 95

Table 4.4 Predominant bacterial species isolated from samples of Iceberg lettuce by maceration and rinse techniques and plating onto TSA

Sample Maceration a Rinse a 1 (i) Pseudomonas pavonaceae (2.3 x 106) (i) Agrobacterium larrymoorei (1.0 x 106) (ii) Agrobacterium larrymoorei (5.0 x 104) (ii) Pseudomonas pavonaceae (7.5 x 105) (iii) Stenotrophomonas maltophilia (2.0 x 104) 2 (i) Pseudomonas cichorii (6.5 x 105) (i) Pseudomonas cichorii (4.5 x 105) (ii) Zoogloea ramigera (4.0 x 104) (ii) Pseudomonas putida (6.0 x 104) (iii) Pseudomonas syringae (7.5 x 104) (iii) Zoogloea ramigera (4.0 x 104) 3 (i) Agrobacterium larrymoorei (1.0 x 105) (i) Agrobacterium larrymoorei (9.8 x 105) (ii) Pseudomonas straminea (9.0 x 104) (ii) Pantoea stewartii (2.7 x 105) (iii) Pantoea stewartii (3.5 x 104) (iii) Pseudomonas straminea (4.0 x 104) 4 (i) Acinetobacter rhizosphaerae (1.7 x 104) (i) Microbacterium testaceum (1.0 x 104) (ii) Curtobacterium flaccumfaciens (1.8 x 103) (ii) Curtobacterium flaccumfaciens (iii) Bacillus humi (1.0 x 102) (4.2 x 103) 5 (i) Burkholderia cepacia (1.9 x 103) ND (ii) Pantoea agglomerans (2.0 x 102) 6 (i) Pseudomonas putida (1.0 x 102) ND (i) Pantoea agglomerans (1.0 x 102) 7 ND (i) Pseudomonas synxantha/fluorescens (1.5 x 102) (ii) Pseudomonas sp. (1.0 x 102) 8 (i) Curtobacterium flaccumfaciens (2.8 x 104) (i) Pseudomonas viridiflava (1.9 x 104) (ii) Burkholderia cepacia (4.7 x 103) (ii) Curtobacterium flaccumfaciens (iii) Erwinia rhapontici (3.2 x 103) (9.5 x 103) (ii) Pseudomonas synxantha/fluorescens (9.5 x 103) 9 (i) Burkholderia cepacia (1.7 x 103) (i) Burkholderia cepacia (4.1 x 105) (ii) Pseudomonas synxantha/fluorescens (ii) Pseudomonas synxantha/fluorescens (1.2 x 103) (8.0 x 104) (iii) Curtobacterium flaccumfaciens (1.0 x 103) 10 (i) Curtobacterium flaccumfaciens (1.0 x 104) (i) Burkholderia cepacia (6.5 x 103) (ii) Burkholderia cepacia (5.4 x 103) (ii) Stenotrophomonas maltophilia (1.5 x 103) (iii) Pseudomonas viridiflava (2.9 x 103) 11 (i) Serratia marcescens (3.6 x 104) (i) Serratia marcescens (1.2 x 104) (ii) Pantoea ananatis/agglomerans (9.0 x 103) (ii) Pseudomonas fluorescens (9.0 x 103) (iii) Acinetobacter johnsonii (1.0 x 103) (iii) Flavobacterium mizutaii (4.0 x 103) 12 (i) Pseudomonas flavescens (7.5 x 104) (i) Serratia marcescens (3.2 x 104) (ii) Acinetobacter baumannii (2.2 x 104) (ii) Pseudomonas synxantha/fluorescens (iii) Acinetobacter johnsonii (2.0 x 104) (5.5 x 103) (iii) Acinetobacter johnsonii (5.0 x 103) 13 (i) Serratia marcescens (3.1 x 105) (i) Serratia marcescens (1.2 x 105) (ii) Stenotrophomonas maltophilia (9.0 x 104) (ii) Pseudomonas putida (4.0 x 104) (iii) Pantoea ananatis (2.0 x 104) a Population (cfu/g) of species is shown in brackets after the identity ND- No bacterial species detected in lettuce (<100 cfu/g) Bold type within table represents species detected by both sampling methods.

Table 4.5 Predominant bacterial species isolated from samples of Iceberg lettuce by maceration with Tween 80 and rinse with Tween 80 and plating onto TSA

Sample Maceration + Tween a Rinse + Tween a 1 (i) Pseudomonas pavonaceae (1.0 x 106) (i) Pseudomonas pavonaceae (6.6 x 105) (ii) Agrobacterium larrymoorei (7.5 x 104) (ii) Agrobacterium larrymoorei (4.6 x 105) (iii) Microbacterium testaceum (4.0 x 104) 2 (i) Pseudomonas cichorii (1.1 x 105) (i) Pseudomonas cichorii (6.0 x 105) (ii) Zoogloea ramigera (9.5 x 104) (ii) Pseudomonas syringae (6.0 x 104) (iii) Pseudomonas fragi (4.0 x 104) (iii) Zoogloea ramigera (4.0 x 104) 3 (i) Agrobacterium larrymoorei (2.0 x 106) (i) Agrobacterium larrymoorei (4.0 x 105) (ii) Pantoea stewartii (6.5 x 105) (ii) Pantoea stewartii (3.0 x 105) (iii) Pseudomonas straminea (2.3 x 105) (iii) Pseudomonas straminea (2.2 x 105) 4 (i) Pseudomonas migulae (2.9 x 103) (i) Microbacterium testaceum (2.5 x 104) (ii) Curtobacterium flaccumfaciens (ii) Curtobacterium flaccumfaciens (2.2 x 103) (5.0 x 103) (iii) Chryseobacterium scophthalan (5.0 x 102) 5 (i) Pseudomonas synxantha/fluorescens (i) Curtobacterium herbarum (1.0 x 102) (8.0 x 102) (ii) Curtobacterium herbarum (3.5 x 102) (iii) Serratia proteamaculans (2.0 x 102) 6 (i) Pseudomonas synxantha/fluorescens (i) Arthrobacter sp. (4.0 x 102) (7.4 x 103) (ii) Microbacterium keratanolyticum (ii) Serratia proteamaculans (8.0 x 102) (3.5 x 102) (iii) Planomicrobium okeanokoites (2.0 x 102) 7 (i) Pseudomonas synxantha/fluorescens (i) Pseudomonas synxantha/fluorescens (1.0 x 102) (1.1 x 103) (ii) Arthrobacter sp. (1.0 x 102) (ii) Burkholderia cepacia (2.0 x 102) 8 (i) Curtobacterium flaccumfaciens (5.9 x (i) Pseudomonas synxantha/fluorescens 103) (1.5 x 105) (ii) Pseudomonas fluorescens (4.3 x 103) (ii) Pseudomonas putida (8.8 x 104) (iii) Burkholderia cepacia (3.0 x 103) 9 (i) Pseudomonas synxantha/fluorescens (i) Pseudomonas synxantha/fluorescens (2.8 x 103) (2.2 x 104) (ii) Pseudomonas sp. (1.4 x 103) (ii) Stenotrophomonas maltophilia (1.0 x 104) (iii) Pseudomonas putida (9.5 x 102) 10 (i) Curtobacterium flaccumfaciens (i) Pseudomonas synxantha/fluorescens (1.7 x 104) (1.2 x 104) (ii) Burkholderia cepacia (8.0 x 103) (ii) Pseudomonas viridiflava (1.0 x 104) (iii) Pseudomonas synxantha/fluorescens (7.5 x 103) 11 (i) Stenotrophomonas maltophilia (i) Klebsiella sp. (3.9 x 104) (1.8 x 104) (ii) Stenotrophomonas maltophilia (ii) Buttiauxella noackiae/agrestis (9.5 x 103) (2.5 x 104) (iii) Chryseobacterium balustinum (8.5 x 103) 12 (i) Stenotrophomonas maltophilia (6.5 x 104) (i) Serratia marcescens (3.4 x 105) (ii) Acinetobacter johnsonii (4.0 x 104) (ii) Pseudomonas sp. (4.0 x 104) (ii) Delftia acidovorans (4.0 x 104) (iii) Acinetobacter haemolyticus (3.5 x 104) 13 (i) Acinetobacter johnsonii (1.1 x 105) (i) Serratia marcescens (1.4 x 104) (ii) Stenotrophomonas maltophilia (7.5 x 104) (ii) Pantoea ananatis (4.5 x 103) (iii) Pseudomonas synxantha/fluorescens (7.0 x 104) a Population (cfu/g) of species is shown in brackets after the identity Bold type within table represents species detected by both sampling methods.

Table 4.6 Predominant bacterial species isolated from samples of Iceberg lettuce by maceration and maceration with Tween 80 and plating onto TSA

Sample Maceration a Maceration + Tween a 1 (i) Pseudomonas pavonaceae (2.3 x 106) (i) Pseudomonas pavonaceae (1.0 x 106) (ii) Agrobacterium larrymoorei (5.0 x 104) (ii) Agrobacterium larrymoorei (7.5 x 104) 2 (i) Pseudomonas cichorii (6.5 x 105) (i) Pseudomonas cichorii (1.1 x 105) (ii) Pseudomonas syringae (7.5 x 104) (ii) Zoogloea ramigera (9.5 x 104) (iii) Zoogloea ramigera (4.0 x 104) (iii) Pseudomonas fragi (4.0 x 104) 3 (i) Agrobacterium larrymoorei (1.0 x 105) (i) Agrobacterium larrymoorei (2.0 x 106) (ii) Pseudomonas straminea (9.0 x 104) (ii) Pantoea stewartii (6.5 x 105) (iii) Pantoea stewartii (3.5 x 104) (iii) Pseudomonas straminea (2.3 x 105) 4 (i) Acinetobacter rhizosphaerae (1.7 x 104) (i) Pseudomonas migulae (2.9 x 103) (ii) Curtobacterium flaccumfaciens (ii) Curtobacterium flaccumfaciens (1.8 x 103) (2.2 x 103) (iii) Bacillus humi (1.0 x 102) 5 (i) Burkholderia cepacia (1.9 x 103) (i) Pseudomonas synxantha/fluorescens (ii) Pantoea agglomerans (2.0 x 102) (8.0 x 102) (ii) Curtobacterium herbarum (3.5 x 102) (iii) Serratia proteamaculans (2.0 x 102) 6 (i) Pseudomonas putida (1.0 x 102) (i) Pseudomonas synxantha/fluorescens (i) Pantoea agglomerans (1.0 x 102) (7.4 x 103) (ii) Serratia proteamaculans (8.0 x 102) 7 ND (i) Arthrobacter sp. (1.0 x 102) (ii) Pseudomonas synxantha/fluorescens (1.0 x 102) 8 (i) Curtobacterium flaccumfaciens (i) Curtobacterium flaccumfaciens (2.8 x 104) (5.9 x 103) (ii) Burkholderia cepacia (4.7 x 103) (ii) Pseudomonas fluorescens (4.3 x 103) (iii) Erwinia rhapontici (3.2 x 103) (iii) Burkholderia cepacia (3.0 x 103) 9 (i) Burkholderia cepacia (1.7 x 103) (i) Pseudomonas synxantha/fluorescens (ii) Pseudomonas synxantha/fluorescens (2.8 x 103) (1.2 x 103) (ii) Pseudomonas sp. (1.4 x 103) (iii) Curtobacterium flaccumfaciens (iii) Pseudomonas putida (9.5 x 102) (1.0 x 103) 10 (i) Curtobacterium flaccumfaciens (i) Curtobacterium flaccumfaciens (1.0 x 104) (1.7 x 104) (ii) Burkholderia cepacia (5.4 x 103) (ii) Burkholderia cepacia (8.0 x 103) (iii) Pseudomonas viridiflava (2.9 x 103) (iii) Pseudomonas synxantha/fluorescens (7.5 x 103) 11 (i) Serratia marcescens (3.6 x 104) (i) Stenotrophomonas maltophilia (1.8 x 104) (ii) Pantoea ananatis/agglomerans (ii) Buttiauxella noackiae/agrestis (9.5 x 103) (9.0 x 103) (iii) Chryseobacterium balustinum (8.5 x 103) (iii) Acinetobacter johnsonii (1.0 x 103) 12 (i) Pseudomonas flavescens (7.5 x 104) (i) Stenotrophomonas maltophilia (6.5 x 104) (ii) Acinetobacter baumannii (2.2 x 104) (ii) Acinetobacter johnsonii (4.0 x 104) (iii) Acinetobacter johnsonii (2.0 x 104) (ii) Delftia acidovorans (4.0 x 104) 13 (i) Serratia marcescens (3.1 x 105) (i) Stenotrophomonas maltophilia (7.5 x 104) (ii) Stenotrophomonas maltophilia (ii) Pseudomonas synxantha/fluorescens (9.0 x 104) (7.0 x 104) (iii) Acinetobacter johnsonii (1.1 x 105) a Population (cfu/g) of species is shown in brackets after the identity ND- No bacterial species detected in lettuce (<100 cfu/g) Bold type within table represents species detected by both sampling methods. 98

Table 4.7 Predominant bacterial species isolated from samples of Iceberg lettuce by rinsing and rinsing with Tween 80 and plating onto TSA

Sample Rinse a Rinse + Tween a 1 (i) Agrobacterium larrymoorei (1.0 x 106) (i) Pseudomonas pavonaceae (6.6 x 105) (ii) Pseudomonas pavonaceae (7.5 x 105) (ii) Agrobacterium larrymoorei (4.6 x 105) (iii) Stenotrophomonas maltophilia (2.0 x 104) (iii) Microbacterium testaceum (4.0 x 104) 2 (i) Pseudomonas cichorii (4.5 x 105) (i) Pseudomonas cichorii (6.0 x 105) (ii) Pseudomonas putida (6.0 x 104) (ii) Pseudomonas syringae (6.0 x 104) (iii) Zoogloea ramigera (4.0 x 104) (iii) Zoogloea ramigera (4.0 x 104) 3 (i) Agrobacterium larrymoorei (2.8 x 105) (i) Agrobacterium larrymoorei (4.0 x 105) (ii) Pantoea stewartii (2.7 x 105) (ii) Pantoea stewartii (3.0 x 105) (iii) Pseudomonas straminea (4.0 x 104) (iii) Pseudomonas straminea (2.2 x 105) 4 (i) Microbacterium testaceum (1.0 x 104) (i) Microbacterium testaceum (2.5 x 104) (ii) Curtobacterium flaccumfaciens (ii) Curtobacterium flaccumfaciens (4.3 x 103) (5.0 x 103) (iii) Chryseobacterium scophthalan (5.0 x 102) 5 ND (i) Curtobacterium herbarum (1.0 x 102) 6 ND (i) Arthrobacter sp. (4.0 x 102) (ii) Microbacterium keratanolyticum (3.5 x 102) (iii) Planomicrobium okeanokoites (2.0 x 102) 7 (i) Pseudomonas synxantha/fluorescens (i) Pseudomonas synxantha/fluorescens (1.5 x 102) (1.1 x 103) (ii) Pseudomonas sp. (1.0 x 102) (ii) Burkholderia cepacia (2.0 x 102) 8 (i) Pseudomonas viridiflava (1.9 x 104) (i) Pseudomonas synxantha/fluorescens (ii) Pseudomonas synxantha/fluorescens (1.5 x 105) (9.5 x 103) (ii) Pseudomonas putida (8.8 x 104) (iii) Curtobacterium flaccumfaciens (9.5 x 103) 9 (i) Burkholderia cepacia (4.1 x 105) (i) Pseudomonas synxantha/fluorescens (ii) Pseudomonas synxantha/fluorescens (2.2 x 104) (8.0 x 104) (ii) Stenotrophomonas maltophilia (1.0 x 104) 10 (i) Burkholderia cepacia (6.5 x 103) (i) Pseudomonas synxantha/fluorescens (ii) Stenotrophomonas maltophilia (1.5 x 103) (1.2 x 104) (ii) Pseudomonas viridiflava (1.0 x 104) 11 (i) Serratia marcescens (1.2 x 104) (i) Klebsiella sp. (3.9 x 104) (ii) Pseudomonas fluorescens (9.0 x 103) (ii) Stenotrophomonas maltophilia (2.5 x 104) (iii) Flavobacterium mizutaii (4.0 x 103) 12 (i) Serratia marcescens (3.2 x 104) (i) Serratia marcescens (3.4 x 105) (ii) Pseudomonas synxantha/fluorescens (ii) Pseudomonas sp. (4.0 x 104) (5.5 x 103) (iii) Acinetobacter haemolyticus (3.5 x 104) (iii) Acinetobacter johnsonii (5.0 x 103) 13 (i) Serratia marcescens (1.2 x 105) (i) Serratia marcescens (1.4 x 104) (ii) Pseudomonas putida (4.0 x 104) (ii) Pantoea ananatis (4.5 x 103) (iii) Pantoea ananatis (2.0 x 104) a Population (cfu/g) of species is shown in brackets after the identity ND- No bacterial species detected in lettuce (<100 cfu/g) Bold type within table represents species detected by both sampling methods.

Table 4.8 Frequency of isolation of bacterial species from Iceberg lettuce samplesa by plating onto TSA

Analytical procedure Species Maceration Rinse + Total Maceration Rinse + Tween 80 Tween 80 Ac. baumannii 1 0 0 0 1 Ac. haemolyticus 0 0 0 1 1 Ac. johnsonii 2 2 1 0 5 Ac. rhizosphaerae 1 0 0 0 1 Ag. larrymoorei 2 2 2 2 8 Arthrobacter spp. 0 1 0 1 2 B. humi 1 0 0 0 1 Burk. cepacia 4 2 2 1 9 Butt. noackiae/agrestis 0 1 0 0 1 Curt. flaccumfaciens 4 3 2 1 10 Curt. herbarum 0 1 0 1 2 Chrys. balustinum 0 1 0 0 1 Chrys. scophthalan 0 0 0 1 1 D. acidovorans 0 1 0 0 1 Erw. rhapontici 1 0 0 0 1 Fl. mizutaii 0 0 1 0 1 Klebsiella spp. 0 0 0 1 1 Mic. keratanolyticum 0 0 0 1 1 Mic. testaceum 0 0 1 2 3 P. agglomerans 3 0 0 0 3 P. ananatis 0 0 1 1 2 P. stewartii 1 1 1 1 4 Pl. okeanokoites 0 0 0 1 1 Ps. cichorii 1 1 1 1 4 Ps. flavescens 1 0 0 0 1 Ps. fragi 0 0 0 1 1 Ps. migulae 0 1 0 0 1 Ps. pavonaceae 1 1 1 1 4 Ps. putida 0 1 2 1 4 Ps. straminea 1 1 1 1 4 Ps. synxantha/fluorescens 1 7 5 4 17 Ps. syringae 1 0 0 1 2 Ps. viridiflava 1 0 1 1 3 Ser. marcescens 2 0 3 2 7 Ser. proteamaculans 0 2 0 0 2 St. maltophilia 1 3 2 2 8 Z. ramigera 1 1 1 1 4 Total 32 34 29 32 127 a For each analytical procedure 13 samples of lettuce were analysed

100

4.3.5 Comparison of bacterial species isolated from macerates and rinses of lettuce samples

There were 26 paired samples where bacterial species were isolated from lettuce samples by macerating and rinsing (Table 4.4 and 4.5). For 13 of these samples (Table

4.5), Tween 80 was included in the rinse and macerates solution and was not included for the other 13 samples (Table 4.4). The effect of Tween 80 on the isolation and occurrence of individual species will be discussed in a subsequent section. The first, second and third most predominant species for each sample were recorded on the basis of population (cfu/g) and listed in Tables 4.4, 4.5, 4.6 and 4.7.

There were only 10 out of the 26 samples where the same first dominant species was recovered by both maceration and rinsing methods (samples 2, 3, 9, 11 & 13, Table 4.4; samples 1, 2, 3, 7 & 9, Table 4.5). Only three of the 26 samples gave the same first and second dominant species (sample 9, Table 4.4; samples 1 & 3, Table 4.5) by both methods. The most frequently isolated bacterial species by maceration were Burk. cepacia, Ps. synxantha/fluorescens and Curt. flaccumfaciens (Tables 4.8 and 4.9).

Rinsing of the lettuce samples led to an increase in the isolation of Ser. marcescens, along with detection of Microbacterium spp. and a similar frequency of isolation for Ps. synxantha/fluorescens. However, less isolations of Burk. cepacia and Curt. flaccumfaciens were obtained by rinsing.

101

4.3.6 Comparison of bacterial species isolated from macerates and rinses of lettuce samples in the presence of Tween 80

There were 11 out of 26 samples where the same most predominant species of bacteria was observed for macerates and rinses, with and without the addition of Tween 80 to the solution (samples 1, 2, 3, 8 & 10, Table 4.6; samples 2, 3, 4, 7, 12 & 13, Table 4.7).

Only five out of the 26 samples gave the same first and second predominant species

(samples 1 & 10, Table 4.6; samples 3, 4 & 12, Table 4.7) by both methods. The addition of Tween 80 to the macerate solution led to an increased recovery of Ps. synxantha/fluorescens and St. maltophilia (Tables 4.8 and 4.9), however upon rinsing with Tween 80 in the solution a decrease in the isolation of Burk. cepacia and P. agglomerans was observed (Table 4.8).

102

Table 4.9 Frequency of isolation of predominant bacterial species from Iceberg lettuce samplesa by plating onto TSA

Population Times Isolated Species Isolated Range Maceration Rinse + Maceration Rinse (cfu/g) + Tween 80 Tween 80 1.0 x 103 – Ac. johnsonii 2 2 1 0 1.1 x 105 5.0 x 104 – Ag. larrymoorei 2 2 2 2 2.0 x 106 2.0 x 102 – Burk. cepacia 4 2 2 1 4.1 x 105 1.0 x 103 – Curt. flaccumfaciens 4 3 2 1 2.8 x 104 3.5 x 104 – P. stewartii 1 1 1 1 6.5 x 105 Ps. fluorescens/ 1.0 x 102 – 1 7 5 4 synxantha 1.5 x 105 6.5 x 105 – Ps. pavonaceae 1 1 1 1 2.3 x 106 9.5 x 102 – Ps. putida 0 1 2 1 8.8 x 104 4.0 x 104 – Ps. straminea 1 1 1 1 2.3 x 105 1.2 x 104 – Ser. marcescens 2 0 3 2 3.4 x 105 1.5 x 103- St. maltophilia 1 3 2 2 9.0 x 104 4.0 x 104 – Z. ramigera 1 1 1 1 9.5 x 104

103

4.3.7 Comparison of agar culture and PCR-DGGE methods for profiling the bacterial ecology of lettuce

Maceration and rinses of the same lettuce samples were analysed for the presence of bacterial species by agar plate culture and by PCR-DGGE as described in Materials and

Methods (Section 4.2). These species are listed and compared for each sample in Table

4.10 (isolates from lettuce macerates) and Table 4.11 (isolates from lettuce rinses).

These Tables also show the influence of including Tween 80 in macerate and rinse suspensions on the species recovered from the lettuce samples.

4.3.8 Species isolated and detected in lettuce samples

There were only four out of 13 samples where species isolated by plate culture were also detected by PCR-DGGE (Table 4.10). These were samples 2, 11, 12 and 13. In sample 2, Ps. cichorii was detected by both methods but PCR-DGGE did not detect Ps. syringae, Ps. fragi and Z. ramigera, despite populations of 104-105 cfu/g, that were recovered from the sample by plate culture. Acinetobacter johnsonii and Ser. marcescens were detected by both methods in sample 11. Acinetobacter baumannii and

Ac. johnsonii were detected by both methods in sample 12, and Ser. marcescens and Ac. johnsonii were detected by both methods in sample 13. Species of Curtobacterium,

Agrobacterium and Burkholderia, frequently isolated by plating on TSA, were not detected by PCR-DGGE, even though they were present sometimes at 104-105 cfu/g

(Table 4.10).

104

Table 4.10 Comparison of culture and PCR-DGGE methods for the isolation of predominant bacterial species from samples of Iceberg lettuce prepared by maceration and maceration with Tween 80

Sample Maceration Maceration + Tween Culture a PCR-DGGE Culture a PCR-DGGE 1 Ag. larrymoorei (5.0 x 104) Ps. stutzeri Ag. larrymoorei (7.5 x 104) Ps. stutzeri Ps. pavonaceae (2.3 x 106) Ps. pavonaceae (1.0 x 106) 2 Ps. cichorii (6.5 x 10 5) Ps. cichorii Ps. cichorii (1.1 x 105) Ps. cichorii Ps. syringae (7.5 x 104) Ps. Ps. fragi (4.0 x 104) Ps. Z. ramigera (4.0 x 104) brassicacearum Z. ramigera (9.5 x 104) brassicacearum 3 Ps. straminea (9.0 x 104) Pseudomonas sp. Ps. straminea (2.3 x 105) Pseudomonas sp. Ag. larrymoorei (1.0 x 105) Enterobacter sp. Ag. larrymoorei (2.0 x 106) Enterobacter sp. P. stewartii (9.0 x 104) P. stewartii (6.5 x 105) 4 Curt. flaccumfaciens ND Curt. flaccumfaciens ND (1.8 x 103) (2.2 x 103) Ac. rhizosphaerae (3.5 x 104) Ps. migulae (2.9 x 103) B. humi (1.0 x 102) 5 Burk. cepacia (1.9 x 103) ND Curt. herbarum (3.5 x 102) ND P. agglomerans (2.0 x 102) Ps. synxantha/ fluorescens (8.0 x 102) Ser. proteamaculans (2.0 x 102) 6 P. agglomerans (1.0 x 102) ND Ps. synxantha/ fluorescens ND Ps. putida (1.0 x 102) (7.4 x 103) Ser. proteamaculans (8.0 x 102) 7 ND ND Arthrobacter sp. (1.0 x 102) ND Ps. synxantha/ fluorescens (1.0 x 102) 8 Burk. cepacia (1.7 x 103) ND Burk. cepacia (3.0 x 103) ND Curt. flaccumfaciens Curt. flaccumfaciens (2.8 x 104) (5.9 x 103) Erw. rhapontici (3.2 x 103) Ps. fluorescens (4.3 x 103) 9 Burk. cepacia (1.7 x 103) ND Pseudomonas sp. (1.4 x 103) ND Curt. flaccumfaciens Ps. synxantha/ fluorescens (1.0 x 103) (2.8 x 103) Ps. synxantha/ fluorescens Ps. putida (9.5 x 102) (1.2 x 103) 10 Burk. cepacia (5.4 x 103) ND Burk. cepacia (8.0 x 103) ND Curt. flaccumfaciens Curt. flaccumfaciens (1.0 x 104) (1.7 x 104) Ps. viridiflava (2.9 x 103) Ps. synxantha/ fluorescens (7.5 x 103) 11 Ac. johnsonii (1.0 x 103) Ac. johnsonii Butt. noackiae/agrestis Ac. baumannii Ser. marcescens (3.6 x 104) Ser. marcescens (9.5 x 103) Ac. johnsonii P. ananatis/ agglomerans Ac. baumannii Chrys. balustinum (8.5 x 103) Bacillus sp. (9.0 x 103) St. maltophilia (1,8 x 104) Ser. marcescens 12 Ac. baumannii (2.2 x 104) Ac. baumannii Ac. johnsonii (4.0 x 104) Ac. johnsonii Ac. johnsonii (2.0 x 104) Ac. johnsonii St. maltophilia (6.5 x 104) Ac. baumannii Ps. flavescens (7.5 x 104) Ser. marcescens D. acidovorans (4.0 x 104) Ser. marcescens 13 Ser. marcescens (3.1 x 105) Ser. marcescens Ac. johnsonii (1.1 x 105) Ac. johnsonii St. maltophilia (9.0 x 104) Ac. baumannii St. maltophilia (7.5 x 104) Ac. baumannii Ac. johnsonii Ps. synxantha/ fluorescens Ser. marcescens Bacillus sp. (7.0 x 104) a Population (cfu/g) of species is shown in brackets after the identity ND- No bacterial species detected by PCR-DGGE Bold type within table represents species detected by both culture and PCR-DGGE. 105

Table 4.11 Comparison of culture and PCR-DGGE methods for the isolation of predominant bacterial species from samples of Iceberg lettuce prepared by rinsing and rinsing with Tween 80

Sample Rinse Rinse + Tween Culture a PCR-DGGE Culture a PCR-DGGE 1 Ag. larrymoorei (1.0 x 106) Ps. stutzeri Ag. larrymoorei (4.6 x 105) Ps. stutzeri Ps. pavonaceae (7.5 x 105) Mic. testaceum (4.0 x 104) St. maltophilia (2.0 x 104) Ps. pavonaceae (6.6 x 105) 2 Ps. cichorii (4.5 x 105) Ps. cichorii Ps. cichorii (6.0 x 105) Ps. cichorii Ps. putida (6.0 x 104) Ps. Ps. syringae (6.0 x 104) Ps. brassicacearum Z. ramigera (4.0 x 104) brassicacearum Z. ramigera (4.0 x 104) 3 Ps. straminea (4.0 x 104) Pseudomonas sp. Ps. straminea (2.2 x 105) Pseudomonas sp. Ag. larrymoorei (9.8 x 105) Enterobacter sp. Ag. larrymoorei (4.0 x 105) Enterobacter sp. P. stewartii (2.7 x 105) P. stewartii (3.0 x 105) 4 Curt. flaccumfaciens ND Chrys. scophthalan ND (4.2 x 103) (5.0 x 102) Mic. testaceum (1.0 x 104) Curt. flaccumfaciens (5.0 x 103) Mic. testaceum (2.5 x 104) 5 ND ND Curt. herbarum (1.0 x 102) ND 6 ND ND Arthrobacter sp. ND (4.0 x 102) Mic. keratanolyticum (3.5 x 102) Pl. okeanokoites(2.0 x 102) 7 Pseudomonas sp. ND Burk. cepacia (2.0 x 102) ND (1.0 x 102) Ps. synxantha/fluorescens Ps. synxantha/ fluorescens (1.1 x 103) (1.5 x 102) 8 Curt. flaccumfaciens ND Ps. synxantha/ fluorescens Pseudomonas sp. (9.5 x 103) (1.1 x 103) Ps. synxantha/ fluorescens Ps. putida (8.8 x 104) (9.5 x 103) Ps. viridiflava (1.9 x 104) 9 Ps. synxantha/ fluorescens Pseudomonas sp. Ps. synxantha/ fluorescens Pseudomonas sp. (8.0 x 104) (2.2 x 104) Burk. cepacia (4.1 x 105) St. maltophilia (1.0 x 104) 10 Burk. cepacia (6.5 x 103) ND Ps. synxantha/ fluorescens Pseudomonas sp. St. maltophilia (1.5 x 103) (1.2 x 104) Ps. viridiflava (1.0 x 104) 11 Ser. marcescens (1.2 x 104) Ser. marcescens Klebsiella sp. (3.9 x 104) Ac. baumannii Fl. mizutaii (4.0 x 103) Ac. baumannii St. maltophilia (2.5 x 104) Ac. johnsonii Ps. fluorescens (9.0 x 103) Ac. johnsonii Bacillus sp. Bacillus sp. Ser. marcescens 12 Ac. johnsonii (5.0 x 103) Ac. johnsonii Ser. marcescens Ser. marcescens Ser. marcescens (3.2 x 104) Ser. marcescens (3.4 x 105) Ps. synxantha/ fluorescens Ac. baumannii Ac. haemolyticus(3.5 x 104) Ac. baumannii (5.5 x 103) Pseudomonas sp. Ac. johnsonii (4.0 x 104) 13 Ser. marcescens (1.2 x 105) Ser. marcescens P. ananatis (4.5 x 103) Ac. baumannii P. ananatis (2.0 x 104) Ac. baumannii Ser. marcescens (1.4 x 104) Ac. johnsonii Ps. putida (4.0 x 104) Ac. johnsonii Bacillus sp. a Population (cfu/g) of species is shown in brackets after the identity ND- No bacterial species detected by PCR-DGGE Bold type within table represents species detected by both culture and PCR-DGGE.

106

Seven out of the 13 samples (samples 4, 5, 6, 7, 8, 9 & 10) did not show the presence of any bacterial species by PCR-DGGE of macerates with and without the addition of

Tween 80, despite the distinct presence of bacterial species by agar plate culture.

Although these samples generally had lower total populations then the other samples, there were cases where Curt. flaccumfaciens was present at approximately 103-104 cfu/g and not detected by PCR-DGGE (e.g. samples 4, 8, 9 & 10). PCR-DGGE of lettuce rinses generally gave the same data as obtained with macerates (Table 4.11). However, there were some samples where an unidentified Pseudomonas species was detected by

PCR-DGGE of lettuce rinses, but was not detected by PCR-DGGE of lettuce macerations (samples 8, 9 & 10, Table 4.11).

Figure 4.2 shows the PCR-DGGE profiles obtained from the total DNA extracted from several lettuce samples. The Figure also shows the PCR-DGGE profiles for DNA extracted from pure cultures of bacterial species that were isolated by plate culture.

PCR-DGGE of DNA extracted from macerates and rinses of sample 1 gave two, dominant bands of bacterial DNA, shown as band 1 and band 2 in Figure 4.2. On excision and sequencing, band 1 corresponded to the species Ps. stutzeri. The sequence of band 2 corresponded to the DNA of plant chloroplasts. The main cultural isolates from macerations of sample 1 were Ps. pavonaceae (lane 5) and Ag. larrymoorei (lane

6) which on extraction and PCR-DGGE gave DNA bands with mobilities that were, respectively, very similar to those found for Ps. stutzeri and plant chloroplast, as obtained by PCR-DGGE of macerate biomass (lanes 1 & 2, Figure 4.2).

107

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 Sample 1 Sample 2 MT M RT R Pp Al Mt Sm MT M RT R Pc Psy Zr

2 3

1 2 3 4 5 6 7 1 2 3 4 5 6 7 8 9 10 Sample 3 Sample 4

MT M RT R Al Pas Ps MT M RT R Pm Cf Ar Bh Mt Cs

2 3

Figure 4.2 Electrophoresis gels from PCR-DGGE analysis of (i) total biomass obtained from macerates and rinses of lettuce samples and (ii) pure cultures of bacterial species isolated from the same lettuce samples by agar culture M- analysis of biomass in macerates; R- analysis of biomass in rinses; MT- analysis of biomass in macerates with Tween 80; RT- analysis of biomass in rinses with Tween 80; Al- pure culture of Agrobacterium larrymoorei; Ar- pure culture of Acinetobacter rhizosphaerae; Bh- pure culture of Bacillus humi; Cf- pure culture of Curtobacterium flaccumfaciens; Cs- pure culture of Chryseobacterium scophthalan; Mt- pure culture of Microbacterium testaceum; Pas- pure culture of Pantoea stewartii; Pc- pure culture of Pseudomonas cichorii; Pm- pure culture of Pseudomonas migulae; Pp- pure culture of Pseudomonas pavonaceae; Ps- pure culture of Pseudomonas straminea; Psy- pure culture of Pseudomonas syringae; Sm- pure culture of Stenotrophomonas maltophilia; Zr- pure culture of Zoogloea ramigera

108

PCR-DGGE of DNA extracted from macerates and rinses of sample 2 gave one strong band which sequenced as chloroplast DNA (band 3) and two weak bands (bands 1 & 2,

Figure 4.2) that sequenced as Ps. cichorii and Ps. brassicacearum. Culture analysis of this lettuce sample gave isolates of Ps. syringae and Z. ramigera as well as Ps. cichorii.

PCR-DGGE bands from pure cultures of these species are shown in the Figure (lanes 5,

6 & 7), but there are no corresponding bands for PCR-DGGE analysis of biomass extracted from the sample for Zoogloea.

PCR-DGGE analysis of biomass obtained from macerates and rinses of sample 3 gave one strong band which sequenced as chloroplast DNA (band 2, Figure 4.2), three weak bands that sequenced as Enterobacter spp. (band 3, Figure 4.2) and one weak band that sequenced as Pseudomonas species (band 1, Figure 4.2). The closest species for the unidentified Pseudomonas was Ps. putida which had a similar mobility as the pure culture isolate, Ps. straminea (lane 5, Figure 4.2). Other cultural isolates from macerations and rinses of sample 3 were P. stewartii and Ag. larrymoorei and the DNA bands of the PCR-DGGE analysis are shown by lanes 6 and 7 (Figure 4.2). Band 3, which sequenced as Enterobacter species from the PCR-DGGE of macerates and rinses had a similar mobility to P. agglomerans, identified from pure culture isolation. These two species of bacteria are closely related as the genus Pantoea was formerly in the genus Enterobacter (Gavini et al. 1989).The DNA band for the pure culture of Ag. larrymoorei showed similar mobility to the DNA band for the plant chloroplast DNA and therefore may have been masked (lane 5, Figure 4.2).

PCR-DGGE analysis of biomass obtained from macerates and rinses of sample 4 gave only one distinct DNA band that corresponded to chloroplast DNA. Curtobacterium flaccumfaciens and Ac. rhizosphaerae were the main cultural isolates from macerates

109

and rinses of this lettuce sample. PCR-DGGE bands from pure cultures of these species are shown in Figure 4.2, lanes 6 and 7, however there are no corresponding bands of the same mobility for PCR-DGGE analysis of extracts of total biomass from the sample.

PCR-DGGE of DNA extracted from macerates and rinses of samples 11, 12 and 13 gave DNA bands for Ac. baumannii (band 1, Figure 4.3a), Ac. johnsonii (band 2, Figure

4.3a) and two separate bands for Ser. marcescens (bands 4 & 6, Figure 4.3a). Bacillus species was detected in one of the samples but did not predominate across all samples

(band 3, Figure 4.3). Predominant bacterial species isolated by plate culture were

Serratia marcescens (lane 4, Figure 4.3b), Ac. johnsonii (lane 5, Figure 4.3b) and Ac. baumannii (lane 6, Figure 4.3b) which had similar mobilities as the bands identified by

PCR-DGGE.

110

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 a) Maceration Maceration Rinse Rinse b)b & Tween & Tween Lettuce Sample 11 12 13 11 12 13 11 12 13 11 12 13 +ve -ve -ve Sm Aj Ab

1 2 3 4 5 6

Figure 4.3. Electrophoresis gels from PCR-DGGE analysis of: (a) total biomass obtained from macerates and rinses of lettuce samples with and without the addition of Tween 80, and (b) pure cultures isolated from lettuce samples +ve- positive control, un-inoculated lettuce sample -ve- negative control Sm- pure culture of Serratia marcescens Aj- pure culture of Acinetobacter johnsonii Ab- pure culture of Acinetobacter baumannii

4.3.9 Comparison of maceration and rinsing methods, and addition of Tween 80

on detection of bacterial species in lettuce samples by PCR-DGGE

As mentioned in previous discussions, there is a greater possibility that plant macerates,

compared to rinses, will contain substances likely to inhibit the performance of the PCR

assay. Also, inclusion of Tween 80 in the macerate and rinse diluents could facilitate

release and suspension of bacterial cells from lettuce samples, especially for the rinsing

method. Table 4.12 shows the frequency of detection of bacterial species by the PCR-

DGGE assay of macerates and rinses of lettuce samples. 111

Based on DNA band mobilities, a total of nine bacterial species were detected by PCR-

DGGE, across the 13 samples of lettuce examined. These same species were detected in lettuce macerates or rinses, with or without the use of Tween 80 (Table 4.12). The most frequently detected species were Ac. johnsonii, Ac. baumannii and Ser. marcescens and they were isolated in similar frequencies from macerates, rinses, with or without the addition of Tween 80.

Other species in Table 4.12 were also detected at similar frequencies from macerates or rinses (with/without Tween 80), with the exception of Pseudomonas spp. where rinsing and rinsing with Tween 80 gave increased frequency of detection. However, many more samples would need to be examined in order to conclude a definitive influence of these variables on the recovery of individual species, and this could be a direction for further research.

As mentioned previously, plant chloroplast DNA was a prominent band in DGGE gels

(Figure 4.2). It was considered that the presence of plant DNA would be less in lettuce rinses than in lettuce macerates and, therefore, bands of chloroplast DNA would be substantially less in intensity or even absent in the PCR-DGGE assay of lettuce rinses.

This was not found to be the case as chloroplast bands at similar intensities were found in PCR-DGGE assays of both lettuce macerates and lettuce rinses (Figures 4.2 & 4.3).

112

Table 4.12 Frequency of detection of bacterial species in samplesa of Iceberg lettuce by PCR-DGGE method

Sample Preparation Species Maceration Rinse + Total Maceration Rinse + Tween Tween Ac. baumannii 2 3 3 3 11 Ac. johnsonii 3 3 3 3 12 Bacillus spp. 1 1 1 2 5 Enterobacter spp. 1 1 1 1 4 Ps. brassicacearum 1 1 1 1 4 Ps. cichorii 1 1 1 1 4 Ps. stutzeri 1 1 1 1 4 Pseudomonas spp. 1 1 2 4 8 Ser. marcescens 3 3 3 2 11 Total 14 15 16 18 63 a For each analytical procedure, 13 samples were analysed

113

4.3.10 Overall comparison of culture and PCR-DGGE assays for detection of bacterial species in lettuce samples

Table 4.13 compares the frequency of detection of bacterial species in samples of lettuce by plate culture and PCR-DGGE methods. The most frequently detected species by agar culture were Ps. synxantha/fluorescens, Curt. flaccumfaciens, Burk. cepacia,

Ag. larrymoorei and St. maltophilia. None of these species were detected by PCR-

DGGE of total DNA in macerates or rinses. The most frequently detected species by

PCR-DGGE were Ac. johnsonii, Ac. baumannii, Ser. marcescens, Ps. stutzeri and an unidentifiable species of Pseudomonas. The Acinetobacter species were also detected by plate culture, but only on seven occasions compared with 23 occasions by PCR-

DGGE. Pseudomonas stutzeri was not detected by plate culture. Serratia marcescens and Ps. cichorii were the only two species that gave similar or near similar detection frequencies by both culture and PCR-DGGE methods (Table 4.12). Bacillus spp. and

Enterobacter spp. were detected on several occasions by PCR-DGGE but not by plate culture.

114

Table 4.13 Frequency of detection of bacterial species in samples of Iceberg lettuce by plate culture and PCR-DGGE methods

Species Culture PCR-DGGE Ac. baumannii 1 11 Ac. haemolyticus 1 0 Ac. johnsonii 5 12 Ac. rhizosphaerae 1 0 Ag. larrymoorei 8 0 Arthrobacter spp. 2 0 Bacillus spp. 0 5 B. humi 1 0 Burk. cepacia 9 0 Butt. noackiae/agrestis 1 0 Curt. flaccumfaciens 10 0 Curt. herbarum 2 0 Chrys. balustinum 1 0 Chrys. scophthalan 1 0 D. acidovorans 1 0 Enterobacter spp. 0 4 Erw. rhapontici 1 0 Fl. mizutaii 1 0 Klebsiella spp. 1 0 Mic. keratanolyticum 1 0 Mic. testaceum 3 0 P. agglomerans 3 0 P. ananatis 2 0 P. stewartii 4 0 Pl. okeanokoites 1 0 Pseudomonas spp. 0 8 Ps. brassicacearum 0 4 Ps. cichorii 4 4 Ps. flavescens 1 0 Ps. fragi 1 0 Ps. migulae 1 0 Ps. pavonaceae 4 0 Ps. putida 4 0 Ps. straminea 4 0 Ps. stutzeri 0 4 Ps. synxantha/ fluorescens 17 0 Ps. syringae 2 0 Ps. viridiflava 3 0 Ser. marcescens 7 11 Ser. proteamaculans 2 0 St. maltophilia 8 0 Z. ramigera 4 0 Total 123 63 115

4.3.11 PCR-DGGE detection of bacterial species associated with lettuce samples

The poor correlation between bacterial ecological data obtained by cultural and PCR-

DGGE analysis of lettuce samples suggested the need for a more detailed investigation of the PCR-DGGE method.

4.3.11.1 Limit of detection of pure cultures by PCR-DGGE

PCR-DGGE analyses were conducted on pure cultures of several species that had been isolated from lettuce samples by agar culture. Fresh cultures of these species were grown, diluted to give a range of populations (cfu/ml) and then samples of the diluted culture were assayed for the detection of the organism by PCR-DGGE. Using the standard one-step PCR-DGGE assay (Materials and Methods Section 4.2.10), DNA bands on gels were clearly visible for cultures diluted down to 103-104 cfu/ml for species of Ps. fluorescens and P. stewartii (Lane 6, Figure 4.4a; Lane 5, Figure 4.4c).

Thus, the lowest limit of detection would be a population of 104 cfu/ml. In an attempt to improve on this limit of detection, the diluted cultures were also examined by a nested

PCR procedure (Materials and Methods Section 4.2.10). The limit of detection was improved but double banding patterns were obtained at the lower cell populations and would confuse data interpretation (Lane 4, Figure 4.4b; Lane 2, Figure 4.4d; Lane 3,

Figure 4.4e; Lane 5, Figure 4.4f). Species of Ps. fluorescens, P. stewartii, Ps. pavonaceae and Curt. flaccumfaciens all had double banding patterns for populations of

103-106 cfu/ml. The limit of detection for the nested PCR therefore would be 105-106 cfu/ml based on the presence of single bands for the pure cultures.

116

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 a b 109108 107 106 105 104 103 102 10 1 109108 107 106105104103102 10 1

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 c d 108 107 106 105 104 103 102 10 1 108 107 106 105104 103 102 10 1

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 e f

108 107 106 105 104 103102 10 1 108 107 106 105 104 103 102 10 1

Figure 4.4 PCR-DGGE analysis of pure cultures at different populations; 1-109 cfu/ml, Pseudomonas fluorescens analysed by (a) one-step PCR- DGGE and (b) nested PCR-DGGE; 1-108 cfu/ml, Pantoea stewartii analysed by (c) one-step PCR-DGGE and (d) nested PCR-DGGE; (e) 1-108 cfu/ml, Pseudomonas pavonaceae analysed by nested PCR-DGGE and (f) 1-108 cfu/ml, Curtobacterium flaccumfaciens analysed by nested PCR- DGGE 117

Detection limits by PCR-DGGE were further investigated after known populations of individual species were inoculated into samples of lettuce (Materials and Methods

Section 4.2.9). The lettuce samples were then macerated or rinsed and examined by

PCR-DGGE (Figure 4.5). The Figure shows the DNA band obtained for pure cultures of each species (lane 1). Lane 2 shows the banding pattern for a negative control, which is for PCR-DGGE analysis of lettuce samples not inoculated with any species. Each negative control gave a distinct band towards the “bottom” of the gel. On excision and sequencing, this band corresponded to chloroplast DNA. Bands for pure cultures of Ps. fluorescens, P. stewartii, Ac. johnsonii and Burk. cepacia are clearly separated from the band of chloroplast DNA. However, DNA bands for pure cultures of Curt. flaccumfaciens and Ag. larrymoorei had similar mobilities to the band of chloroplast

DNA (Figures 4.5 (c) & 4.5 (d), lanes 1 & 6).

Populations of Ps. fluorescens and P. stewartii were not recovered by PCR-DGGE analysis of macerates or rinses of lettuce samples, even though they had been inoculated at levels up to 106 cfu/g (lanes 3, 4, 5, 8, 9 & 10, Figures 4.5 (a) & 4.5 (b)). It was not possible to draw any conclusions about the recovery of Curt. flaccumfaciens and Ag. larrymoorei inoculated into lettuce samples because their bands could not be distinguished from that of chloroplast DNA. Inoculated Ac. johnsonii was not detected in macerates of lettuce (lanes 4-7, Figure 4.5 (e)), despite populations of 103-106 cfu/g, but it was detected in rinses of lettuce at the 106 cfu/g inoculum level (lanes 12 & 13,

Figure 4.5 (e)). Such data demonstrate an inhibitory effect of substances in the macerate on performance of PCR.

118

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

a Ps. fluorescens (M) P. stewartii (M) b Ps. fluorescens (R) P. stewartii (R) + - 102 104 106 + - 102 104106 + - 102 104 106 + - 102 104106

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 c C. flaccumfaciens Ag. larrymoorei d C. flaccumfaciens Ag. larrymoorei + - 102 104 106 + - 102 104106 + - 102 104 106 + - 102 104106

1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13

e Ac. johnsonii (M) Ac. johnsonii (R) f Burk cepacia (M) Burk cepacia (R) + - - 103103 106 106 - - 103 103106106 + - - 103 103 106 106 - - 103 103106106

Figure 4.5 PCR-DGGE analysis of (a) & (b) Pseudomonas fluorescens and Pantoea stewartii, (c) & (d) Curtobacterium flaccumfaciens and Agrobacterium larrymoorei, (e) Acinetobacter johnsonii and (f) Burkholderia cepacia inoculated onto Iceberg lettuce and sampled using maceration (M) and rinse (R) methods + = positive control, pure culture of target microorganism; -= negative control, lettuce sample with no inoculate pure cultures; 103 cfu/g; 106 cfu/g 119

Burkholderia cepacia inoculated onto lettuce at 103-106 cfu/g was detected by PCR-

DGGE analysis of both macerates and rinses (lanes 4, 5, 6, 7, 10, 11, 12 & 13, Figure

4.5 (f)). As might be expected stronger band signals were obtained when the cells were inoculated at 106 cfu/g (lanes 6, 7, 12 & 13, Figure 4.5 (f)). For some negative control samples, positive bands were observed apart from the strong band identified as chloroplast DNA (lanes 3, 8 & 9, Figure 4.5 (f)). These particular bands correspond to the natural microflora of the lettuce.

Several conclusions can be drawn from these experiments: (i) the “universal primers” used to amplify DNA of the bacterial ribosomal DNA gene, also amplify the DNA of chloroplasts from plant cells. The mobility of this chloroplast DNA during DGGE is similar to the amplified ribosomal DNA of several bacterial species associated with lettuce; (ii) the PCR is inhibited or decreased by substances that are extracted from lettuce either by maceration or rinsing.

4.3.12 The use of additives in the PCR mixture to enhance amplification of bacterial DNA

In an attempt to increase the amplification of bacterial DNA and reduce the presence of chloroplast DNA in PCR-DGGE gels, betaine and dimethyl sulfoxide (DMSO) were supplemented into the PCR mastermix. Both of these agents have been found to enhance PCR by affecting the specificity of the PCR reaction and increasing the amplification of GC-rich DNA sequences (Filichkin and Gelvin 1992; Rees et al. 1993;

Baskaran et al. 1996; Henke et al. 1997; Chakrabarti and Schutt 2001b).

120

To assess the effect of the addition of betaine and DMSO to the master mix, serial dilutions (1:10, 1:100) and undiluted DNA of the total biomass from lettuce sample 1 were analysed by PCR-DGGE. The electrophoresis gel shows the results of the analysis with a control sample (lanes 1, 5 & 9, Figure 4.6) for comparison with the different additives, 1.4 M betaine (lanes 2, 6 & 10); 5% DMSO (lanes 3, 7 & 11) and 1.4 M betaine and 5% DMSO (lanes 4, 8 & 12). For all of the different concentrations of the template DNA (undiluted, 1:10, 1:100), no difference in the DNA banding pattern was observed, where all the DNA bands had similar mobilities for the additives and the control.

1 2 3 4 5 6 7 8 9 10 11 12 C B D BD C B D BD C B D BD

Figure 4.6 The effect of betaine, DMSO and template DNA dilution on the recovery of bacterial species from lettuce by PCR-DGGE analysis Lanes 1-4: undiluted template DNA Lanes 5-8: 1:10 dilution of template DNA Lanes 9-12: 1:100 dilution of template DNA C, control; PCR mastermix containing added: B, betaine (1.4 M); D, DMSO (5%); BD, betaine (1.4 M) + DMSO (5%)

121

4.4 Discussion

To understand and manage the impact of microorganisms on the quality, shelf-life and safety of lettuce and lettuce products, it is important to have reliable knowledge about the species of bacteria that are present on this produce and the quantitative populations at which they occur. Using a combination of culture and culture-independent molecular methods for bacterial analyses, this Chapter has provided data to meet this goal. The reliability of the data was extended by using both maceration and rinsing methods for sample preparation, and including Tween 80 in diluents to assist suspension and recovery of the bacterial species.

4.4.1 Populations of bacterial species associated with lettuce

To study the bacterial populations associated with lettuce, samples were analysed using maceration, maceration with Tween 80, rinse and rinse with Tween 80, followed by culturing onto TSA.

The total bacterial populations recovered from lettuce samples ranged from 0-107 cfu/g with significant differences observed between batches of lettuce examined. Most samples had populations in the range 104-105 cfu/g and this is consistent with results reported by previous researchers (Riser et al. 1984; Garg et al. 1990; Pingulkar et al.

2001). Numerous factors are reported to affect the populations of bacteria on leaves taken from lettuce heads and these include contact with soil, faeces, irrigation water, water used to reconstitute pesticides, dust, insects, inadequate composted manure, wild and domestic animals, human handling, harvesting and processing equipment (Lund

1992; Singaglia et al. 1999; Beuchat 2002). Leaves on the outer parts of the heads 122

generally have higher populations than those from the inner locations (Lund 1993;

Aycicek et al. 2006), as might be expected. In this Chapter, variations between inner and outer leaves were minimized by analysing mixtures of sub-samples taken from both locations, and by using 25 g as the basic analytical sample rather than the standard 10g sample (ES ISO 6887-1:2001; AS 5013-2004; AS 5013.11.1-2004).

Maceration of food samples is considered to give the highest recovery of microbial counts, as a greater proportion of the microbial cells associated with the product are brought into suspension (Jay and Margitic 1979; Fleet 1999; Nedoluha et al. 2001).

However, maceration also blends food constituents that could inhibit the recovery of some microorganisms (Pederson and Fisher 1944a, 1944b; Jay and Margitic 1979;

Beuchat and Brackett 1990; Nguyen-the and Lund 1992; Kyung and Fleming 1994; Han et al. 2004; Beuchat 2006; Brandi et al. 2006). In comparison, rinsing of samples may minimize the amount of anti-microbial food constituents in the final suspension, and facilitate recovery of species that might be inhibited. Nevertheless, there is always the possibility that rinsing might not remove species strongly adsorbed to the food surface, thereby giving an underestimation of the true ecology (Jay and Margitic 1979; Burnett and Beuchat 2001; Nedoluha et al. 2001). Based on the data of Table 4.2, it could not be concluded with statistical confidence that either maceration or rinsing gave higher population recovery, although slightly higher populations were obtained in more samples by rinsing. In this context, the issues of antimicrobial inhibitors or decreased release and suspension of bacterial cells are not significant for the lettuces examined.

Similar results have been observed for comparison of sample preparation methods for recovering Salmonella from fruits, vegetables and herbs (Burnett and Beuchat 2001)

123

and the isolation of aerobes and Enterobacteriaceae from poultry (Cox et al. 1976;

Lillard 1988b).

Surfactants can enhance the recovery of bacteria from food matrices, but selection of the type of surfactant is important, as some can be inhibitory to some species of bacteria.

Tween 80, used at low concentrations, is one surfactant that seems to be particularly effective (Adams et al. 1989; Hwang and Beuchat 1995; Yu et al. 2001 Lillard 1988;

Eginton et al. 1998; Raiden et al. 2003; do Socorro Rocha Bastos et al. 2005; El-

Baradei et al. 2007; Nielsen et al. 2007), so it was selected for evaluation in this study.

The data of Table 4.2, as described in Section 4.3.3 clearly show that incorporation of

Tween 80 into macerates or rinses gave higher recoveries of bacterial populations. The surfactant may facilitate dispersion and separation of bacterial cells from food particles or facilitate the dispersion of clumps or chains of bacterial cells (Eginton et al. 1998;

Ukuku and Fett 2002; do Socorro Rocha Bastos 2005). In the past, the addition of surfactants has led to greater recovery of microorganisms from various food matrices, such as poultry (Lillard 1988a; Hwang and Beuchat 1995), cantaloupe (Beuchat and

Scouten 2004; do Socorro Rocha Bastos 2005), and strawberries (Yu et al. 2001).

4.4.2 Diversity of bacterial species associated with lettuce

The predominant bacterial species isolated from lettuce were identified by sequencing of the 16S rDNA and comparison of the data to the BLAST database.

Independent of sampling method, the most frequently isolated species from all lettuce batches were Ps. synxantha/fluorescens, Curt. flaccumfaciens, St. maltophilia and Ag. larrymoorei (Table 4.8). The predominance of these species is consistent with other

124

studies conducted on raw and processed lettuce, where Ps. fluorescens was observed to be the most frequently isolated species (Magnuson et al. 1990; Hamilton-Miller and

Shah 2001). The same authors found other bacterial species at lower frequencies, and these included Curt. flaccumfaciens, Ser. marcescens, P. agglomerans, St. maltophilia and Acinetobacter. In this study of Iceberg lettuce, Gram negative bacteria were found to predominate the bacterial populations recovered from lettuce samples (Table 4.3), which is similar to the observations of Szabo and Coventry (2001).

The species which were consistently found at high populations were Ps. synxantha/fluorescens, Ag. larrymoorei and P. stewartii (Table 4.9). Pseudomonas fluorescens is regarded as a pectolytic fluorescent pseudomonad which is responsible for substantial proportion of soft-rot disorders in vegetables (Liao and Wells 1987; Liao

2006). The importance of pectolytic fluorescent pseudomonads as the leading cause of spoilage of refrigerated produce is primarily due to their psychrotrophic nature, nutritional diversity, and predominant presence on the surfaces of fresh produce

Serratia marcescens is commonly isolated from plants and has been found to be a nosocomial and iatrogenic pathogen of humans (Grimont et al. 1981). Spoilage of vegetable products by species of Serratia is usually due to temperatures abuse, which allows the growth of Serratia over psychrotrophic species of pseudomonads (Murray

2001).

125

Agrobacterium species are commonly associated with the roots of over 800 plant species, including crops (Canfield and Moore 1991; Pulawska and Sobiczewski 2005) and have been proposed to be incorporated into the genus Rhizobium (Young et al.

2001). Agrobacterium species are bacteria that are responsible for the formation of crown galls resulting in great economic losses in nursery production of fruit trees, roses and grapevines (Kennedy and Alcorn 1980). The presence of galls inhibits the physiological function of infected tissue, especially roots, and upon breakdown creates wounds on roots that can become entry points for other soil-borne pathogens (Pulawska and Sobiczewski 2005). On-farm harvesting and packaging operations are likely to bring lettuce roots and associated soil into contact with lettuce leaves. Therefore, the isolation of Agrobacterium species from lettuce leaves would not be unexpected, as found in this Chapter, although it has not been previously reported. The significance of this finding requires further research, as it is not known if such species would grow on lettuce surfaces, contribute to spoilage reactions, or affect the growth and survival of other lettuce associated bacteria.

The genus Pantoea consists of several species which were formerly classified in the genera Enterobacter and Erwinia (Gavini et al. 1989; Hauben et al. 1998). Pantoea species are frequently isolated from the surface of raw and processed vegetables

(Wright et al. 1976; Riser et al. 1984; Brocklehurst et al. 1987; Magnuson et al. 1990;

Hamilton-Miller and Shah 2001). These species, in particular P. agglomerans, are considered to be a major cause of spoilage of ready-to-eat vegetables upon storage and refrigeration (Bennik et al. 1998; Hamilton-Miller and Shah 2001). Pantoea agglomerans has also been used as a biological control agent against Curt. flaccumfaciens, which causes bacterial wilt of bean (Hsieh et al. 2005). Mixed

126

populations of P. agglomerans and Curt. flaccumfaciens were found to reduce the infection by Curt. flaccumfaciens on beans. Interestingly, for the samples of lettuce analysed in this Chapter, species of Pantoea and Curt. flaccumfaciens were never isolated from the same lettuce sample (e.g. samples 3, 4, 5, 6, 8, 9, 10 & 11; Tables 4.4,

4.5, 4.6 & 4.7), suggesting the occurrence of some antagonism by Pantoea against Curt. flaccumfaciens. There would be merit in further research into the effect of this relationship on the bacterial ecology of lettuce.

Curtobacterium flaccumfaciens (formerly a species of Corynebacterium) is a Gram positive bacterium and has been found infrequently by other researchers on the surface of lettuce (Collins and Jones 1983, Magnuson et al. 1990). Curtobacterium flaccumfaciens is an invasive phytopathogenic microorganism that produces pectolytic enzymes leading to the spoilage of beans (Hsieh et al. 2005). It has also been isolated from the surface of potatoes (Krechel et al. 2002) and pear blossoms (Stockwell et al.

2002). Curtobacterium flaccumfaciens has been found to be infrequently isolated from pear blossoms, however when it is isolated it occurs at relatively high populations

(Stockwell et al. 2002). This does not coincide with the results of this study where Curt. flaccumfaciens occurred at high populations and was isolated from several lettuce samples. This could be due to the growth of Curt. flaccumfaciens on the surface of the lettuce and presence as a epiphytic microorganism, leading to potential spoilage. Future research could focus on the effect of this microorganism on the bacterial ecology of lettuce.

Burkholderia cepacia, formerly known as Ps. cepacia (Palleroni and Holmes 1981;

Kersters et al. 1996), causes of spoilage of fresh and stored vegetable products due to the production of pectinases (Marchetti et al. 1992; Freire and Robbs 2000; Tekoriené

127

2003). Burkholderia cepacia has been reported in soil environments (McArthur et al.

1988; Palumbo et al. 2007), as a plant pathogen (González et al. 1997), and as a N2- fixing maize endophyte in Mexico (Estrada et al. 2002), as well as a sugarcane endophyte in South Africa, where in vitro tests revealed antifungal activity against

Ustilago and Fusarium (van Antwerpen et al. 2002). This species has also been shown to have biological activity against the spoilage fungus, Penicillium digitatum, on citrus fruit (Huang et al. 1993) as well as being acknowledged as a source of sepsis in clinical environments (Ricker et al. 1991).

The isolation of Acinetobacter species from the surface of lettuce led to further research into its prevalence on lettuce and ready-to-eat vegetable salads which is described in more detail in Chapter 6. Acinetobacter species have been found to be part of the microflora of other vegetables including lettuce and tomatoes (Khan et al. 1992;

Brackett 1998; Hamilton-Miller and Shah 2001). Generally, the presence of

Acinetobacter species in foods is an indicator of spoilage, although they occur at varying populations, depending upon the food type (Gennari et al. 1992). They are significant nosocomial pathogens and are associated with the increasing incidence of hospital acquired infections such as bacteremia and pneumoniae, endocardation and meningitis (Bergogne-Bérézin and Towner 1996). Their public health significance has become of particular concern in recent years because of their ability to quickly develop resistance to antibiotics such as penicillin, cephalosporins, quinolones and aminoglycosides (Van Looveren et al. 2004; Hanlon 2005). Outbreaks of Acinetobacter infection in hospitals, caused by multi-resistant strains of these bacteria are widely reported and have drawn attention to the environmental sources of these species.

Species of main interest are Ac. baumannii and Ac. calcoaceticus which have been

128

isolated from raw vegetables, fish and meat (Van Looveren et al. 2004; Abbo et al.

2005; Hanlon 2005).

Broadly, the same diversity of bacterial species was obtained from lettuces by either maceration or rinsing and by the use of Tween 80, but for some samples and species, some differential influences were observed (Tables 4.8 and 4.9). For example, Burk. cepacia, P. agglomerans, Curt. flaccumfaciens and Ac. johnsonii were more frequently detected from maceration samples, whereas, species of Ser. marcescens, Ps. putida and

Microbacterium testaceum were more frequently isolated from rinse samples (Table

4.8). These results suggest that there is a difference in the bacterial species recovered by maceration and rinse methods and this may be due to variations in attachment of microorganisms to the surface of the lettuce (Frank 2001). Addition of Tween 80 to the macerate and rinse solutions led to an increased recovery of Ps. synxantha/fluorescens,

St. maltophilia and detection of Arthrobacter species and Curt. herbarum (Tables 4.8 and 4.9), suggesting that the surfactant gave an enhanced removal of these bacteria from the lettuce samples.

The use of surfactants, such as Tween 80, and detergents can assist in the removal of total microbial populations from the surface of strawberries, cantaloupe and lettuce

(Adams et al. 1989; Beuchat and Scouten 2004; do Socorro Rocha Bastos 2005). The addition of Tween 80 to the solution used for washing has been shown to reduce the total microbial populations on the surface of lettuce by 99.6% (Adams et al. 1989).

However, the authors failed to identify the microbial species removed from the surface of the lettuce. For some species, Tween 80 could have an inhibitory effect. For example, there was a decreased frequency of isolation of Burk. cepacia and P. agglomerans from rinses which included Tween 80 (Table 4.8). Tween 80 has been observed to have a

129

sanitising effect, with a 5% solution having an increased effect on the removal of

Salmonella spp. from chicken skin (Cheng-An and Beuchat 1995) and 100-200ppm of

Tween 80 leading to a 1.1-1.2-log reduction of E. coli O157:H7 on the surface of strawberries (Yu et al. 2001). These data suggest that the concentration of Tween 80 should be kept low and well controlled.

Although the data do suggest some differential influences of maceration, rinsing and the use of Tween on the recovery of some species from lettuces, the trends are not definitive. Many more samples would need to be examined to overcome inherent variations that occur both within the one lettuce sample and between different lettuce samples. In addition, such research would benefit from a more targeted approach, by focussing on the behaviour of only one or two key species.

4.4.3 PCR-DGGE

PCR-DGGE was used in conjunction with cultural methods to study the microbial ecology of lettuce. DGGE is a culture-independent method for the detection of microorganisms that may not be obtained by cultivation on agar plates (Muyzer and

Smalla 1998; Muyzer 1999; Giraffa 2004) and, recently, it has been applied to the study of the microbial ecology of foods and beverages (Cocolin et al. 2001; Dewettinck et al.

2001; Ercolini et al. 2003b; Lopez et al. 2003; Miambi et al. 2003; Lee et al. 2005).

Compared to plate culture, PCR-DGGE gave very poor recovery or detection of bacterial species associated with lettuce samples. None of the 13 samples, analysed by either maceration or rinsing, gave comparable data when examined by either culture or

PCR-DGGE methods. Most samples gave no detectable bands of bacterial DNA by

130

PCR-DGGE analysis. However, there were four samples where at least one species was detected by both methods.

The detection limit of bacterial species in food and beverage ecosystems by PCR-

DGGE is reported to be about 103-104 cells/g or ml of product (Dewettinck et al. 2001;

Vanbroekhoven et al. 2004; Pulawska and Sobiczewski 2004; Theunissen et al. 2005).

This limitation might account for the failure of PCR-DGGE to detect bacteria in some samples, as their populations were around 103-104 cfu/g, but there were some samples where bacterial populations of 104-105 cfu/g were not detected.

Although a diversity of 37 species were detected by plate culture (Table 4.8), only nine species were detected by PCR-DGGE (Table 4.12). The most frequently detected species by plate culture were Ps. synxantha/fluorescens, Curt. flaccumfaciens, Burk. cepacia, Ag. larrymoorei and St. maltophilia. However, none of these were detected by

PCR-DGGE of total DNA in either macerates or rinses of lettuce samples. The most frequently detected species by PCR-DGGE were Ac. baumannii, Ac. johnsonii and Ser. marcescens. The Acinetobacter species were also detected by plate culture but only on seven occasions compared with 23 occasions by PCR-DGGE. Bacillus spp., Ps. stutzeri and Ps. brassicacearum were only detected by PCR-DGGE. For the two species of

Pseudomonas isolated by PCR-DGGE, Ps. stutzeri and Ps. brassicacearum, there were no closely related species of the same genus that were isolated by culture (Anzai et al.

2000; Yamamoto et al. 2000). This conclusion was based on a comparison of the 16S ribosomal DNA sequences for the isolates obtained by culture and for bands obtained from DGGE gels. The fact that PCR-DGGE analyses did detect bacteria not found by plate culture does suggest that lettuce may harbour some bacterial species in a viable but not culturable form. However, it is important to note that PCR-DGGE detects the DNA

131

of all cells, whether they are dead or alive (Giraffa 2004). Therefore, bacterial species detected by PCR-DGGE and not detected by culture may not always be due to the microorganism being viable but non-culturable, because they could represent dead cells.

In food microbiology, studies of the bacterial species recovered by of culture-dependent and PCR-DGGE methods have focussed on potatoes (Garbeva et al. 2001), wine grapes

(Bae et al. 2006), cheese (Cocolin et al. 2004) and fermentation of cassava dough

(Miambi et al. 2003), maize dough (Ampe et al. 1999), and cocoa (Nielsen et al. 2007).

The results of some of these studies show good correlation between culture on agar and

PCR-DGGE analysis during the evaluation of the late blowing of cheese and fermentation of cocoa (Cocolin et al. 2004: Nielsen et al. 2007). However, the results of some of these studies are similar to those observed in this Chapter, where differences in the profile of bacterial species recovered by culture on agar and PCR-DGGE were observed. Such conclusions highlight the need to use a combination of methods in order to obtain a more reliable and accurate description of the microbial ecology of foods

(Garbeva et al. 2001; Miambi et al. 2003; Bae et al. 2006). This Chapter was the first study to look at the complex microflora of lettuce by PCR-DGGE and has exposed some technical difficulties that need to be overcome before reliable data can be obtained.

A series of controlled experiments were conducted to determine what might be causing the failure of the PCR-DGGE analysis of bacteria in lettuce samples. Pure cultures of the various bacterial species isolated from lettuce could be detected by PCR-DGGE at populations as low as 103-104 cfu/ml, and at lower levels if a nested PCR assay was used. Nested PCR uses two sets of primers in sequential reactions, the second set of primers amplifying a region within the amplicon generated using the first set of primers,

132

which is useful in improving the specificity of a particular microorganism. The first

PCR generates an amplicon product at concentrations greater than the initial target. Due to an excess, the amplicon, rather than the initial target, is far more likely to function as the target in the second PCR, greatly reducing the likelihood of amplification of nonspecific products (Boon et al. 2002; Thompson et al. 2002; Cox and Fleet 2003;

Vanbroekhoven et al. 2004).

While the nested PCR detected lower populations (1-10 cfu/ml, Figures 4.4b, 4.4d, 4.4e,

4.4f), multiple DNA bands were obtained on the electrophoretic gels, thereby complicating the analyses. This result can be due to artificial or natural microheterogeneity in the DNA sequence (Speksnijder et al. 2001), where a single band may be composed of several species (Van Hannen et al. 1998; Sekiguchi et al. 2001), or that several bands are generated from a single species. The application of DGGE is also dependent upon unique coupling between Tm (band retardation) and the phylogenetic identity. Due to the sensitivity of the DGGE, this relationship must hold down to the level of single bases (i.e. at the microvariation level) (Kisand and Wikner 2003). The multiple bands for a single species may be due to the presence of a so-called “wobble base” (either a C or a T) in the reverse primer (Kowalchuk et al. 1997) or because of multiple copies of the 16S rDNA gene with sequence microheterogeneity, which can overestimate the community diversity detected by DGGE (Nübel et al. 1996). For example, multiple bands are displayed in DGGE profiles of the 16S rDNA V1 region by some species of lactobacilli and staphylococci of food origin (Cocolin et al. 2001b); moreover, multiple bands were also found for species of staphylococci by DGGE analysis of the V3 region (Cocolin et al. 2001a; Blaiotta et al. 2003). Further research is

133

required on the effect of dilution of the target DNA and the effect of different primers on the PCR-DGGE banding pattern in the case of the nested PCR assay.

Although the PCR-DGGE method could detect bacteria at 103-104 cfu/ml in pure culture, such populations and higher (up to 106 cfu/ml) were not detected when they were inoculated into lettuce samples and the standard PCR-DGGE method applied.

Plant chloroplast DNA emerged as a significant factor that interfered with the PCR-

DGGE assay. It seemed that the universal bacterial primers selected for PCR amplification of the 16S unit of the bacterial ribosomal gene also recognized sequences in plant DNA, giving strong amplification of plant chloroplast DNA. The mobility of this DNA band on DGGE corresponded with that of several bacterial species that are frequently associated with lettuce (e.g. Curt. flaccumfaciens and Ag. larrymoorei,

Figures 4.5c & 4.5d). The strength of the band of chloroplast DNA effectively masked the detection of these species. Nevertheless, there were some bacterial species that gave

DNA amplicons with mobilities that were well separated from chloroplast DNA. Even in these cases, detection of inoculated cells was significantly diminished, with Ps. fluorescens and P. stewartii not being detected at 106 cfu/g for both maceration and rinse samples. Inoculated Ac. johnsonii was not detected in macerates of lettuce, despite populations of 103-106 cfu/g, but it was detected in rinses of lettuce at 106 cfu/g inoculum level. In this case, it seems that macerates contains plant constituents that may be inhibitory to the PCR. The presence of large amounts of plant chloroplast DNA from the lettuce homogenates and rinses in the PCR mastermix could lead to a decreased efficacy of the PCR amplification by inhibiting binding to the bacterial ribosomal DNA by the PCR primers (Reysenbach et al. 1992; Suzuki and Giovannoni 1996; von

Wintzingerode et al. 1997; Wilson 1997; Sambrook and Russell 2001; Ercolini 2004).

134

The presence of plant chloroplast DNA may also lead to differential or preferential amplification of non-target DNA. This has been observed from mixed bacterial DNA samples where some of the original members of the community were not amplified during PCR, due to preferential amplification of other bacterial DNA present (Ercolini

2004).

Plant chloroplast DNA has been found to interfere with microbial signals during studies of the microbial ecology of whole seeds of sugar beets and citrus plants (Dent et al.

2004; Lacava et al. 2006). The use of several different conditions of PCR, namely, standard, hot start and touchdown, was not successful in reducing the interference of plant chloroplast DNA from extracts of sugar beet seeds (Dent et al. 2004). It has been suggested that bacterial PCR primers may also target and amplify regions of plant and yeast DNA that are likely to occur in total DNA extracts (Lopez et al. 2003). An evaluation of commonly used bacterial primers found that those which target the V6-V8 region of bacterial DNA are able to amplify plant DNA, and that primers which target the V3 region of bacterial DNA are also able to amplify yeast DNA (Lopez et al. 2003).

The present study used primers targeting the V3 region, and therefore were not expected to amplify the chloroplast DNA, as seems to have occurred. The use of bacterial V3 primers has also led to the detection of plant chloroplast DNA from sugar beet (Dent et al. 2004), cassava starch (Ampe et al. 2001), maize dough (Ampe et al. 1999), kimchi

(Lee et al. 2005) and barley (Normander and Prosser 2000; Yang and Crowley 2000), demonstrating that chloroplast is able to be amplified using V3 primers.

Further research is needed to understand what factors in lettuce homogenates and rinses interfere with PCR-DGGE assays for bacteria. One alternative is to try and decrease the amount of plant DNA in the total extract from lettuce, thereby allowing better

135

performance of the PCR assay. This will be considered in the next chapter (Chapter 5).

Other possibilities include better selection of primers and trying to improve the efficacy of the DNA amplification by PCR.

The conditions of PCR assay involve many variables (e.g. reagent mixture and concentrations, buffer pH, time, temperature). As mentioned already, the selection and specificity of primer couples are fundamental to successful PCR reactions. Targeting of different 16S variable regions may lead to different results in species composition of the same sample (Ercolini et al. 2003). The process of trialling different primer couples is very time consuming and this strategy may not be successful when large amounts of plant DNA are present relative to bacterial DNA. Consequently, this direction was not further investigated in this study. However, future research could focus on targeting different regions of bacterial DNA for amplification, with the goal of detecting all the species likely to be present in lettuces, but not amplifying plant DNA.

This study investigated the effect of adding betaine, DMSO and formamide to the PCR reagent mixture with the goal of improving PCR efficacy, but the outcomes were not successful. These results do not coincide with past research where the addition of these reagents has led to increased specificity and amplification of DNA bands (Sarkar et al.1990; Smith et al.1990; Pomp and Medrano 1991; Filichkin and Gelvin 1992; Rees et al.1993; Varadaraj and Skinner 1994; Baskaran et al.1996; Henke et al.1997;

Chakrabarti and Schutt 2001a; Chakrabarti and Schutt 2001b). The reason for this discrepancy may be due to the GC-content of the target DNA being too low. The addition of betaine, DMSO and formamide has been acknowledged as enhancing the amplification of GC-rich DNA (Sarkar et al. 1990; Smith et al. 1990; Varadaraj and

Skinner 1994; Henke et al. 1997). However, if the GC-contents of the target bacterial

136

and plant chloroplast DNA were similar, differential amplification would not be achieved.

In conclusion, the bacterial populations recovered from Iceberg lettuce samples ranged from 0-107 cfu/g using maceration, maceration with Tween 80, rinse and rinse with

Tween 80. No difference was observed between the recoveries of total populations between maceration and rinsing, however upon the addition of Tween 80 higher bacterial populations were usually observed. The most frequently isolated bacterial species by culturing onto TSA were Ps. synxantha/fluorescens, Curt. flaccumfaciens and Ag. larrymoorei. PCR-DGGE analysis of the same lettuce samples was unable to detect the same incidence of bacterial species as culturing onto TSA. However, PCR-

DGGE was able to detect Bacillus species, Pseudomonas species, Ser. marcescens and enhance the detection of Acinetobacter species from lettuce samples. Failure of the

PCR-DGGE analyses to reliably detect bacterial species in lettuce extracts is probably related to interference of plant DNA and possibly other plant constituents on the PCR assay, and further research is needed to investigate these factors.

137

CHAPTER 5

ELIMINATION OF CHLOROPLAST DNA FROM THE

PCR-DGGE ASSAY OF BACTERIA IN LETTUCE

HOMOGENATES

5.1 Introduction

From the previous chapter (Chapter 4), PCR-DGGE analysis of the bacterial flora associated with lettuce samples showed that plant chloroplast DNA interfered with the detection of bacterial species. Plant chloroplast DNA had the same mobility in DGGE gels as the DNA bands of epiphytic microorganisms, such as Curtobacterium flaccumfaciens and Agrobacterium larrymoorei, thereby masking the presence of these organisms. Moreover, it is possible that, chloroplast DNA is also recognized and amplified by the bacterial primers used for PCR, thereby diminishing their availability for the amplification of bacterial DNA. In previous studies of the microbial ecology of foods by molecular methods, chloroplast DNA was detected during the fermentation of maize dough and sour cassava starch, and from sugar beet (Ampe et al. 1999; 2001;

Dent et al. 2004). Dent et al. (2004) found that the DNA of sugar beet (Beta vulgaris subsp. Vulgaris) had a significant effect on the detection of bands of bacterial DNA by

DGGE. It was suggested that an extra step to target the depletion of plant rDNA would be needed to increase the limit of detection of bacterial species in plant materials.

Methods that may be used to reduce or eliminate interference from chloroplast DNA during PCR-DGGE can be divided into the following categories: i) application before

138

extraction of DNA; ii) application after DNA extraction and iii) and application during

PCR.

Before extraction of DNA from the food matrix, bacteria and chloroplasts occur as intact whole cells or entities and various methods have been proposed for the physical separation of bacteria from the plant material. These methods include physical, chemical, physico-chemical and biological methods of extraction (Stevens and Jaykus

2004). Examples of these methods include differential or density gradient centrifugation, adsorption onto metal hydroxides, and filtration (Lucore et al. 2000;

Stevens and Jaykus 2004). Metal hydroxide solutions such as zirconium (IV) hydroxide, hafnium hydroxide or titanium (III) hydroxide, have been used to separate bacterial cells from other matrices (Lucore et al. 2000; Cullison and Jaykus 2002; Stevens and

Jaykus 2004). All of these methods have advantages and disadvantages and require evaluation for specific purposes.

After the extraction of DNA by chemical or mechanical methods, chloroplast and bacterial DNA can be separated on the basis of size. Plant DNA (chloroplast) is much larger than bacterial DNA. Preparative agarose gel electrophoresis is an example of a method able to separate genomic DNA on the basis of size and could be applied before use of the DNA in PCR. Low percentage gels (0.7-1.0 %) prepared with low melt agarose are used to facilitate easy recovery of the DNA after electrophoresis.

Modification of the PCR assay is another strategy that can be used to reduce or eliminate interference from chloroplast DNA in PCR-DGGE. The addition of reagents such as betaine, glycerol, formamide or dimethyl sulfoxide (DMSO) to the PCR master mix can give a less biased reaction and lead to greater amplification of bacterial DNA

(Smith et al. 1990; Pomp and Medrano 1991; Baskaran et al. 1996; Chakrabarti and

139

Schutt 2001a, 2001b). Such initiatives were examined in the previous chapter (Chapter

4), but were unsuccessful.

This chapter investigates various methods for eliminating or reducing the interference from chloroplast DNA during PCR-DGGE analysis of bacteria in lettuce samples.

5.2 Materials and Methods

5.2.1 Preparation of lettuce homogenates

Whole Iceberg lettuces were purchased from local supermarkets, stored at 4°C and used within 24 h. To produce homogenates of lettuce, samples (25 g) were taken and placed into a filter-lined stomacher bag (Seward, Norfolk, UK) to which 225 ml of 0.1%

Bacteriological Peptone water (Oxoid, Adelaide, Australia) were added. The sample was then macerated in a Stomacher (Seward) for one minute and the resulting homogenate used in experiments.

5.2.2 Effect of sample preparation on the presence of chloroplasts in lettuce suspensions

Sample preparation methods, such as maceration and rinsing, can lead to differences in the release of antimicrobial compounds into the final suspension (Fleet 1999). The effect of sample preparation methods (maceration and rinse) on the release of chloroplast material into the final suspension was compared using spinach leaves as the starting material. Whole baby spinach leaves were chosen for investigation in this comparison because, unlike lettuce, they could be purchased in a “whole”, intact state

140

and examined without cutting. Whole, undamaged baby spinach leaves were purchased from local supermarkets and replicate sub-samples (15 g) were aseptically taken. These sub-samples were either cut with a knife into pieces (2 cm by 2 cm) or examined whole.

Bacteriological Peptone water (1:10, 0.1%) was added and the samples were either macerated for one minute in a Stomacher, or rinsed by placing the sample into a 500 ml conical flask and shaken on an orbital shaker (150 rpm, Model P03422, Paton

Industries) for 30 min at room temperature. An aliquot (10 ml) of the suspension was transferred to a Petri dish and assessed for the presence of the colour green using a

Minolta Colorimeter (Minolta Co. Ltd, Japan). Chloroplast cells are green in colour

(Campbell 1996); therefore, more intense green colour in the suspension indicates that more chloroplast cells are released during sample preparation.

5.2.3 Removal of chloroplast cells by differential centrifugation

Differential centrifugation is, mainly, centrifugation at a relatively low speed to sediment larger sized particles, followed by centrifugation of the supernatant at a higher speed to sediment smaller sized particles (Stevens and Jaykus 2004). Differential centrifugation was examined as a method for the removal of larger chloroplast cells from smaller bacterial cells. Lettuce homogenates were placed into a sterile 250 ml centrifuge tube and centrifuged at 1,000 g for 10 mins at 4°C (Avanti J-E, Beckman

Coulter Refrigerated Centrifuge, Beckman-Coulter Inc., California USA) to remove chloroplasts. The supernatant was then transferred to a clean sterile centrifuge tube (250 ml) and centrifuged at 10,000 g for 15 mins at 4°C to sediment bacterial cells. The supernatant was discarded and the cell pellet transferred to an Eppendorf tube after

141

suspension in distilled water (1-2 ml). The resulting cell pellet was used to assess the reduction in the chloroplast cells by comparison to a single step centrifugation process

(10,000g for 15 mins at 4°C). The cell suspensions were visually compared and assessed for colour and the presence of chloroplast cells (green colour). The DNA was also mechanically extracted from these cell suspensions and analysed by PCR-DGGE as described in Section 5.2.10.

5.2.4 Use of titanium hydroxide to separate bacterial and chloroplast cells

A suspension of titanium (III) hydroxide was prepared by the addition of 356 µl of titanium (III) chloride (Sigma-Aldrich, Australia) to 200 ml of water to produce a 1.3 mM suspension. The pH of this suspension was adjusted to 7.0 ± 0.2 by addition of 5 M ammonium hydroxide, using continuous agitation. The sediment was then washed three times to remove excess ammonium ions. During this process, 200 ml of sterile saline

(0.9% NaCl) was added to the suspension and gently mixed. The suspension was then left for 10 mins at room temperature allowing the sediment to settle to the bottom of the flask. The top 40% of the supernatant was then decanted and the sediment washed another two times. The final volume of the titanium (III) hydroxide suspension was 200-

300 ml. The suspension could be stored in the dark at room temperature for up to six months (Lucore et al. 2000).

Lettuce homogenates were placed into a sterile centrifuge tube (250 ml) and spun at

10000g for 15 mins at 4°C. The pellet was transferred to a sterile tube and resuspended in 3 ml of sterile saline and 6 ml of titanium (III) hydroxide suspension. The tube was agitated for 10 mins at room temperature on a horizontal shaking vortex mixer. After

142

shaking, the tube was centrifuged for 10 mins at 500g at room temperature to sediment the titanium hydroxide particles with adsorbed bacteria, and the supernatant carefully removed. The remaining pellet was transferred to a sterile Eppendorf tube and resuspended in 1 ml of sterile saline. The absorbed bacterial cells were then extracted for their DNA and analysed using PCR-DGGE, as described in Section 5.2.10.

5.2.5 Preparative agarose gel electrophoresis

Preparative agarose gel electrophoresis is similar to normal agarose electrophoresis but with a substitution of low melt agarose. Low melt agarose is similar to normal agarose, however the agarose melts at a much lower temperature, thereby allowing easier recovery of the DNA after electrophoresis (Weil and Hampel 1973; Cho and Kim

2000). Low melt agarose was obtained from Sigma Aldrich and used to prepare gels of concentrations ranging from 0.7-1.0% agarose in 1X Tris Acetate Buffer (TAE). The agarose solution was melted in a microwave for 1-2 mins and allowed to cool before pouring into a gel caster. After one hour at room temperature, the gel was transferred to the electrophoresis unit containing 1X TAE with sufficient volume to cover the gel.

Samples (10-20 µl) of crude DNA extracted by both mechanical and chemical methods were mixed with loading dye and loaded into individual wells. The gel was run at 60V until the dye had reached the far end of the gel (~6 h) and then stained with ethidium bromide (0.5 µg/ml) for 10-15 mins. After staining, the gel was visualised on a UV illuminator and the bands of interest excised.

143

5.2.6 Recovery of DNA from low melt agarose gels

After excision of the relevant bands with a sterile scalpel, the gel fragments were placed into separate sterile Eppendorf tubes. Five volumes of elution buffer (20 mM Tris-HCl,

1 mM EDTA) were added to the gel slice and the tube placed in a waterbath at 65°C for

5 mins to melt the agarose (Sambrook et al. 1989). The tube and contents were cooled to room temperature and an equivalent amount of equilibrated phenol (pH 8.0; Sigma-

Aldrich) was added. The tube was mixed on a vortex for 20 s and the aqueous phase recovered by centrifugation at 4000g for 10 mins at 20°C. The agarose precipitates at the interface of the two solutions as a white powder and, therefore, can be removed. The aqueous phase (top layer) was taken to a new Eppendorf tube and extracted once with phenol:chloroform (1:1, Sigma-Aldrich) and once with chloroform (Sigma-Aldrich), with transfer to a new Eppendorf tube after each extraction. Ammonium acetate (0.2 volumes, 10 M) and 2 volumes of absolute ethanol at 4°C were added to the final aqueous phase for precipitation of the DNA. The mixture was stored at room temperature for 10 mins and DNA was recovered by centrifugation at 13000g for 10 mins at 4°C. The supernatant was discarded, and the DNA pellet washed with 70% ethanol and dissolved in TE buffer (pH 8.0). The DNA was amplified by PCR and analysed by DGGE, as described in Section 5.2.10.

5.2.7 Reference strains of bacteria used for the assessment of preparative gel electrophoresis

Reference strains of bacteria used in this study are listed in Table 5.1 and were cultured and maintained on Nutrient Agar (NA, Oxoid). Bacteria were grown in 5 ml of Nutrient

144

Broth (NB, Oxoid) at 30°C for 24 h. Culture (1.0 ml) was transferred to a 1.5 ml cryogenic tube and centrifuged at 16,000g for 10 min at 4°C to sediment the bacterial cells. Cell pellets were stored at -20°C until extraction of DNA by chemical lysis, and used for evaluation of preparative gel electrophoresis.

Table 5.1 Bacterial cultures used for the evaluation of preparative gel electrophoresis

Bacterial species Source Acinetobacter baumannii School of Chemical Sciences and Engineering, UNSWb Curtobacterium flaccumfaciens School of Chemical Sciences and Engineering, UNSW Pseudomonas fluorescens School of Biotechnology and Biomolecular (ATCCa 13525) Sciences, UNSW Stenotrophomonas maltophilia School of Chemical Sciences and Engineering, UNSW a American Type Culture Collection b University of New South Wales, Sydney Australia

5.2.8 DNA Extraction Methods

Two different methods (physical and chemical) were used to extract the DNA from bacterial cell pellets.

5.2.8.1 Physical extraction of DNA

The DNA from the cell pellets was physically extracted according to the method described earlier (Section 3.2, Chapter 3).

5.2.8.2 Chemical extraction of DNA

For extraction of the DNA by chemical lysis, the microbial pellet was washed with TE

Buffer (10 mM Tris, 1 mM EDTA, pH 8.0) and centrifuged at 15,000g for 10 mins at

145

4°C to recover the pellet. The supernatant was discarded and 500 µl of breaking buffer

(10 mM Tris, 1 mM EDTA, 1% SDS, 0.2 mg/ml Proteinase K) added (Sambrook et al.

1989). The tube was then incubated in a waterbath (55°C) overnight for cellular digestion. Solutions of 5 M NaCl, CTAB/NaCl (4.1 g NaCl + 10.0 g CTAB in 100 ml sterile water) and the digested pellet were pre-warmed in a waterbath (65°C) for 5 mins and 90 µl of 5M NaCl and CTAB/NaCl solutions added to the digested pellet. The tube was mixed slowly by inverting for 5 mins and then incubated for 10 mins at 65°C to allow CTAB to bind protein and form a precipitate. An equal amount (700 µl) of chloroform:isoamyl alcohol (Sigma-Aldrich) was added and mixed slowly for 15 mins.

The tube was then centrifuged at 13,000 rpm for 10 mins at 4°C and the upper layer transferred to a fresh Eppendorf tube. This step was repeated depending upon the clarity of the upper layer. An equivalent amount of isopropanol (600 µl) was added to the tube and mixed slowly and the tube was placed in a freezer overnight to aid in precipitation of DNA. The tube was spun at 13,000 rpm for 10 mins at 4°C and the supernatant discarded and the pellet washed with 1 ml of 70% ethanol. After centrifuging at 13,000 rpm for 10 mins at room temperature, the supernatant was discarded and the pellet allowed to dry at room temperature for 15 mins. The DNA was resuspended in 100 µl of

TE buffer, allowed to dissolve (5 mins), and stored in the freezer (-20°C) until needed.

146

5.2.9 PCR amplification of DNA for PCR-DGGE analysis

To investigate the reduction of chloroplast from PCR-DGGE fingerprints, the primer set

341F-GC and 518R was used to amplify the extracted DNA according to the method described earlier (Section 3.2, Chapter 3).

The presence of PCR products was confirmed by agarose gel electrophoresis of 2-10 µl of sample loaded into a 1.5% agarose gel and run at 120V for 40 min. The gel was then stained in ethidium bromide (1.0 µg/ml) for 10-15 min and visualised on a UV illuminator (300 nm).

5.2.10 Denaturing gradient gel electrophoresis (DGGE)

The PCR amplicons were separated by DGGE using a D-Code Universal Mutation

Detection System (Bio-rad, Hercules, CA, USA) according to the method described earlier (Section 3.2, Chapter 3).

147

5.3 Results

5.3.1 Effect of sample preparation on the release of chloroplasts from spinach leaves

Samples of baby spinach were either cut or left whole, and macerated or rinsed to assess the release of chloroplasts. Leaves which had been cut and macerated gave more intense green colour due to the release of chloroplast cells (Table 5.2). Chloroplast cells are green in colour and more intense green colour corresponds to a greater release of chloroplasts during sample preparation. The uncut rinsed samples gave the least amount of chloroplast release, followed by the cut and rinsed samples. The data in Table 5.2 show values taken using a Minolta Colorimeter. The L-values correspond to the lightness of the sample, the a-values correspond to the red to green colour shift, and the b-values correspond to the yellow to blue colour shift. Therefore, the lower or more negative the a-value for a sample, the greener the colour of the solution. Comparison of the different sample techniques shows that maceration and cutting of whole baby spinach leaves released the most chloroplast cells and gave the lowest a-values. Rinsing of whole baby leaves of spinach had the greatest effect on limiting the release of chloroplast cells. Nevertheless, even rinsing of whole leaves released obvious chloroplast material.

148

Table 5.2 Effect of maceration, rinse and cutting of whole spinach leaves on release of chloroplasts as measured by a Minolta colorimeter

Sampling method L value a value b value Control 60.24 +0.46 -2.22 Maceration and cut 59.63 -1.01 +1.19 Maceration 59.06 0.00 -1.07 Rinse + Cut 58.76 -0.59 +0.50 Rinse 59.03 +0.09 -1.02

L-value = lightness; a value= red (+)-green (-); b value= yellow (+)-blue (-)

5.3.2 Removal of chloroplasts by differential centrifugation

Differential centrifugation was used to separate the bacterial and chloroplast cells from lettuce homogenates on the basis of their mass and size. The efficacy of the method to separate the bacterial and chloroplast cells was measured by the intensity of the green colour. Resuspended differential centrifugation sediments of the fraction expected to contain mostly bacterial cells had a green colour intensity similar to that of the control where only a single centrifugation step was completed. This suggests that the differential centrifugation of the lettuce homogenates only removes the larger particles of lettuce while leaving the bacterial and chloroplast cells still in the supernatant. This was confirmed by PCR-DGGE analysis of the DNA extracted by mechanical lysis from the sediment expected to contain mostly bacterial DNA, as it gave intense bands of chloroplast DNA (Figure 5.1, lanes 5 & 6, band 2).

149

1 2 3 4 5 6 7 8

Figure 5.1. Comparison of PCR-DGGE analysis of lettuce homogenates separated by titanium (III) hydroxide and differential centrifugation and the effect of DNA extracted by mechanical and chemical lysis Lanes 1 & 2: DNA extracted from lettuce homogenates by mechanical lysis Lanes 3 & 4: DNA extracted from the sediment expected to contain mainly bacterial cells, which were separated by the use of titanium (III) hydroxide and extracted by mechanical lysis Lanes 5 & 6: DNA extracted from the sediment expected to contain mainly bacterial cells, which was separated by differential centrifugation and extracted by mechanical lysis Lanes 7 & 8: DNA extracted from lettuce homogenates by chemical lysis

150

5.3.3 Removal of chloroplasts by titanium (III) hydroxide

Titanium (III) hydroxide was used to adsorb and sediment bacterial cells from lettuce homogenates. Efficacy of the method to separate bacterial cells from chloroplast cells was measured by the intensity of green colour. Resuspended titanium hydroxide sediments had a green colour intensity just slightly less than that of the control assay where no titanium hydroxide was added to the homogenate. This suggests that chloroplast cells were also entrapped within the sedimented metal hydroxide complex.

This was confirmed by PCR-DGGE analysis of DNA extracted from the sediment.

DNA extracted by the mechanical lysis procedure gave intense bands of chloroplast

DNA (Figure 5.1, lanes 3 & 4, band 2).

5.3.4 Removal of chloroplast by preparative gel electrophoresis

Preparative agarose gel electrophoresis was used to separate the plant and bacterial

DNA after total DNA extraction from lettuce macerates. The efficacy of preparative agarose electrophoresis to separate bacterial and chloroplast cells was evaluated by the separation of DNA fragments by PCR-DGGE of the fragment corresponding to bacterial DNA. Lysis of the lettuce homogenates using the mechanical procedure led to a range of DNA fragments being formed (Figure 5.2, lanes 8 & 9). This is due to the shearing of the DNA during the physical breaking open of the cells using the Bead

Beater. The extraction of the DNA from lettuce homogenates by chemical lysis and processing by preparative gel electrophoresis gave separate bands for the DNA fragments (Figure 5.2, lanes 2 & 3, band 2). The other band (Figure 5.2, lanes 2 & 3, band 1), corresponds to fragments larger than 10,000 base pairs which can consist of 151

plant DNA. The other bands, visible for all samples in the preparative agarose gel are

26S RNA (Figure 5.2, band 3), 16S RNA (Figure 5.2, band 4) and 5S RNA (Figure 5.2, band 5) (Griffiths et al. 2000).

The ability of preparative gel electrophoresis to distinguish bacterial DNA was evaluated using pure cultures of Pseudomonas fluorescens, Curtobacterium flaccumfaciens, Acinetobacter baumannii and Stenotrophomonas maltophilia, which were extracted using chemical lysis (Figure 5.2, lanes 4, 5, 6 & 7). Preparative agarose electrophoresis of these DNA extracts led to the separation and formation of a distinct band corresponding to bacterial DNA (Figure 5.2, band 2). Other bands corresponding to 26S RNA, 16S RNA and 5S RNA were also visible (Figure 5.2, bands 3, 4 & 5). The top visible band (Figure 5.2; band 1) corresponds to the rest of the bacterial cell contents such as ribosomes.

152

1 2 3 4 5 6 7 8 9 10 1

3 4

Figure 5.2 Preparative agarose gel electrophoresis of DNA extracted from lettuce homogenates and DNA extracted from pure cultures. The DNA was extracted by either mechanical or chemical methods Lane 1: Hind III marker Lanes 2 & 3: Lettuce homogenates, DNA extracted using chemical lysis Lane 4: Pseudomonas fluorescens, DNA extracted using chemical lysis Lane 5: Acinetobacter baumannii, DNA extracted using chemical lysis Lane 6: Stenotrophomonas maltophilia, DNA extracted using chemical lysis Lane 7: Curtobacterium flaccumfaciens, DNA extracted using chemical lysis Lanes 8 & 9: Lettuce homogenates, DNA extracted using mechanical lysis Lane 10: Size marker

After preparative gel electrophoresis, the bacterial DNA was extracted from the band

(Figure 5.2, band 2) and analysed by PCR-DGGE. These DNA amplicons were compared to the control, untreated lettuce homogenates to which preparative gel electrophoresis had not been carried out (Figure 5.3). PCR-DGGE of bacterial DNA obtained after preparative gel electrophoresis did not show the presence of bands of chloroplast DNA and allowed the amplification and separation of a more distinct bacterial band (Figure 5.3, lanes 5 & 6, band 1). A band of similar mobility was barely visible for the control (Figure 5.3, lanes 1 & 2, band 1) with a more distinct band present for the chloroplast (Figure 5.3, lanes 1 & 2, band 2). Upon excising and 153

sequencing, band 1 was identified as Serratia marcescens. These results suggest that the chloroplast DNA is separated from the bacterial DNA during preparative gel electrophoresis.

1 2 3 4 5 6

1 2

Figure 5.3 PCR-DGGE analysis of DNA extracts from lettuce homogenates Lanes 1 & 2: untreated lettuce homogenate, no preparative agarose gel electrophoresis; note band 2, chloroplast DNA Lanes 3 & 4: negative controls, PCR master mix with no template DNA Lanes 5 & 6: bacterial DNA obtained from lettuce homogenates by preparative agarose gel electrophoresis, and then subjected to PCR-DGGE

154

5.4 Discussion

Plant chloroplast DNA was observed to have an effect on the detection of bacterial species in samples of Iceberg lettuce by PCR-DGGE analysis (Chapter 4). The objective of this chapter was to evaluate the reduction of plant chloroplast DNA in lettuce homogenates by differential centrifugation, the use of titanium (III) hydroxide and preparative gel electrophoresis. The efficacy of these methods was evaluated by the presence of a green colour, indicating the presence of chloroplast, and confirmed by

PCR-DGGE analysis of the resulting solutions.

The first researchers to acknowledge the presence of chloroplasts and their effect on

PCR-DGGE analysis of bacteria in plant products were Dent et al. (2004). They noted that the release of plant DNA from embryo tissue created too much noise (background interference) for PCR-DGGE to be useful. This interference affected the detection of low populations of bacterial species on sugar beet seeds and the researchers suggested that future efforts should be targeted on removal of plant DNA by specific oligonucleotide-labelled magnetic beads. These observations were reported after completion of the current project, but confirm the difficulties experienced with plant

DNA in this project.

A simple strategy to decrease the presence of plant DNA in extracts for bacterial analysis would be to minimize disruption of the plant matrix. Rinsing of plant surfaces in contrast to maceration of the sample should achieve this objective (Fleet 1999). As measured by the release of chlorophyll (green colour), this goal was achieved, but the decrease was insufficient because enough plant DNA was also released to significantly interfere with the PCR-DGGE assay (Table 5.2 & Figure 5.1).

155

Differential centrifugation is based on differences in the sedimentation rate of particles of differing sizes and densities. In the past, differential centrifugation has been used to separate meat particles, fat and water layers from samples prior to PCR analysis of bacteria (Stevens and Jaykus 2004). Since chloroplast cells are much larger than bacterial cells (Campbell 1996), it was considered that differential centrifugation of plant macerates or rinses would be one mechanism that might decrease the amount of chloroplast DNA relative to bacterial DNA. In other research, differential centrifugation has been used to physically enrich bacterial cells in relation to other matrices (Thomas

1988; Meyer et al. 1991; Neiderhauser et al. 1992; Stevens and Jaykus 2004).

Unfortunately, it was not effective in removing enough of the chloroplast cells from the bacterial cells, because fractions expected to be enriched in bacterial cells exhibited interference from chloroplast DNA on PCR-DGGE analysis (Figure 5.1). Questions could also be raised about the efficacy of the method as bacterial cells may be lost during transfer of the supernatant between centrifuge tubes and may also be removed during the first centrifugation spin at 1000 g for 10 mins. Further research could consider optimisation of centrifugation speed and g forces to improve separation of the different types of cells, but this was not investigated in this project.

Titanous and zirconium hydroxides have been used to immobilise bacterial cells from different sample matrices including non-fat dry milk, enrichment broths and dairy products (Lucore et al. 2000; McKillip et al. 2000; Cullison and Jaykus 2002). Titanium

(III) hydroxide suspension, prepared according to the procedure of Lucore et al. (2000), also adsorbed chloroplast cells from lettuce extracts. The suspension complex was green, and DNA extracted and analysed by PCR-DGGE also gave a prominent band of chloroplast DNA (Figure 5.1). This may be due to the binding of chloroplast cells along

156

with bacterial cells during reaction with the titanium (III) hydroxide. Even though a low centrifugal force was used to sediment the titanium hydroxide particles, it is possible some chloroplast cells also sedimented under these conditions to contaminate the bacterial cells on their release from the titanous hydroxide. Consequently, the initiative to separate chloroplast and bacterial cells by adsorption to metal hydroxides, in order to improve bacterial analysis in lettuce extracts by PCR-DGGE was not effective.

Processing of DNA extracts from lettuce macerates by preparative gel electrophoresis proved to be an effective strategy for separating bacterial DNA from chloroplast DNA.

Bacterial DNA prepared from these gels could be analysed by PCR-DGGE without the production of bands of chloroplast DNA and allowed the identification of Serratia marcescens. The detection of only one band after the elimination of chloroplast may be due to the low bacterial populations (Figure 5.3; band 1) on the surface of the lettuce sample examined. Alternatively, one band can correspond to several species of bacteria

(Van Hannen et al. 1998; Sekiguchi et al. 2001; Speksnijder et al. 2001), however this was not the case as no problems were encountered upon identification of the excised band by DNA sequencing.

The use of preparative gel electrophoresis to clean up bacterial DNA from complex matrices, such as environmental samples, prior to molecular analysis by PCR methods has been reported previously (Weil and Hampel 1973; Kedersha and Rome 1986;

Liesack and Stackebrandt 1992; Kowalchuk et al. 1997; Griffiths et al. 2000). However, this is the first time it has been used to process DNA from plant foods as a prelude to subsequent PCR analysis. Preparative gel electrophoresis was best performed on DNA that was extracted from bacteria in the lettuce macerates by a chemical lysis method.

Physical disruption of the cells by mechanical shaking lead to unacceptable shearing of

157

the DNA, both from bacterial and plant cells. This decreased the efficacy of electrophoretic fractionation, and also gave sheared fragments of DNA that can lead to artefacts of the PCR assay such as multiple bands and chimeric PCR products (Liesack et al. 1991). Similar results for the shearing of DNA were observed by Leff et al. (1995) when using a similar mechanical lysis extraction method.

From a practical perspective, application of preparative gel electrophoresis adds a further layer of analytical complexity that adds to the cost and time of microbiological examination. The use of preparative agarose electrophoresis was not taken further in this Chapter due to time limitations for the overall thesis. However further evaluation of its efficacy as a preparative step before PCR-DGGE analyses of microorganisms in plant produce is warranted.

From this Chapter, it can be concluded that preparative agarose gel electrophoresis was the only method capable of separating plant chloroplast DNA from bacterial cell DNA to a degree that would enable meaningful PCR-DGGE analyses. The use of differential centrifugation and titanium (III) hydroxide was ineffective in separating bacterial cells and chloroplast cells for the purposes of PCR-DGGE analyses. More research is required to apply preparative agarose gel electrophoresis as a prelude to more detailed study of the bacterial ecology of lettuce and other plant produce by PCR-DGGE.

158

CHAPTER 6

OCCURRENCE AND DETECTION OF ACINETOBACTER

SPECIES IN LETTUCE AND VEGETABLE SALADS;

EVALUATION OF MEDIA

6.1 Introduction

Acinetobacter species are a group of bacteria in the family Neisseriaceae. They are

Gram negative, non-motile, non-sporulating, catalase positive, oxidase negative cocco- bacilli with an aerobic metabolism. Current genomic taxonomy recognises 19 species within the genus, with Acinetobacter calcoaceticus, Ac. baumannii, Ac. haemolyticus,

Ac. junii, Ac. johnsonii, Ac. lwoffii and Ac. radioresistens being the most well-known species (Bouvet and Grimont 1986; Nemec et al. 2001; Carr et al. 2003; Hanlon 2005).

Acinetobacter species are ubiquitous in the environment. They are readily isolated from soil, water, hospital environments, and naturally occur on the skin of healthy individuals in about 20-25% of the population (Berlau et al. 1999b).

Acinetobacter species have been isolated from various foods including fresh meats, poultry, soft drinks, cheese, natural mineral water, milk and nuts. They have been linked to the spoilage of bacon, fresh meat, poultry, fish and eggs, even when these products are stored at refrigeration temperature (Gennari et al. 1992; Gennari and Lombardi

1993; Kämpfer 2000; Addis et al. 2001; Houang et al. 2001; Cantoni and Iacumin 2003,

Betts 2006). Generally, the presence of Acinetobacter in foods is an indicator of

159

spoilage, although they occur at varying populations, depending upon the food type

(Gennari et al. 1992).

Acinetobacter species have not been associated with outbreaks of foodborne disease but they do have a record of public health concern. They are significant nosocomial pathogens and are associated with the increasing incidence of hospital acquired infections such as bacteremia and pneumoniae, endocardation and meningitis

(Bergogne-Bérézin and Towner 1996).

Their public health significance has become of particular concern in recent years because of their ability to quickly develop resistance to antibiotics such as penicillin, cephalosporins, quinolones and aminoglycosides (Van Looveren et al. 2004; Hanlon

2005). Outbreaks of Acinetobacter infection in hospitals, caused by multi-resistant strains of these bacteria are widely reported and have drawn attention to the environmental sources of these species. Species of main interest are Ac. baumannii and

Ac. calcoaceticus which have been isolated from raw vegetables, fish and meat (Van

Looveren et al. 2004; Abbo et al. 2005; Hanlon 2005).

Foods, including salad vegetables, have been implicated as possible sources of

Acinetobacter in the hospital environment (Berlau et al. 1999a; Houang et al. 2001).

Salad vegetables are of particular significance because they are not further processed and consumed as fresh products.

Surveys of fresh salad vegetables have found that Acinetobacter species occur at low populations (50-1000 cfu/g) in 17-50% of samples, with Ac. baumannii, Ac. calcoaceticus and Ac. johnsonii being the most frequently isolated (Berlau et al. 1999a;

Houang et al. 2001). Lettuces, which are the main constituent of ready-to-eat salads, have a low inherent population of Acinetobacter species, leading researchers to propose

160

that uncooked vegetables are a potential source of nosocomial infections to vulnerable patients, and that multiple antibiotic resistance is common in epiphytic bacteria as found on lettuce (Jiwa et al. 1981; Khan et al. 1992; Hamilton-Miller and Shah 2001).

Because of the increasing concern about Acinetobacter species as antibiotic-resistant pathogens, more information is required about their incidence and population in foods, especially high risk ready-to-eat salads. This study reports the prevalence of

Acinetobacter species in iceberg lettuce and ready-to-eat salads obtained from retail outlets in Sydney, Australia, and examine the efficacy of several plating and enrichment media for detecting these species in such habitats.

6.2 Materials and Methods

6.2.1 Salad vegetable samples

Iceberg lettuce and prepacked ready-to-eat vegetable salads were randomly purchased from supermarkets and retailers in Sydney, Australia. They were stored at 4°C and analysed within 24 h of purchase. All samples were collected and examined within their

‘best before date’ or ‘display date’.

6.2.2 Media for isolation of Acinetobacter species

Media for the selective enrichment and agar plate culture of Acinetobacter species were prepared from basic ingredients, as they were not commercially available. Two selective agar plating media and two selective enrichment broths were used to isolate

Acinetobacter species. These were modified Leeds Acinetobacter agar (MLAA),

161

minimal salts agar (MSA), Baumann enrichment medium (BEM) and minimal salts solution (MSS) (Table 6.1).

MLAA was modified from the original Leeds Acinetobacter medium (LAM), first described by Jawad et al. (1994) for the selective isolation of Acinetobacter species from hospital environments. The modification consisted of using a reduced concentration of vancomycin (3 mg/L) in the formulation (Houang et al. 2001).

Antibiotics were purchased from Sigma Aldrich (Sydney, Australia), reconstituted in water, and filter sterilised before addition to the sterilised medium. Peptone, casein hydrolysate and agar were purchased from Oxoid (Adelaide, Australia).

Table 6.1 Composition of media used for the isolation of Acinetobacter species from salad vegetables

Medium Ingredients Modified Leeds Acinetobacter 10 g bacteriological agar, 15 g acid casein hydrolysate, agar (MLAA) 5 g bacteriological peptone, 5 g NaCl, 5 g D-fructose, 5 g sucrose, 5 g D-mannitol, 1 g L-phenylalanine, 0.4 g ferric ammonium citrate, 0.02 g phenol red, 3 mg vancomycin, 15 mg cefsulodin, 50 mg cephradine (pH 7.0 ± 0.1) (Jawad et al. 1994; Houang et al. 2001) Minimal salts agar (MSA) 250 ml minimal salts solution (MSS), 20 g bacteriological agar, 10 ml of 20% glucose solution, 1 g sodium acetate, 750 ml distilled water (pH 7.0 ± 0.1) (Berlau et al. 1999a) Baumann enrichment medium 2 g sodium acetate, 2 g KNO3, 0.2 g MgSO4.7H2O, 1 L (BEM) 0.04 M phosphate buffer (pH 6.0) (Baumann 1968) Minimal salts solution (MSS) 20 g NH4Cl, 5 g KNO3, 8 g Na2SO4, 12 g K2HPO4, 4 g KH2PO4, 0.4 g MgSO4.7H2O (pH 7.0 ± 0.1) (Berlau et al. 1999a)

Unless otherwise noted, all media were reconstituted in 1 L of distilled water and autoclaved for 15 mins at 121°C

162

6.2.3 Analysis for Acinetobacter species

Sub-samples (25 g) of vegetable salads and lettuce were taken aseptically and transferred to 225 ml of 0.1% Bacteriological Peptone water. The lettuce samples comprised a mixture of sections of inner and outer leaves. Samples were placed in a

Stomacher bag (Seward, Norfolk, UK) and homogenised for one minute in a Stomacher

(Model BA6021, Seward).

The homogenate (1 ml) was serially diluted in 0.1% Bacteriological Peptone, from which samples (0.1 ml) were spread inoculated in duplicate onto plates of plate count agar (PCA, Oxoid, Adelaide, Australia), MLAA or MSA. The plates were incubated for

48 h at 30°C and presumptive Acinetobacter colonies were isolated and purified by streaking onto nutrient agar (NA, Oxoid). Presumptive Acinetobacter species were taken as pink colonies with a mauve background on MLAA and raised, white colonies on MSA. On MLAA, Acinetobacter species are distinguished by a combination of selective and differential effects. Enterobacteriaceae are differentiated by fermentation of the sugars (sucrose, fructose and mannitol) to produce yellow colonies, whereas

Acinetobacter do not ferment these sugars. The production of pink colonies with a mauve background for Acinetobacter species is due to the liberation of ammonium ions from nitrogenous materials. This liberation increases the alkalinity of the medium and changes the colour of the phenol red indicator to mauve (Mandel et al. 1964; Holton

1983; Jawad et al. 1994). MSA exploits the ability of Acinetobacter species to grow on low amounts of acetate (0.1%) and on media containing ammonium salts (Lautrop

1974).

Presumptive isolates that were Gram negative, oxidase negative, catalase positive, Hugh and Leifson’s oxidative or non-glucose utilising, and non-motile, were further identified 163

by sequencing regions of the 16S rDNA gene. Biomass from a 24 h culture on NA was inoculated into nutrient broth and incubated at 30°C for 24 h. Culture (1 ml) was transferred to a cryo-vial and centrifuged for 10 min at 10,000 g at 4°C to obtain a cell pellet. The supernatant was discarded and the cell pellet stored in a freezer (-80°C) until

DNA was extracted. Extraction of DNA, amplification by PCR using bacterial specific primers (341F/907R) and identification by sequencing followed the same methods as described in Chapter 4.

6.2.4 Evaluation of selective plating media for the isolation of Acinetobacter species

MLAA and MSA were evaluated for their ability to isolate species of Acinetobacter.

Both media were streaked with pure cultures of Acinetobacter baumannii, Ac. calcoaceticus, Ac. johnsonii, Ac. junii, Pseudomonas fluorescens, Ps. putida, Ps. fragi,

Citrobacter freundii, Stenotrophomonas maltophilia, Burkholderia cepacia, and

Klebsiella pneumoniae (Table 6.2) and incubated at 30°C for 48 h. Colony phenotypes of the different species were recorded. Both media were then assessed for their ability to detect known populations of Ac. baumannii. Acinetobacter baumannii was grown overnight at 30°C in nutrient broth (NB, Oxoid) and serially diluted to obtain populations of approximately 102, 103 and 106 cfu/g. These cell suspensions were used to inoculate samples of Iceberg lettuce. Lettuce samples (25 g) consisting of a mixture of inner and outer leaves in a Stomacher bag, were inoculated with 1 ml of cell suspension and thoroughly mixed. Bacteriological Peptone (225 ml, 0.1%, Oxoid) was then added to the Stomacher bag and the mixture macerated for one minute in a

164

Stomacher. The macerate was serially diluted and 0.1 ml of the diluted homogenate was spread inoculated, in duplicate, onto plates of PCA, MLAA and MSA incubated for 48 h at 30°C.

Table 6.2 Reference cultures used to evaluate media for the isolation of Acinetobacter species

Bacterial species Source Acinetobacter baumannii subsp. anitratus School of Biotechnology and (ATCC 15308) Biomolecular Sciences, UNSW Acinetobacter calcoaceticus* School of Chemical Sciences and Engineering , UNSW Acinetobacter johnsonii* School of Chemical Sciences and Engineering , UNSW Acinetobacter junii* School of Chemical Sciences and Engineering , UNSW Burkholderia cepacia* School of Chemical Sciences and Engineering , UNSW Citrobacter freundii (IMVS 1263) School of Chemical Sciences and Engineering , UNSW Klebsiella pneumoniae (ATCC 12657) School of Chemical Sciences and Engineering , UNSW Pseudomonas fluorescens (ATCC 13525) School of Chemical Sciences and Engineering , UNSW Pseudomonas fragi School of Chemical Sciences and Engineering , UNSW Pseudomonas putida School of Chemical Sciences and Engineering , UNSW Stenotrophomonas maltophilia School of Chemical Sciences and Engineering , UNSW

* Culture isolated and identified in this study

6.2.5 Evaluation of selective enrichment broths for growth of Acinetobacter species

MSS and BEM (Table 6.1) were evaluated for their ability to recover Acinetobacter species. Pure cultures of either Ac. baumannii, Ac. junii, Ac. johnsonii, Ps. fluorescens,

Cit. freundii or St. maltophilia were inoculated into tubes containing 10 ml of MSS or

165

BEM and incubated in an orbital shaker (150 rpm) at 30°C for 48 h. Shaking of enrichment broths led to enhanced growth of Acinetobacter (Baumann 1968). After incubation, one loopful of each selective enrichment culture was streaked onto MSA and MLAA and incubated at 30°C for 48 h.

BEM was then assessed for its ability to recover Ac. baumannii at low populations, from lettuce. Acinetobacter baumannii was grown overnight at 30°C in NB and serially diluted to obtain populations of approximately 1, 10 and 100 cfu/ml. These suspensions were used to inoculate samples of Iceberg lettuce. Lettuce samples (25 g) consisting of a mixture of inner and outer leaves in a Stomacher bag were inoculated with 1 ml of suspension and thoroughly mixed. BEM (225 ml) was then added to the Stomacher bag and the mixture macerated in a Stomacher for 1 min. The enrichment broth was then transferred to a sterile 500 ml flask and incubated at 30°C on an orbital shaker (150 rpm) for 48 h. At 24 h and 48 h of incubation, one loopful of enrichment broth was streaked onto plates of MSA and MLAA. These media were incubated at 30°C for 48 h and examined for the growth of Acinetobacter species.

Because Acinetobacter may be present at low populations on lettuce and ready-to-eat salads, a survey was conducted using enrichment culture. Sub-samples (25 g) of Iceberg lettuce (5) and salads (10) were aseptically taken and placed into a Stomacher bag.

BEM (225 ml) was added, and the sample was macerated for 1 min in a Stomacher. The enrichment broth was transferred to a sterile 500 ml flask and incubated on an orbital shaker (150 rpm, 30°C) for 48 h. At 24 h and 48 h of incubation, one loopful of enrichment culture was streaked onto plates of MSA and MLAA and incubated at 30°C for 48 h. Presumptive colonies of Acinetobacter were isolated and re-streaked for purity and identified as described already.

166

6.3 Results

6.3.1 Presence of Acinetobacter species in lettuce and vegetable salads as determined by direct plating onto MLAA and MSA

MLAA as well as a non-modified version have been used previously as plating media for the selective isolation and enumeration of Acinetobacter species from a diversity of clinical and environmental samples, including foods (Houang et al. 2001). Based on this history, MLAA was initially chosen as the plating medium to survey the presence of

Acinetobacter species on lettuce and salad vegetables. No Acinetobacter were detected

(<50 cfu/g) on a total of 10 lettuce and 40 vegetable salad samples when plated onto

MLAA, despite the fact that they had total bacterial populations between 102-106 cfu/g

(Table 6.3).

Colonies with typical Acinetobacter phenotypes were observed on MLAA agar plates, but they were not confirmed as Acinetobacter. On isolation, purification and sequencing, they were identified as species of Pseudomonas and Arthrobacter.

Table 6.3 Presence of Acinetobacter species in lettuce and ready-to-eat vegetable salads as determined by plating onto modified Leeds Acinetobacter agar (MLAA)

Sample No. samples Total aerobic plate count Acinetobacter population examined (cfu/g) (cfu/g) Lettuce 10 3.1 x 102 - 4.9 x 105 ND Ready-to-eat 35 2.3 x 103 - 4.7 x 106 ND salads Other 5 3.5 x 105 - 1.4 x 106 ND vegetable salads§

§ Other salads which were not packaged under modified atmosphere; ND, not detected (<50 cfu/g)

167

In a second trial, lettuce and vegetable salads were examined for the presence of

Acinetobacter species by plating onto MSA which has been used by Berlau et al.

(1999a) for the detection of Acinetobacter on environmental samples including vegetables. No Acinetobacter were detected (<50 cfu/g) in any of the 15 samples of salad vegetables, but six of the 20 lettuce samples gave presumptive colonies of

Acinetobacter, which were subsequently confirmed as Ac. calcoaceticus and Ac. johnsonii (Table 6.4). Depending on the sample, the populations of these Acinetobacter species ranged between 2.0 x 101-1.2 x 104 cfu/g (Table 6.4).

Although the lettuce and vegetable salad samples showed total plate count populations of 2.0 x 102-2.2 x 106 cfu/g, MSA was sufficiently selective to suppress the growth of background flora and allow the detection of Acinetobacter.

Table 6.4 Presence of Acinetobacter species in lettuce and ready-to-eat vegetable salads as determined by plating onto minimal salts agar (MSA)

Sample No. Total aerobic plate Acinetobacter Percentage samples count (cfu/g) population (cfu/g) isolation of examined Acinetobacter Lettuce 20 2.0 x 102 – 2.0 x 101 – 6/20 (30%) 2.2 x 106 1.2 x 104 Ready-to- 12 2.4 x 104 – ND 0 eat salads 6.2 x 106 Other 3 1.5 x 105 – ND 0 vegetable 2.1 x 106 salads§

§ Other salads which were not packaged under modified atmosphere pressure; ND - not detected (<50 cfu/g)

168

6.3.2 Comparison of MLAA and MSA media for the growth of Acinetobacter and other species

Failure to detect Acinetobacter on MLAA suggested that it may not be sufficiently selective to suppress interfering background microflora associated with raw and processed vegetables. To test this possibility, MLAA and MSA were examined for their ability to support the growth of pure cultures of the species listed in Table 6.2.

These results showed that MSA was the superior medium for the isolation of

Acinetobacter. All species of Acinetobacter used in this study were capable of growth on MSA whilst other bacterial species were suppressed (Table 6.5; Figure 6.1). In comparison, MLAA allowed the growth of Ac. baumannii and Ac. calcoaceticus but not

Ac. junii and Ac. johnsonii. MLAA also allowed the growth of species of Pseudomonas,

Citrobacter, Burkholderia and Stenotrophomonas, some of which produced colonies with colour and morphology similar to those of Acinetobacter (Table 6.5; Figure 6.2).

Table 6.5 Growth of Acinetobacter and other bacterial species on minimal salts agar (MSA) and modified Leeds Acinetobacter agar (MLAA)

Microorganism MSA MLAA Acinetobacter baumannii + + Acinetobacter calcoaceticus + + Acinetobacter junii + - Acinetobacter johnsonii + - Pseudomonas fluorescens + + Pseudomonas fragi - + Pseudomonas putida - + Citrobacter freundii - + Stenotrophomonas maltophilia - + Burkholderia cepacia - + Klebsiella pneumoniae - -

169

a b

c d

Figure 6.1 Pure and mixed cultures of bacterial species on MSA (a) mixed culture of bacteria isolated from a lettuce sample, Acinetobacter colony circled; (b) MSA, control, no microorganisms; (c) Acinetobacter baumannii; (d) Pseudomonas fluorescens

170

a b

c d

e f

Figure 6.2 Pure and mixed cultures of bacterial species on MLAA (a) mixed culture from lettuce sample; (b) MLAA, control plate, no microorganisms; (c) pure culture of Acinetobacter calcoaceticus; (d) pure culture of Pseudomonas fluorescens; (e) pure culture of Burkholderia cepacia; (f) pure culture of Stenotrophomonas maltophilia

171

To further evaluate the ability of both selective agar media to recover Acinetobacter, lettuce samples were spiked with known populations of Ac. baumannii (102, 103 and 106 cfu/g). Recovery of Ac. baumannii was observed to differ between both media (Table

6.6). Populations of 102 cfu/g and higher on lettuce were readily detected by plating on

MSA. However for plating on MLAA, Ac. baumannii could not be recovered from lettuce samples inoculated with 102 and 103 cfu/g, but it was recovered from samples inoculated with 106 cfu/g. The difference in performance between the two media was attributable to the growth of non-target species of bacteria on MLAA, therefore, masking the detection of Ac. baumannii colonies when inoculated at low populations.

Table 6.6 Detection of Acinetobacter baumannii inoculated onto Iceberg lettuce and isolated by spread plate culture on MSA and MLAA

Inoculum level per g lettuce Sample 102 cfu 103 cfu 106 cfu MLAA MSA MLAA MSA MLAA MSA 1 - (0) + (45%) - (0) + (25%) + (50%) + (90%) 2 - (0) + (90%) - (0) + (50%) + (45%) + (50%) 3 - (0) + (50%) - (0) + (90%) + (11%) + (75%) 4 - (0) + (75%) - (0) + (95%) + (14%) + (85%) 5 - (0) + (85%) - (0) + (85%) + (35%) + (90%)

+, detection of Acinetobacter species, -, no detection of Acinetobacter species, (%) percent recovery of inoculated Ac. baumannii

172

6.3.3 Presence of Acinetobacter species in lettuce and vegetable salads as determined by enrichment culture

Failure to detect Acinetobacter species in lettuce and vegetable salad samples by direct agar plating also suggested that the organisms were either absent from these products or was present at very low populations (<50-100 cfu/g). To test this possibility, samples were examined using enrichment culture. Two enrichment media, MSS and BEM were used to detect and isolate Acinetobacter species. In a preliminary evaluation with pure cultures, both media suppressed the growth of a range of potential, competitive non-

Acinetobacter species likely to be found in vegetable products (Table 6.7). Both media gave good growth of Ac. johnsonii and Ac. calcoaceticus but MSS did not give growth of Ac. baumannii. For this reason further trials were conducted with BEM only.

Lettuce samples spiked by inoculation with low populations of Ac. baumannii (1 cfu/g,

10 cfu/g, 100 cfu/g) gave recovery of the organism after enrichment in BEM for 48 h at

30°C followed by plating onto either MSA or MLAA (Table 6.8). BEM enrichment was sufficiently selective so that species which interfered with the recovery of Acinetobacter species on MLAA were no longer a factor. Enrichment in BEM followed by plating onto MSA and MLAA was used to survey the presence of Acinetobacter species in samples of lettuce and salad vegetables (Table 6.9). No Acinetobacter were detected in any of the 10 samples of salad vegetables examined (data not shown), but two out of the five samples of lettuce gave positive recovery of Acinetobacter with Ac. calcoaceticus in one sample and Ac. johnsonii in the other sample (Table 6.9).

173

Table 6.7 Growth of Acinetobacter and other bacterial species in enrichment media used for the isolation of Acinetobacter

Species Baumann Enrichment Minimal salts solution medium (BEM) (MSS) Acinetobacter baumannii + - Acinetobacter junii + + Acinetobacter johnsonii + + Pseudomonas fluorescens - - Citrobacter freundii - - Stenotrophomonas - - maltophilia Klebsiella pneumoniae - -

Table 6.8 Detection of low populations of Acinetobacter baumannii inoculated into Iceberg lettuce and recovered by enrichment in BEM, followed by plating onto MSA and MLAA

Population of Ac. baumannii Recovery on MSA MLAA 1 cfu/g + + 10 cfu/g + + 100 cfu/g + +

Table 6.9 Presence of Acinetobacter species in samples of lettuce as determined by enrichment in BEM, followed by plating onto MLAA and MSA

Sample MLAA MSA Species identity 1 + + Ac. calcoaceticus 2 - - ND 3 - - ND 4 + + Ac. johnsonii 5 - - ND

ND, no Acinetobacter species detected in 25 g lettuce

174

6.4 Discussion

Increasing consumer demands for fresh, ready-to-eat vegetable salad products have correlated with an increase in outbreaks of foodborne disease associated with these products (Desmarchelier 1996; Sloan 2000; Sivapalasingam et al. 2004). These trends have stimulated new interest and focus on the microbiological safety of fresh produce that is consumed without further processing or cooking (Beuchat 1998; Nguyen-the and

Carlin 2000; Beuchat 2002; Buck et al. 2003). Salmonella, Listeria monocytogenes and

E. coli have been the main bacteria of concern in this context, but there are also risks from other organisms such as Acinetobacter spp. which are a significant cause of nosocomial infections. Vegetable salads have been considered to be a potential source of these bacteria in hospital environments (Berlau et al. 1999a; Houang et al. 2001).

During a broader study on the microbial ecology of lettuce sold at retail outlets in

Sydney, Australia, (Chapter 4) we occasionally noted the isolation of Acinetobacter baumannii and these observations have provided the basis for the more focussed study reported here.

Over the years, a diversity of agar media has been reported for the selective and differential culture of Acinetobacter spp. (Mandel et al. 1964; Baumann 1968; Holton

1983; Jawad et al. 1994; Kämpfer 2000). Principally, these media were developed for the isolation of Acinetobacter from clinical samples and not environmental samples such as foods. Two such media are Herellea agar (Mandel et al. 1964) and Holton agar

(Holton 1983), but in a comparison of media for the isolation of Acinetobacter from both clinical and environmental samples, Jawad et al. (1994) found these media to be insufficiently selective, and to give lower recoveries of Acinetobacter spp. They

175

reported the formulation of a new medium, Leeds Acinetobacter medium (LAM), which gave superior isolation of Acinetobacter spp. However, this medium does not support the growth of some Acinetobacter spp. such as Ac. johnsonii, Ac. lwoffii, and Ac. haemolyticus (Jawad et al. 1994; Berlau et al. 1999a). Berlau et al. (1999a) reported better recoveries of Acinetobacter spp. from clinical samples on a Minimal Salts agar that was originally formulated by Juni (1972). A more recent study of by Houang et al.

(2004) reported the effective use of a modified Leeds Acinetobacter agar (MLAA) for the isolation of Acinetobacter spp. from clinical and environmental samples including foods. The modification involved decreasing the amount of vancomycin (10 mg/L to 3 mg/L) used as a selective agent in the medium. In a previous study it was reported that the modified medium allowed the growth of a greater range of Acinetobacter reference cultures (Chu et al. 1999).

On the basis of this literature, MLAA was chosen to screen for the presence of

Acinetobacter spp. in salad vegetables and lettuce samples. However, despite the analysis of some 50 samples of lettuce and ready-to-eat salads, and despite the fact that many of these samples contained significant populations of bacteria (103-106 cfu/g), no

Acinetobacter were detected (Table 6.3). It was noted that the isolation plates were overgrown by non-Acinetobacter spp., and that MLAA therefore was not sufficiently selective for the successful culture of Acinetobacter in lettuce or ready-to-eat salads.

Overgrowth by non-Acinetobacter species was not experienced when lettuce and ready- to-eat salad samples were cultured on MSA and, in this case, it was possible to detect

Acinetobacter spp. in lettuce samples (Table 6.4). The superiority of MSA over MLAA was further demonstrated by analysis for growth of pure cultures, where many non-

Acinetobacter species found on vegetables grew on MLAA, but not on MSA (Table

176

6.5). Also, in agreement with reports of Jawad et al. (1994) and Berlau et al. (1999a), some Acinetobacter spp. were not able to grow on MLAA. In controlled challenge studies, MSA was able to detect as few as 102 cfu/g of Ac. baumannii inoculated onto lettuce whereas 102-103 cfu/g were not detected by plating on MLAA.

Although Acinetobacter were found on samples of retail lettuce by plating onto MSA, they were not detected in ready-to-eat salads by this method (Table 6.4). To further investigate the presence of Acinetobacter in ready-to-eat salads an enrichment-plating method was used. BEM was selected as the enrichment medium as it gave good recovery of Acinetobacter spp. as well as preventing the growth of several species of bacteria such as Ps. fluorescens, Cit. freundii, St. maltophilia and Kl. pneumoniae, that are commonly found in ready-to-eat vegetables and interfere with the detection of

Acinetobacter spp. (Table 6.7). MSS is another selective enrichment medium used for the culture of Acinetobacter spp (Berlau et al. 1999a). However, it did not support the growth of the strain of Ac. baumannii used in our evaluation trials (Table 6.7) although this strain grew readily on MSA. It is difficult to explain this discrepancy, but it was repeatedly observed.

Enrichment in BEM followed by plating onto either MSA or MLAA has been effectively used in other studies to isolate Acinetobacter spp. from faecal specimens

(Grehn and von Graevenitz 1978), water sources (Guardabassii et al. 1999) and clinical and environmental samples (Baumann 1968; Dijkshoorn et al. 2005). However, despite the efficacy of this procedure as determined by previous literature and the evaluation trials of this Chapter, Acinetobacter spp. were never detected in samples of ready-to-eat salad (absent in 25 g sample). Nevertheless, they were recovered from lettuce samples using the same enrichment-plating procedure (Table 6.7). Since lettuce is a major

177

component of mixed vegetable salads, it needs to be explained why Acinetobacter are not found in these products. Possibly, they are removed by the operations used to wash the vegetable raw materials before salad preparation (Sapers 2001; Nascimento et al.

2003) or they are unable to survive under the atmospheric limitations of the packaged salad (Garcia-Gimeno and Zurera-Cosano 1997).

To the authors’ knowledge, previous studies have not specifically tested for the presence of Acinetobacter species in ready-to-eat packaged salads. However

Acinetobacter species including Ac. johnsonii, Ac. calcoaceticus, and Ac. baumannii have been isolated from vegetable produce including lettuce by previous researchers

(Berlau et al. 1999a; Houang et al. 2001). The detection of Ac. johnsonii and Ac. calcoaceticus in about 30% of lettuce samples examined are consistent with this previous work. Moreover, these two species as well as Ac. lwoffii frequently occur in cheese, other vegetables, meat and fish (Gennari and Lombardi 1993; Berlau et al.

1999a; Kämpfer 2000; Addis et al. 2001).

In conclusion, this study has shown that lettuce may harbour Acinetobacter spp. at populations up to 104 cfu/g and, therefore, could be a potential source of these bacteria in hospital and other health care institutions where nosocomial infections from

Acinetobacter are a significant issue. MLAA was not sufficiently selective for the examination of Acinetobacter in lettuce or ready-to-eat salads, as it was overgrown by non-Acinetobacter species. MSA was a superior medium for examination of

Acinetobacter in these types of foods, and allowed detection of populations as low as

102 cfu/g by direct plating. Lower populations can be detected by enrichment in BEM followed by plating onto MSA or MLAA. Further research is needed to evaluate the

178

effect of washing of produce and packaging environment on the survival and growth of

Acinetobacter species in ready-to-eat salad products

179

CHAPTER 7

THE EFFECT OF EXTRACTS FROM LETTUCE AND

OTHER VEGETABLES ON THE SURVIVAL OF

BACTERIA OF SPOILAGE AND PUBLIC HEALTH

SIGNIFICANCE

7.1 Introduction

During visits to growers of salad vegetables, it was observed that harvested lettuces were packed into boxes for transport. Sap and other exudates leached from the lettuce heads where the roots had been cut off, and came into contact with other lettuce heads packed in the same box. Such sap and exudates may have antimicrobial properties and, therefore, impact on the growth and survival of microorganisms associated with the surface of the leaves. In addition, processors of lettuce and other vegetables generate large amounts of waste, vegetable product, and have suggested that this may be a potential source of value added antimicrobial substances.

Natural antimicrobials can be obtained from many different sources such as microorganisms, vegetables, herbs, spices and fruits (Beuchat and Golden 1989; Walker

1994; Cowan 1999; Rauha et al. 2000; Burnett and Beuchat 2001; Ceylon and Fung

2004; Davidson and Taylor 2007). Past research has found that only a few types of vegetables may have inhibitory effects towards pathogenic or spoilage microorganisms.

These vegetables include cabbage, carrot, cauliflower, garlic and onion (Pederson and

180

Fisher 1944a, 1944b; Conner et al. 1986; Beuchat and Brackett 1990; Beuchat et al.

1994; Han and Kyung 1995; Cowan 1999; Beuchat 2006; Brandi et al. 2006).

Lettuce may be a potential source of antimicrobial compounds due to the presence of sesquiterpene lactones in some species. These lactones can have antimicrobial activity, but lettuces have not been thoroughly tested in this context. These sesquiterpene lactones include lactucin, lactucopicrin, 8-deoxylactucin and lettucenin A, and occur at different amounts in the latex or sap of different types of lettuce (Crosby 1963; Picman and Towers 1983; Takasugi et al. 1985; Mahmoud et al. 1986; Gromek et al. 1992).

One study found that the naturally associated lettuce microflora, Escherichia coli and

Salmonella were able to grow in lettuce juice stored at 10°C and 20°C, suggesting that microorganisms are not inhibited by the juice released from this vegetable (Maxcy

1982).

Further research is needed to determine if lettuce extracts are a possible source of antimicrobial substances. This chapter evaluates the inhibitory effect of extracts from lettuce and some other salad vegetables on species of bacteria of spoilage and public health significance in ready-to eat salad products.

181

7.2 Materials and Methods

7.2.1 Vegetable samples

Samples of lettuce (Iceberg and Cos) and capsicum (red and green) were purchased from local supermarkets and other shops in Sydney, Australia and analysed on the day of purchase.

7.2.2 Vegetable juice

Vegetable juice was prepared from Iceberg and Cos lettuce and red and green capsicums. For lettuce, a 100 g sample was taken and processed into juice using a

Breville Juice Fountain Professional (Breville Australia) taking about two minutes to process. For capsicum, one whole capsicum was used to produce juice using the

Breville Juice Fountain. The juice was then transferred to a 250 ml centrifuge tube and centrifuged at 13,000 rpm for 10 mins at 4°C (Avanti J-E, Beckman Coulter

Refrigerated Centrifuge, Beckman-Coulter Inc., California USA). The supernatant was decanted and filter sterilised using a 0.45 µm syringe filter. The filtrate was used in experiments and stored at 5°C until required. It was used within 24 hours of preparation.

Samples of the core sections of Iceberg and Cos lettuces were also processed into a filtered extract as just described.

7.2.3 Vegetable powder

Vegetable powder was prepared from fresh samples of lettuce and capsicum. Vegetables were cut into small pieces (~2 cm by 2 cm) and lyophilised in a freeze-dryer (Oerlikon 182

Leybold Vacuum, Germany) for 2-3 days. After drying, the lettuce was powdered using a mill and the capsicum was broken and powdered into smaller pieces using a Waring

Blender (Waring, USA). The powdered vegetable samples were stored in an airtight container at room temperature until required. Examples of Iceberg and Cos lettuce and red and green capsicum powders are shown in Figure 7.1.

a b

c d

Figure 7.1 Example of vegetable powders used for the extraction of antimicrobial compounds (a) Green capsicum; (b) Red capsicum; (c) Cos lettuce; (d) Iceberg lettuce

7.2.4 Solvent extraction of vegetable powders

Vegetable powders were extracted for their antimicrobial compounds using three different solvents, ethanol, methanol and acetone (Sigma-Aldrich, Australia). The 183

vegetable powders (500 mg), lettuce or capsicum, were weighed into a 15 ml centrifuge tube and an aliquot (10 ml) of solvent added. The tube was shaken for five minutes on a vortex mixer and then centrifuged at 3,000 rpm for 10 mins at room temperature. The supernatant was decanted to a round-bottomed flask and the remaining pellet was extracted with a further two aliquots of 10 ml of solvent, using the same procedure as described. After each successive extraction the supernatant was decanted to the same round-bottomed flask.

The flask with extracts was attached to a rotary evaporator (Buchii, Switzerland; waterbath temperature <40°C) and evaporated to dryness. The resulting dried powder was dissolved in a mixture of the same solvent used for extraction and water (50% v/v).

The dissolved extracts were stored in the refrigerator (4°C) until required.

7.2.5 Supercritical fluid extraction

Supercritical carbon dioxide extractions were performed at the Eiffel Technologies Ltd. laboratories, University of New South Wales (Lucien and Foster 2000).

A series of supercritical extractions were performed on each vegetable to determine the solubility of antimicrobial compounds present in the vegetable matrix.

For Iceberg lettuce powder, about 5 g was loaded into a bomb (refer to Figures 7.2 &

7.3) along with an equivalent amount of glass beads (Sigma, 30-500 microns). Each extraction was run at a constant temperature of 40°C with an extraction pressure of 200 bar for one to two hours. Extracts were collected at room temperature (23°C) and atmospheric pressure in either a filter or separator. Extractions of Iceberg lettuce

184

powder were run using i) carbon dioxide, and ii) carbon dioxide plus a modifier

(ethanol, 0.1 mole fraction).

For red capsicum powder, about 5 g was loaded into a bomb (refer to Figures 7.2 & 7.3) along with an equivalent amount of glass beads. The extraction was run using carbon dioxide and a constant temperature of 40°C with an extraction pressure of 200 bar for one to two hours. Extracts were collected at room temperature (23°C) and atmospheric pressure in either a filter or separator.

PI TI

V V2 V3 1 F1 V5 V6 F2 HC SP HPV SEP

H V4

Figure 7.2 Schematic diagram of supercritical fluid extraction apparatus used for extracting antimicrobial compounds from vegetable samples A- Carbon dioxide cylinder Vn- Valves: Sno-trik (V3 & V5), Swaglok (V1, V2, V4 & V5), Metering (V6) HC- Heating coil HPV- High pressure vessel (bomb) PI- Pressure indicator TI- Temperature indicator H- Heater Fn- Filter SEP- Separator (Lucien and Foster 2000)

185

Temperature Controller

Metering gauge & separator Syringe Pump

Extraction Vessel (Bomb)

Heating Coil

Pressure Indicator

Figure 7.3 Supercritical fluid extraction apparatus used for the extraction of antimicrobial compounds from Iceberg lettuce and red capsicum

7.2.6 Microorganisms

The reference microorganisms used throughout this study are listed in Table 7.1.

Microorganisms were maintained by subculture on Tryptone Soya agar (TSA, Oxoid,

Adelaide, Australia) at 30°C for 1-2 days and stored at 4°C. For use in experiments, cultures were activated by transferring loop inocula into 10 ml of Nutrient broth (NB,

Oxoid) and incubated at 30°C for 18-24 h.

186

Table 7.1 Reference cultures used to evaluate antimicrobial activity of vegetable extracts

Microorganism Source Acinetobacter junii* School of Chemical Sciences and Engineering, UNSWa Aeromonas hydrophila* School of Chemical Sciences and Engineering, UNSW Escherichia coli School of Biotechnology and Biomolecular Sciences, UNSW, (ATCCb 11775, UNSW 048200) Escherichia coli O111:NM School of Biotechnology and Biomolecular Sciences, UNSW Escherichia coli O157:H7 School of Biotechnology and Biomolecular Sciences, UNSW, (ATCC 35150) Listeria monocytogenes (L2) 1768 serotype 1/2a, TECRAc, Sydney Listeria monocytogenes (L4) 1771 serotype 4c, TECRA, Sydney (ATCC 19116) Listeria monocytogenes (L5) 1773 serotype 4e, TECRA, Sydney (ATCC 19118) Pseudomonas fluorescens School of Biotechnology and Biomolecular Sciences, UNSW, (UNSW 022600) Pseudomonas fluorescens School of Biotechnology and Biomolecular Sciences, UNSW, (ATCC 13525, UNSW 036800) Pseudomonas fragi* School of Chemical Sciences and Engineering, UNSW Pseudomonas putida* School of Chemical Sciences and Engineering, UNSW Salmonella enteritidis School of Biotechnology and Biomolecular Sciences, UNSW, (UNSW 031901) Salmonella typhimurium School of Chemical Sciences and Engineering, UNSW Serratia marcescens* School of Chemical Sciences and Engineering, UNSW a UNSW; University of New South Wales, Sydney, Australia b ATCC; American Type Culture Collection c TECRA; Tecra International, Frenchs Forest, Sydney, Australia * Isolated and characterized in Chapter 4

7.2.7 Antimicrobial properties of vegetable extracts

Two methods, the hole-in-plate and paper-disc diffusion assays, were used for the assessment of antimicrobial activity. Using Iso-sensitest agar (Oxoid), which is specifically formulated for the detection of susceptibility testing, the presence of antimicrobial activity was assessed based on the zone of clearing around the point of inoculation of the vegetable juice or extract (Deans and Ritchie 1987; Deans 1991).

The sterility of vegetable extracts obtained by supercritical fluid extraction was monitored by spread plating 0.1 ml of the extract across the surface of an Iso-sensitest 187

agar plate. This plate was incubated at 30°C for 24 h and checked for the presence of microbial growth.

Controls for each test consisted of a negative control (the solvent used for extraction of vegetable extract (100%)) and a positive control (300 ppm oxytetracycline, Sigma-

Aldrich).

7.2.8 Hole-in-plate method

Molten Iso-sensitest agar was prepared according to the manufacturers instructions and cooled to 50°C in a waterbath. An aliquot of overnight culture (1 ml) was placed into a sterile petrie dish along with 15 ml of molten Iso-sensitest agar, mixed and allowed to set. Equidistant holes (4 mm) were punched into the agar using a sterile cork borer and

15 µl of vegetable juice/extract was aseptically transferred into the hole.

Oxytetracycline (Sigma-Aldrich; 300 ppm) and sterile water were used as positive and negative controls, respectively. The vegetable juice/extract was allowed to diffuse into the agar by leaving the inoculated plates for 30-60 mins at room temperature before incubation at 30°C for 24 h.

7.2.9 Paper-disc diffusion assay

Plates of Iso-sensitest agar were prepared according to the manufacturers’ instructions and allowed to dry overnight at room temperature prior to use. An aliquot of the test cultures (0.1 ml) was spread across the surface of the agar and allowed to dry. Sterile discs (6 mm) were impregnated with lettuce or capsicum juice/extract (15 µl) before being placed at equidistant intervals onto the surface of the agar. The inoculated plates 188

were left at room temperature for 30-60 mins to allow for diffusion of the vegetable juice/extract into the agar and were incubated at 30°C for 24 h.

7.3 Results

The objective of this chapter was to assess the antimicrobial activity of extracts from lettuce and capsicum. This was achieved by assessing different extracts of the produce which included the juice, solvent extracts and supercritical fluid extracts.

7.3.1 Vegetable juice

Juice samples prepared from Iceberg and Cos lettuces and red and green capsicums did not exhibit any antimicrobial action against the bacteria as shown in Table 7.2 and

Figure 7.3. Figure 7.3 shows examples of the hole-in-plate method and paper disc diffusion assay. The positive controls in Figures 7.3b and 7.3c show the zone of clearing which demonstrate an antimicrobial activity against the reference culture. Similarly, results for the cores of lettuce showed no antimicrobial action (Table 7.2).

7.3.2 Evaluation of antimicrobial activity in solvent extracts from lettuce and capsicum

Solvent extracts prepared from Iceberg and Cos lettuces and red and green capsicums did not exhibit any antimicrobial action against the bacteria as shown in Table 7.3 and

7.4.

189

Table 7.2 Antimicrobial assay* of juice samples prepared from Iceberg and Cos lettuce, red and green capsicum and the core section of Iceberg lettuce

Vegetable Juice Microorganism Iceberg Red Green Lettuce Cos lettuce lettuce Capsicum Capsicum Core Acinetobacter junii No effect No effect No effect No effect No effect Aeromonas No effect No effect No effect No effect No effect hydrophila Escherichia coli No effect No effect No effect No effect No effect Escherichia coli No effect No effect No effect No effect No effect O111:NM Escherichia. coli No effect No effect No effect No effect No effect O157:H7 Listeria No effect No effect No effect No effect No effect monocytogenes (L2) Listeria No effect No effect No effect No effect No effect monocytogenes (L4) Listeria No effect No effect No effect No effect No effect monocytogenes (L5) Pseudomonas No effect No effect No effect No effect No effect fluorescens Pseudomonas No effect No effect No effect No effect No effect fluorescens (ATCC 13525) Pseudomonas fragi No effect No effect No effect No effect No effect Pseudomonas putida No effect No effect No effect No effect No effect Salmonella No effect No effect No effect No effect No effect enteritidis Salmonella No effect No effect No effect No effect No effect typhimurium Serratia marcescens No effect No effect No effect No effect No effect

* Assays by both hole in plate and paper disc diffusion methods

190

1 2 1 2

3 4 3 4

b c

Ethanol Ethanol

+ve -ve +ve -ve

Figure 7.3 Examples of hole-in-plate and paper disc diffusion assays for evaluation of antimicrobial activity of vegetable extracts a- Hole-in-plate and paper disc diffusion analysis for evaluation of antimicrobial activity against Pseudomonas fluorescens. 1. Iceberg lettuce juice; 2. Cos lettuce juice; 3. Iceberg lettuce core; 4. –ve control (sterile water) b- Control plate for hole-in-plate method using Pseudomonas fluorescens, +ve control (oxytetracycline, 300 ppm), -ve control (sterile water) c- Control plate for paper disc diffusion assay using Pseudomonas fluorescens, +ve control (oxytetracycline, 300 ppm), -ve control (sterile water)

191

Table 7.3 Antimicrobial assay* of ethanol, methanol and acetone extracts from Iceberg and Cos lettuce powders

Iceberg lettuce Cos Lettuce Microorganism Ethanol Methanol Acetone Ethanol Methanol Acetone Acinetobacter No effect No effect No effect No effect No effect No effect junii Aeromonas No effect No effect No effect No effect No effect No effect hydrophila Escherichia coli No effect No effect No effect No effect No effect No effect Escherichia coli No effect No effect No effect No effect No effect No effect O111:NM Escherichia coli No effect No effect No effect No effect No effect No effect O157:H7 Listeria No effect No effect No effect No effect No effect No effect monocytogenes (L2) Listeria No effect No effect No effect No effect No effect No effect monocytogenes (L4) Listeria No effect No effect No effect No effect No effect No effect monocytogenes (L5) Pseudomonas No effect No effect No effect No effect No effect No effect fluorescens Pseudomonas No effect No effect No effect No effect No effect No effect fluorescens (ATCC 13525) Pseudomonas No effect No effect No effect No effect No effect No effect fragi Pseudomonas No effect No effect No effect No effect No effect No effect putida Salmonella No effect No effect No effect No effect No effect No effect enteritidis Salmonella No effect No effect No effect No effect No effect No effect typhimurium Serratia No effect No effect No effect No effect No effect No effect marcescens

* Assays by both hole in plate and paper disc diffusion methods

192

Table 7.4 Antimicrobial assay* of ethanol, methanol and acetone extracts from red and green capsicum powders

Red Capsicum Green Capsicum Microorganism Ethanol Methanol Acetone Ethanol Methanol Acetone Acinetobacter junii No effect No effect No effect No effect No effect No effect Aeromonas No effect No effect No effect No effect No effect No effect hydrophila Escherichia coli No effect No effect No effect No effect No effect No effect Escherichia coli No effect No effect No effect No effect No effect No effect O111:NM Escherichia coli No effect No effect No effect No effect No effect No effect O157:H7 Listeria No effect No effect No effect No effect No effect No effect monocytogenes (L2) Listeria No effect No effect No effect No effect No effect No effect monocytogenes (L4) Listeria No effect No effect No effect No effect No effect No effect monocytogenes (L5) Pseudomonas No effect No effect No effect No effect No effect No effect fluorescens Pseudomonas No effect No effect No effect No effect No effect No effect fluorescens (ATCC 13525) Pseudomonas fragi No effect No effect No effect No effect No effect No effect Pseudomonas No effect No effect No effect No effect No effect No effect putida Salmonella No effect No effect No effect No effect No effect No effect enteritidis Salmonella No effect No effect No effect No effect No effect No effect typhimurium Serratia. No effect No effect No effect No effect No effect No effect marcescens

* Assays by both hole in plate and paper disc diffusion methods

193

7.3.3 Evaluation of antimicrobial activity in supercritical fluid extracts of lettuce and capsicum

Supercritical fluid extracts of Iceberg lettuce and red capsicums did not exhibit any antimicrobial action against the bacteria as shown in Table 7.5. These supercritical extracts were obtained using carbon dioxide only and carbon dioxide plus a modifier, ethanol (0.1 mole fraction) for Iceberg lettuce and using carbon dioxide with no modifier for red capsicum.

Table 7.5 Antimicrobial assay* of supercritical fluid extracts of Iceberg lettuce and red capsicum powders

Iceberg Lettuce Red Capsicum Microorganism Normal SFE SFE with modifier Normal SFE Acinetobacter junii No effect No effect No effect Aeromonas hydrophila No effect No effect No effect Escherichia coli No effect No effect No effect Escherichia coli No effect No effect No effect O111:NM Escherichia coli No effect No effect No effect O157:H7 Listeria monocytogenes No effect No effect No effect (L2) Listeria monocytogenes No effect No effect No effect (L4) Listeria monocytogenes No effect No effect No effect (L5) Pseudomonas fluorescens No effect No effect No effect Pseudomonas fluorescens No effect No effect No effect (ATCC 13525) Pseudomonas fragi No effect No effect No effect Pseudomonas putida No effect No effect No effect Salmonella enteritidis No effect No effect No effect Salmonella typhimurium No effect No effect No effect Serratia marcescens No effect No effect No effect

* Assays by both hole in plate and paper disc diffusion methods

194

7.4 Discussion

During visits to growers of salad vegetables, sap and other exudates were observed to leach from the lettuce heads where the roots had been cut off, and come into contact with other lettuce heads packed in the same box. The objective of this chapter was to evaluate the inhibitory effect of extracts from lettuce and capsicum on species of bacteria of spoilage and public health significance in ready-to eat salad products.

Extracts of lettuce and capsicum were produced by juicing, solvent extraction using ethanol, methanol and acetone, and supercritical fluid extraction and assessed using hole-in-plate method and paper-disc diffusion assay.

Two methods were used to assess the antimicrobial activity of lettuce and capsicum extracts, as there is no agreement in the literature on a standard method for such assays

(Deans 1991). Paper-disc diffusion assay involves the soaking of discs with extracts and placing them on the surface of agar seeded with bacteria. This technique does not always allow an accurate determination of the volume of extract added and, although the disc is applied to the agar surface, the extract would have diffused in three dimensions, giving a hemispherical zone difficult to measure accurately (Deans and

Ritchie 1987). In comparison, the hole-in-plate method overcomes some limitations of the paper-disc diffusion assay by punching holes through the agar, creating wells into which measured volumes (10 μl) of the extracts were placed. This ensures that radial diffusion from the well gives a clean and easily measured zone of inhibition (Bennett et al. 1966).

Initial studies on the juice of Iceberg and Cos lettuce showed no antimicrobial action against a range of spoilage and pathogenic microorganisms (Table 7.1). This result was

195

similar to that observed by Maxcy (1982), where naturally occurring lettuce microflora,

E. coli and Salmonella spp. were able to grow and survive in lettuce juice stored at 10°C and 20°C, suggesting that the juice is not inhibitory. The juices for the red and green capsicum also contained no antimicrobial activity. Juices extracted from green capsicum have been previously reported to have no antimicrobial activity against E. coli,

Klebsiella pneumoniae, Staphylococcus aureus, Staphylococcus epidermis, Shigella dysentriae and Salmonella typhosa (Al-Delaimy and Ali 1970; Lee et al. 2003). It was suggested by Al-Delaimy and Ali (1970) that many vegetable extracts contain bactericidal substances, some of which are lost during the extraction procedure or storage. The method and the time of preparing the extracts, the time and temperature of their storage, and the concentration of the extracts used are among the main factors to influence the effectiveness of any antibacterial activity of vegetable juices (Al-Delaimy and Ali 1970). The antimicrobial effects of carrots and cabbage juice are decreased by heating (Beuchat and Brackett 1990; Nguyen-the and Lund 1991; Beuchat et al. 1994;

Kyung and Fleming 1994; Han and Kyung 1995). Cooking, autoclaving, or increasing the temperature of the vegetable juice leads to the destruction or inactivation of the antimicrobial substances present in the raw vegetable (Pederson and Fisher 1944a).

Such factors might explain why no antimicrobial activity was observed for the vegetable juices examined in this study. Possibly, the processing time for the juices was too long, the slight increase in temperature to 30°C, aeration and shear forces may have inactivated any antimicrobial activity. If this were to be the case, however, it would reflect the fragility and weak stability of such constituents. Thus, their impact on the bacterial ecology of the lettuce or other vegetables would be very minimal (as suggested

196

by the data of Chapter 4), and they would have little practical application as novel preservatives.

Solvent extraction has been used in the past to extract antimicrobial compounds from plant materials (Eloff 1998; Nostro et al. 2000; Rauha et al. 2000; Palombo and Semple

2001; Alzoreky and Nakahara 2003; Rojas et al. 2003; Jabar and Al-Mossawi 2007). A range of different solvents can be used for the extraction of these antimicrobial compounds and include acetone, ethanol, methanol, chloroform, methylene dichloride and water (Eloff 1998). In this Chapter, acetone, methanol and ethanol were used to extract antimicrobial compounds from Iceberg and Cos lettuce, as well as red and green capsicums. Dried, powdered lettuce and capsicum were used as the starting material, as the water content of the fresh lettuce and capsicum may affect the solubility of the antimicrobial component and subsequent separation by liquid-liquid extraction (Eloff

1998). Unfortunately, no antimicrobial activity was observed for the ethanol, methanol and acetone extracts of Iceberg and Cos lettuce and red and green capsicum. Past research has shown similar results with methanol extracts of red capsicum (Capsicum annuum L.) having no antimicrobial activity against E. coli O157:H7 (Choi et al. 1998) and ethanol extracts of lettuce (Lactuca sativa) and capsicum having no antimicrobial effect against E. coli (Kim et al. 2001). However, strong antimicrobial activity was observed for ethanol extracts of lettuce against Lactobacillus casei and capsicum against Clostridium perfringens (Kim et al. 2001). Due to these contrasting results, it was considered that the antimicrobial compounds in lettuce may be present at low concentrations, unable to be extracted with sufficient efficiency by solvents. To assess this hypothesis, supercritical fluid extraction was evaluated for its ability to extract antimicrobial compounds from lettuce and capsicum.

197

Supercritical fluid extraction (SFE) has been used in the food industry for the extraction of essential oils from natural matrices (Reverchon 1997; Marongiu et al. 2003), antimicrobial compounds from Capsicum spp. (Yao et al. 1994; Peusch et al. 1997;

Yajima et al. 1997; Sato et al. 1999; Perva-Uzunalić et al. 2004), natural compounds from grape seeds (Palma and Taylor 1999), natural flavourings, antioxidants and decaffeination of coffee and tea (Raventós et al. 2002).

Supercritical fluid extraction is a separation process where the substances are dissolved in a fluid which is able to modify its dissolving power under specific conditions above their critical temperature and pressure (supercritical region). The properties of supercritical fluids provide a good extraction of the compounds due to their high dissolving power and high mass transfer rates of solutes into the fluid (Raventós et al.

2002). Carbon dioxide (CO2) is the ideal supercritical fluid in the food industry due to its critical temperature (31.06 °C), critical pressure (73.84 bar) and critical density is

3 (0.460 g/cm ) being relatively easy to achieve. The advantage of CO2 over conventional organic solvents is that it is non-toxic, non-flammable, non-polluting, completely recoverable, inexpensive, inert and its critical conditions are relatively safe and easy to reach (Raventós et al. 2002). Supercritical fluid extraction using carbon dioxide was applied to dried Iceberg lettuce and red capsicum powders. Unfortunately, no antimicrobial activity was observed for the extracts obtained.

To enhance the selectivity and solubility of potential antimicrobial compounds in

Iceberg lettuce powder, ethanol was added as a co-solvent to the carbon dioxide. A co- solvent has intermediate volatility between the supercritical fluid and the compound to be extracted. Co-solvents are usually liquid at room temperature and improve the selectivity and solute solubility by physical interactions with the solvent as well as by

198

specific chemical interaction with the solute. In food processing applications, ethanol, water and various gases are particularly suitable co-solvents (Modey et al. 1996;

Raventós et al. 2002). Unfortunately, using ethanol as a co-solvent also proved to be unsuccessful in this project with no antimicrobial activity being observed for the lettuce and capsicum extracts and the range of bacteria tested. Antimicrobial compounds have been extracted from capsicum in the past using supercritical fluid extraction; however, fractionation was required following SFE to yield an antimicrobial fraction from the residue (Yajima et al. 1997). Although a range of microorganisms including Bacillus cereus, Escherichia coli, Saccharomyces cerevisiae and Lactobacillus plantarum were tested, only Saccharomyces cerevisiae was inhibited by the antimicrobial fraction

(Yajima et al. 1997). Supercritical fluid extraction has also been used to extract capsaicinoids such as capsaicin and dihydrocapsaicin from capsicum (Yao et al. 1994;

Peusch et al. 1996; Sato et al. 1999; Perva-Uzunalić et al. 2004), which are the compounds known to have antimicrobial activity (Cowan 1999). However, these authors only focussed on the extraction of capsaicinoids from the capsicum and did not assess the extracts for antimicrobial activity. The quantity of active capsaicinoids was also found to be 3.2% for capsaicin and 0.58% for dihydrocapsaicin (Yao et al. 1994), which may be too low in concentration to have antimicrobial activity.

Lettuce may be a potential source of antimicrobial compounds due to the presence of sesquiterpene lactones in some species. However, from the results of this Chapter,

Iceberg and Cos lettuce have no antimicrobial activity. The sesquiterpene lactones present in lettuce include lactucin, lactucopicrin, 8-deoxylactucin and lettucenin A.

They are known to occur at different amounts in the latex or sap of different types of lettuce, which might explain the results obtained in this study (Crosby 1963; Picman

199

and Towers 1983; Takasugi et al. 1985; Mahmoud et al. 1986; Gromek et al. 1992).

The antimicrobial activity of these lactones may also only be active while the plant is growing as they have been isolated from the latex and sap. Unfortunately, a sample of the sap of freshly harvested lettuce was not taken for analysis, and this could be a topic for future research.

In conclusion, samples of Iceberg lettuce, Cos lettuce, red and green capsicum were evaluated for the presence of antimicrobial compounds by juicing, extraction with ethanol, methanol and acetone, and supercritical fluid extraction using carbon dioxide.

This was the first study to use supercritical fluid extraction on lettuce for the preparation of antimicrobial compounds. Unfortunately, no antimicrobial activity was observed for the range of bacteria tested, in any of the extracts from Iceberg and Cos lettuce, and red and green capsicum samples. Based on these data from several different extraction methods, it would appear that lettuce and capsicum do not contain constituents with significant antimicrobial activity. Future research might focus on freshly extracted sap or latex from lettuce as these fractions may contain substances overlooked in this investigation. On the other hand, microbial growth promoting substances may occur in extracts of lettuce and capsicums, and these may be worthy of investigation.

200

CHAPTER 8

CONCLUSIONS

The original objective of this project and thesis was to undertake a detailed investigation of the bacterial ecology of lettuce as it evolved and changed throughout the total processing chain, with particular emphasis on developments during the stage of cultivation at the farm. Unfortunately, this comprehensive ecological investigation could not be realised during the time available for this study because climatic drought conditions affected the cultivation of lettuce and availability of appropriate samples for analyses. Nevertheless, some new and significant information about the microbiology of lettuce and salad produce was obtained.

Pre-harvest contamination is recognized as a significant source of microorganisms in salad produce, including lettuce. Chapter 3 reports new information demonstrating that some pesticide preparations, commonly used during the cultivation of lettuces could, unintentionally, add to the microbial contamination of this produce, thereby impacting on their shelf-life and public health safety. Although commercial pesticide concentrates

(fungicides, herbicides and insecticides) were effectively sterile, their reconstitution and dilution in various sources of agricultural waters lead to their contamination. Such contaminants, which may include bacterial species of spoilage and public health significance, could grow to populations as high as 105-107 cfu/ml in some of the reconstituted pesticides prior to their application to the produce. The species that grew and their extent of growth varied with the type of pesticide and source of water. Further research is needed to understand what factors affect the growth and survival of

201

microorganisms in pesticide preparations. From this research, three strategies were proposed to minimise the risk of microbial contamination of lettuce and other salad produce by this mechanism: (i) design pesticide formulations with “inerts” or additives that do not support the growth of microorganisms, (ii) reconstitute pesticides in potable water rather than sources of agricultural water, and (iii) educate farmers to apply pesticides immediately after reconstitution, and to minimise storage conditions that would encourage microbial growth in these preparations.

The studies reported in Chapter 4 were undertaken to obtain basic information on the bacterial ecology of fresh heads of Iceberg lettuce, as available from supermarkets. To maximize the reliability of this information, both macerates and rinses of lettuce subsamples were examined, and Tween 80 was included in sample preparation methods to facilitate detachment and suspension of bacterial cells. Also, lettuce samples were analysed using conventional plate culture methods, and PCR-DGGE as a culture- independent, molecular method. The populations and species of bacteria recovered from lettuce varied with the sample, with populations ranging from 0-107 cfu/g. No differences were observed between the recoveries of total populations using maceration and rinsing methods, but incorporation of Tween 80 into the analyses usually recovered higher bacterial populations. Although a diversity of 37 bacterial species was isolated from the lettuce samples by culturing onto TSA, the most frequently isolated species were Pseudomonas synxantha/fluorescens, Curtobacterium flaccumfaciens and

Agrobacterium larrymoorei. The effect of maceration, rinse and the addition of Tween

80 on the recovery of total populations is novel and the presence of these particular species can lead to spoilage of the lettuce. There were only 10 out of 26 samples where the same most dominant species was recovered by maceration and rinsing methods. The

202

most frequently isolated species by maceration were Burkholderia cepacia, Ps. synxantha/fluorescens and Curt. flaccumfaciens. Rinsing of the lettuce samples led to an increase in the isolation of Serratia marcescens, along with detection of

Microbacterium spp. and a similar frequency of isolation for Ps. synxantha/fluorescens.

The addition of Tween 80 to the macerate solution led to an increased recovery of Ps. synxantha/fluorescens and Stenotrophomonas maltophilia, while rinsing with Tween 80 in solution led to a decrease in the isolation of Burk. cepacia and Pantoea agglomerans.

There is a need for more research into the effect of sampling methods on the recovery of predominant species from other vegetable products such as ready-to-eat salads.

PCR-DGGE analysis of the same lettuce samples was unable to detect the same incidence and profile of bacterial species as found by culturing onto TSA. A majority of samples gave no detection of bacterial species by PCR-DGGE. In some cases, low bacterial populations on the lettuce may have accounted for these data, as they were below the limit of detection by PCR-DGGE (e.g. 103-104 cfu/g). However, there were cases where lettuce had 104-105 cfu/g of some bacterial species, but these were not detected by PCR-DGGE, suggesting that other factors were contributing to the poor performance of PCR-DGGE analyses. Nevertheless, PCR-DGGE was able to detect some Bacillus and Pseudomonas species, as well as Ser. marcescens that were not detected by plate culture, and enhance the detection of Acinetobacter species.

Consequently, there is the possibility that viable but non culturable organisms may occur on lettuce, and this concept requires more detailed investigation.

Failure of the PCR-DGGE analyses to reliably detect bacterial species in lettuce extracts was probably related to interference of plant DNA and possibly other plant constituents with the PCR assay. In particular, plant chloroplast DNA was detected in DGGE gels,

203

and gave a strong band with mobility similar to some of the main bacterial species (e.g.

Curt. flaccumfaciens and Ag. larrymoorei) found by culture methods. This factor would mask the detection of bacterial species. Further research was conducted in Chapter 5 to determine if removal of the plant chloroplast cells from lettuce macerates would improve the efficacy of PCR-DGGE for bacterial analysis. The use of differential centrifugation or titanium (III) hydroxide adsorption was ineffective in separating bacterial cells and chloroplast cells for the purposes of PCR-DGGE analyses.

Preparative agarose gel electrophoresis was found to be effective in separating plant chloroplast DNA from bacterial cell DNA in lettuce macerates. The bacterial DNA obtained by this procedure could be analysed by PCR-DGGE, without interference from chloroplast DNA. Insufficient time was available to apply and evaluate this new observation in a more detailed study of the bacterial ecology of lettuce and other vegetable produce, and this remains as a direction for further research.

Data from bacterial analyses conducted in Chapters 3 and 4 suggested the presence of

Acinetobacter species in some lettuce samples and agricultural waters used during their cultivation. Because nosocomial infections with some antibiotic resistant species of

Acinetobacter have become a significant and recent issue in Australian hospitals as well as hospitals in other countries, along with suggestions that salad vegetables may be a source of these organisms, Chapter 6 presented a more detailed study of the occurrence of Acinetobacter species in lettuce and salad products. Several plate culture and enrichment media such as modified Leeds Acinetobacter agar, minimal salts agar,

Baumann enrichment medium, and minimal salts solution, previously used to detect and isolate Acinetobacter species from environmental and food samples were found to be unreliable due to overgrowth by other bacterial species (e.g. Pseudomonas and

204

Arthrobacter) commonly found on lettuce. Plating onto minimal salts agar, or enrichment in Baumann enrichment medium followed by plating on modified Leeds

Acinetobacter agar or minimal salts agar, gave superior data for the detection and enumeration of Acinetobacter species in lettuce and salad products. Using these conditions, Ac. baumannii and Ac. johnsonii were detected in 30% of lettuce samples at populations up to 104 cfu/g, demonstrating that lettuce produce could be a potential source of these pathogens in hospital environments. Notably, no Acinetobacter species were detected in ready-to-eat salads, despite the prominent presence on lettuce. The reasons for this unexpected observation are not evident and require further study.

Extracts of lettuce and other salad produce may contain antimicrobial constituents that could impact on measurements of their bacterial ecology and could also have commercial potential as novel preservation agents. In Chapter 7, samples of Iceberg lettuce, Cos lettuce, and red and green capsicums were evaluated for the presence of antimicrobial compounds in their juices and in ethanol, methanol and acetone extracts.

No antimicrobial activity against a range of bacteria of spoilage and public health significance was detected. Further evaluations were conducted using extracts obtained by supercritical fluid extraction with carbon dioxide and, again, no antimicrobial activity was detected. Based on these data from several different extraction methods, it would appear that lettuce and capsicum do not contain constituents with significant antimicrobial activity.

205

CHAPTER 9

BIBLIOGRAPHY

Abbo, A., Navon-Venezia, S., Hammer-Muntz, O., Krichali, T., Siegman-Igra, Y.,

Carmeli, Y. 2005. Multidrug-resistant Acinetobacter baumannii. Emerging and

Infectious Diseases 11: 22-29

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. Journal of Infectious Diseases 177: 1588-1593

Adams, M.R., Hartley, A.D., Cox, L.J. 1989. Factors affecting the efficacy of washing procedures used in the production of prepared salads. Food Microbiology 6: 69-77

Adams, M.R., Moss, M.O. 1995. Food Microbiology. The Royal Society of Chemistry,

Cambridge: 181-182

Addis, E., Fleet, G.H., Cox, J.M., Kolak, D., Leung, T. 2001. The growth, properties and interactions of yeasts and bacteria associated with the maturation of Camembert and blue-vein cheeses. International Journal of Food Microbiology 69: 25-36

206

Ait Melloul, A., Hassani, L., 1999. Salmonella infection in children from the wastewater-spreading zone of Marrakesh city (Morocco). Journal of Applied

Microbiology 87: 536-539

Albrecht, J.A., Hamouz, F.L., Sumner, S.S., Melch, V. 1995. Microbial evaluation of vegetable ingredients in salad bars. Journal of Food Protection 58: 683-685

Al-Delaimy, K.S., Ali, S.H. 1970. Antibacterial action of vegetable extracts on the growth of pathogenic bacteria. Journal of the Science of Food and Agriculture 21: 110-

112

Alonso, J.L., Amoros, I., Alonso, M.A. 1996. Differential susceptibility of aeromonads and coliforms to cefsulodin. Applied and Environmental Microbiology 62: 1885-1888

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J. 1990. Basic local alignment search tool. Journal of Molecular Biology 215: 403-410

Alzoreky, N.S., Nakahara, K. 2003. Antibacterial activity of extracts from some edible plants commonly consumed in Asia. International Journal of Food Microbiology 80:

223-230

Amann, R., Kuhl, M. 1998. In situ methods for assessment of microorganisms and their activities. Current Opinions in Microbiology 1: 352-358

207

Ampe, F., Omar, N.B., Moizan, C., Wacher, C., Guyot, J-P. 1999. Polyphasic study of the spatial distribution of microorganisms in Mexican Pozol, a fermented maize dough, demonstrates the need for cultivation-independent methods to investigate traditional fermentations. Applied and Environmental Microbiology 65: 5464-5473

Ampe, F., Sirvent, A., Zakhia, N. 2001. Dynamics of the microbial community responsible for traditional sour cassava starch fermentation studied by denaturing gradient gel electrophoresis and quantitative rRNA hybridization. International Journal of Food Microbiology 63: 45-54

Andrews, J.H., Kenerley, C.M. 1978. The effects of a pesticide program on non-target epiphytic microbial populations of apple leaves. Canadian Journal of Microbiology 24:

1058-1072

Anonymous. 2002. Bouncing Baby Lettuces. In Department of Agriculture, Fisheries and Forestry. Made in Australia, p. 17-18

Anzai, Y., Kim, H., Park, J-Y., Wakabayashi, H., Oyaizu, H. 2000. Phylogenetic affiliation of the pseudomonads based on 16S rRNA sequence. International Journal of

Systematic and Evolutionary Microbiology 50: 1563-1589

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: 239-248

208

Arumugaswamy, R.K., Ali, G.R.R., Hamid, S.N.B.A. 1994. Prevalence of Listeria monocytogenes in foods in Malaysia. International Journal of Food Microbiology 23:

117-121

Arvanitidou, M., Stathopoulos, G.A., Constantinidis, T.C., Katsouyannopoulos, V.

1995. The occurrence of Salmonella, Campylobacter and Yersinia spp. in river and lake waters. Microbiology Research 150: 153-158

AS 5013.1-2004. Examination for specific organisms- Standard Plate Count

AS 5013.11.1-2004. Microbiology of food and animal feeding stuffs- Preparation of test samples, initial suspension and decimal dilutions for microbiological examination-

General rules for the preparation of the initial suspension and decimal dilutions

Australian Food Statistics. 2006. Food and Agriculture Division, Australian

Government Department of Agriculture, Fisheries and Forestry

Aycicek, H., Oguz, U., Karci, K. 2006. Determination of total aerobic and indicator bacteria on some raw eaten vegetables from wholesalers in Ankara, Turkey.

International Journal of Environmental Health 209: 197-201

Babic, I., Roy, S., Watada, A.E., Wergin, W.P. 1996. Changes in microbial populations on fresh cut spinach. International Journal of Food Microbiology 31: 107-119

209

Bae, S., Fleet, G.H., Heard, G.M. 2006. Lactic acid bacteria associated with wine grapes from several Australian vineyards. Journal of Applied Microbiology 100: 712-727

Barriga, M.I., Trachy, G., Williemot, C., Simard, R.E. 1991. Microbial changes in shredded iceberg lettuce stored under controlled temperatures. Journal of Food Science

56: 1586-1588

Baskaran, N., Kandpal, R.P., Bhargava, A.K., Glynn, M.W., Bale, A., Weissman, S.M.

1996. Uniform amplification of a mixture of deoxyribonucleic acids with varying GC content. Genome Research 6: 633-638

Bastos, R.K.X., Mara, D.D. 1995. The bacterial quality of salad crops drip and furrow irrigated with waste stabilisation pond effluent: an evaluation of the WHO guidelines.

Water Science and Technology 31: 425-430

Bastos, M.S.R., Soares, N.F.F., Andrade, N.J., Arruda, A.C., Alves, R.E. 2005. The effect of the association of sanitisers and surfactant in the microbiota of the cantaloupe

(Cucumis melo L.) melon surface. Food Control 16: 369-373

Baumann, P. 1968. Isolation of Acinetobacter from soil and water. Journal of

Bacteriology 96: 39-42

210

Beckmann, J., Tazik, P., Gorden, R. 1984. Effects of two herbicides on selected aquatic bacteria. Bulletin of Environmental Contamination and Toxicology 32: 243-250

Bennett, J.V., Brodie, J.L., Benner, E.J., Kirby, W.M.M. 1966. Simplified, accurate method for antibiotic assay of clinical specimens. Applied Microbiology 14: 170-177

Bennik, M.H.J., Vorstman, W., Smid, E.J., Gorris, L.G.M. 1998. The influence of oxygen and carbon dioxide on the growth of prevalent Enterobacteriaceae and

Pseudomonas species isolated from fresh and controlled-atmosphere-stored vegetables.

Food Microbiology 15: 459-469

Benoit, V., Raiche, P., Smith, M.G., Guthrie, J., Donnelly, E.F., Julian, E.M., Lee, R.,

DiMaio, S., Rittmann, M., Matyas, B.T. 1994. Foodborne outbreaks of enterotoxigenic

Escherichia coli – Rhode Island and New Hampshire, 1993. Journal of the American

Medical Association 271: 652-654

Bergmire-Sweat, D., Marengo, L., Pendergrass, P., Hendricks, K., Garcia, M.,

Drumgoole, R., Baldwin, T., Kingsley, K., Walsh, B., Lang, S., Prine, L., Busby, T.,

Trujillo, L., Perrotta, D., Hathaway, A., Jones, B., Jaiyeola, A., Bengtson, S. 2000.

Escherichia coli O111:H8 outbreak among teenage campers- Texas, 1999. Morbidity and Mortality Weekly Report 49: 321-324

211

Bergogne-Bérézin, E., Towner, K.J. 1996. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical and epidemiological features. Clinical Microbiological

Reviews 9: 148-165

Berlau, J., Aucken, H.M., Houang, E., Pitt, T.L. 1999a. Isolation of Acinetobacter spp including A. baumannii from vegetables: implications for hospital-acquired infections.

Journal of Hospital Infection 42: 201-204

Berlau, J., Aucken, H., Malnick, H., Pitt, T. 1999b. Distribution of Acinetobacter species on skin of healthy humans. European Journal of Clinical and Microbiological

Infectious Diseases 18: 179-183

Besser, R.E., Lett, S.M., Weber, J.T., Doyle, M.P., Barrett, T.J., Wells, J.G., Griffin,

P.M. 1993. An outbreak of diarrhea and hemolytic uremic syndrome from Escherichia coli O157:H7 in fresh-pressed apple cider. Journal of the American Medical Association

269: 2217-2220

Betts, G. 2006. Other spoilage bacteria. In Food Spoilage Microorganisms, ed.

Blackburn, C. de W. pp. 668-694. Woodhead Publishing Co., Cambridge,

Beuchat, L.R. 1996a. Pathogenic organisms associated with fresh produce. Journal of

Food Protection 59: 204-216

212

Beuchat, L.R. 1996b. Listeria monocytogenes: incidence on vegetables. Food Control 7:

223-228

Beuchat, L.R. 1998. Surface decontamination of fruits and vegetables eaten raw: a review. Food Safety Unit, World Health Organisation.

Beuchat, L.R. 2002. Ecological factors influencing survival and growth of human pathogens on raw fruits and vegetables. Microbiology and Infection 4: 413-423

Beuchat, L.R. 2006. Sampling, detection, and enumeration of pathogenic and spoilage microorganisms. In Sapers, G.M., Gorny, J.R., Yousef, A.E. (eds). Microbiology of

Fruits and Vegetables. CRC Press, Boca Raton, FL: 543-564

Beuchat, L.R., Brackett, R.E. 1990. Inhibitory effects of raw carrot on Listeria monocytogenes. Applied and Environmental Microbiology 56: 1734-1742

Beuchat, L.R., Brackett, R.E., Doyle, M.P. 1994. Lethality of carrot juice to Listeria monocytogenes as affected by pH, sodium chloride and temperature. Journal of Food

Protection 57: 470-474

Beuchat, L.R., Farber, J.M., Garrett, E.H., Harris, L.J., Parish, M.E., Suslow, T.V.,

Busta, F.F. 2001. Standardisation of a method to determine the efficacy of sanitizers in inactivating human pathogenic microorganisms on raw fruits and vegetables. Journal of

Food Protection 64: 1079-1084

213

Beuchat, L.R., Golden, D.A. 1989. Antimicrobials occurring naturally in foods. Food

Technology 43: 134-142

Beuchat, L.R., Nail, B.V., Adler, B.B., Clavero, M.R.S. 1998. Efficacy of spray application of chlorinated water in killing pathogenic bacteria on raw apples, tomatoes, and lettuce. Journal of Food Protection 61: 1305-1311

Beuchat, L.R., Scouten, A.J. 2004. Factors affecting survival, growth, and retrieval of

Salmonella Poona on intact and wounded cantaloupe rind and in stem scar tissue. Food

Microbiology 21: 683-694

Bhagwhat, A.A. 2004. Rapid detection of Salmonella from vegetable rinse-water using real-time PCR. Food Microbiology 21: 73-78

Bidawid, S.P., Edeson, J.F.B., Ibrahim, J., Matossian, R.M. 1978. The role of non-biting flies in the transmission of enteric pathogens (Salmonella species and Shigella species) in Beirut, Lebanon. Annals of the Tropical and Medical Parasitology 72: 117-121

Blaiotta, G., Pennacchia, C., Ercolini, D., Moschetti, G., Villiani, F. 2003. Combining denaturing gradient gel electrophoresis of 16S rDNA V3 region and 16S-23S rDNA spacer region polymorphism analyses for the identification of staphylococci from Italian fermented sausages. Systematic and Applied Microbiology 26: 423-433

214

Blaser, M.J., Hardesty, H.L., Powers, B., Wang, W-L.L. 1980. Survival of

Campylobacter fetus subsp. jejuni in biological milieus. Journal of Clinical

Microbiology 11: 309-313

Bolton, F.J., Coates, D., Hutchinson, D.N., Godfree, A.F. 1987. A study of thermophilic campylobacters in a river system. Journal of Applied Bacteriology 62: 167-176

Boon, N., De Windt, W., Verstraete, W., Top, E.M. 2002. Evaluation of nested PCR-

DGGE (denaturing gradient gel electrophoresis) with group-specific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants.

FEMS Microbiology Ecology 39: 101-112

Bouvet, P.J.M., Grimont, P.A.D. 1986. Taxonomy of the genus Acinetobacter with the recognition of Acinetobacter baumannii sp. nov., Acinetobacter haemolyticus sp. nov.,

Acinetobacter johnsonii sp. nov., and Acinetobacter junii sp. nov. and emended description of Acinetobacter calcoaceticus and Acinetobacter lwoffii. International

Journal of Systematic Bacteriology 36: 228-240

Brackett, R.E. 1997. Fruits, vegetables and grains. In Doyle, M.P., Beuchat, L.R.,

Montville, T.J. (eds). Food Microbiology, Fundamentals and Frontiers. American

Society of Microbiology Press, Washington DC: 117-126

Brackett, R.E. 1998. Changes in the microflora of packaged fresh tomatoes. Journal of

Food Quality 11: 89-105

215

Brackett, R.E. 1999. Incidence, contributing factors, and control of bacterial pathogens in produce. Postharvest Biology and Technology 15: 305-311

Brandi, G., Amagliani, G., Schiavano, G.F., De Santi, M., Sisti, M. 2006. Activity of

Brassica oleracea leaf juice on foodborne pathogenic bacteria. Journal of Food

Protection 69: 2274-2279

Breidt, Jr., F. 2006. Safety of minimally processed, acidified, and fermented vegetable products. In Sapers, G.M., Gorny, J.E., Yousef, A.E. (eds). Microbiology of Fruits and

Vegetables. CRC Press, Boca Raton, FL: 313-335

Brocklehurst, T.F., Lund, B.M. 1981. Properties of pseudomonads causing spoilage of vegetables stored at low temperature. Journal of Applied Bacteriology 50: 259-266

Brocklehurst, T.F., Zaman-Wong, C.M., Lund, B.M. 1987. A note on the microbiology of retail packs of prepared salad vegetables. Journal of Applied Bacteriology 63: 409-

415

Brunk, C.F., Eis, N. 1998. Quantitative measure of small-subunit rRNA gene sequences of the kingdom Korarchaeota. Applied and Environmental Microbiology 64: 5064-5066

Bryan, F.L. 1977. Diseases transmitted by foods contaminated by wastewater. Journal of Food Protection 40: 45-56

216

Buck, J.W., Walcott, R.R., Beuchat, L.R. 2003. Recent trends in microbiological safety of fruits and vegetables. Plant Health Progress, 21 January 2003

Burks, C., Cinkosky, M.J., Fischer, W.M., Gilna, P., Hayden, J.E.-D., Keen, G.M.,

Kelly, M., Kristofferson, D., Lawrence, J. 1992. Genbank. Nucleic Acids Research 20

(Supplement): 2065-2069

Burnett, A.B., Beuchat, L.R. 2001. Comparison of sample preparation methods for recovering Salmonella from raw fruits, vegetables and herbs. Journal of Food Protection

64: 1459-1465

Calvente, V., Benuzzi, D., Obuchowicz, N., Hough, G., Sanz de Tosetti, M.I. 1999.

Changes in surface microflora of apple and pear fruits by application of pesticides and their relation with biocontrol of post-harvest diseases. Agro-food Industry Hi-Tech 10:

30-33

Campbell, J.V., Mohle-Boetani, J., Reporter, R., Abbott, S., Farrar, J., Brandl, M.,

Mandrell, R., Werner, S.B. 2001. An outbreak of Salmonella serotype Thompson associated with fresh cilantro. Journal of Infectious Diseases 183: 984-987

Campbell, N.A. 1996. Biology, 4th Edition. Benjamin Cummings, California

217

Canfield, M.L., Moore, L.W. 1991. Isolation and characterisation of opine-utilising strains of Agrobacterium tumefaciens and fluorescent strains of Pseudomonas spp. from rootstocks of Malus. Phytopathology 81: 440-443

Cantoni, C., Iacumin, L. 2003. Acinetobacter spp. contamination of soft drinks.

Industrie delle Bevande 32: 35-37

Capuano, D.M., Okino, M.H.T., Bettini, M.C.B., Mangini, A.C.S. 2001. Occurrence of

Cryptosporidium spp. in vegetables commercialised in Ribeirao Preto City, Sao Paulo

State, Brazil. Reviews of the Institute of Adolfo Lutz 60: 89-91

Carlin, F., Nguyen-the, C., Abreu da Silva, A. 1995. Factors affecting the growth of

Listeria monocytogenes on minimally processed endive. Journal of Applied

Bacteriology 78: 636

Carlin, F., Nguyen-the, C., Da Silva, A.A., Cochet, C. 1996. Effects of carbon dioxide on the fate of Listeria monocytogenes, aerobic bacteria and on the development of spoilage in minimally processed fresh endive. International Journal of Food

Microbiology 32: 159-172

Carr, E.L., Kämpfer, P., Patel, B.K.C., Gürtler, V., Seviour, R.J. 2003. Seven novel species of Acinetobacter isolated from activated sludge. International Journal of

Systematic and Evolutionary Microbiology 53: 953-963

218

Centers for Disease Control and Prevention (CDC). 2006. Ongoing multistate outbreak of Escherichia coli serotype O157:H7 infections associated with consumption of fresh spinach- United States 2006. Morbidity and Mortality Weekly Report 55: 1045-1046

Ceponis, M.J., Friedman, B.A. 1958. Pectolytic enzymes of Pseudomonas marginalis and their effects on lettuce. Phytopathology 49: 141-144

Ceylon, E., Fung, D.Y.C. 2004. Antimicrobial activity of spices. Journal of Rapid

Methods and Automation in Microbiology 12: 1-55

Chakrabarti, R., Schutt, C.E. 2001. The enhancement of PCR amplification by low molecular-weight sulfones. Gene 274: 293-298

Chakrabarti, R., Schutt, C.E. 2001. The enhancement of PCR amplification by low molecular weight amides. Nucleic Acids Research 29: 2377-2381

Chandler, D.P., Fredrickson, J.K., Brockman, F.J. 1997. Effect of PCR template concentration on the composition and distribution of total community 16S rDNA clone libraries. Molecular Ecology 6: 475-482

Cheng-An, H., Beuchat, L.R. 1995. Efficacy of selected chemicals for killing pathogenic and spoilage microorganisms on chicken skin. Journal of Food Protection

58: 19-23

219

Cherry, J.P. 1999. Improving the safety of fresh produce with antimicrobials. Food

Technology 53: 54-56, 59

Cho, J-C., Kim, S-J. 2000. Increase in bacterial community diversity in subsurface aquifers receiving livestock wastewater input. Applied and Environmental

Microbiology 66: 956-965

Choi, O-K., Joo, I., Kim, K., Sung, C.K. 1998. Screening of antimicrobial activity against enterohaemorrhagic Escherichia coli O157:H7 from plants in Korea. Journal of

Food Science and Nutrition 3: 324-328

Chu, Y.W., Leung, C.M., Houang, E.T.S., Ng, K.C., Leung, C.B., Leung, H.Y., Cheng,

A.F.B. 1999. Skin carriage of Acinetobacters in Hong Kong. Journal of Clinical

Microbiology 37: 2962-2967

Cichewicz, R.H., Thorpe, P.A. 1996. The antimicrobial properties of chile peppers

(Capsicum species) and their uses in Mayan medicine. Journal of Ethnopharmacology

52: 61-70

Cieslak, P.R., Barrett, T.J., Griffin, P.M., Gensheimer. K.F., Beckett, G., Buffington, J.,

Smith, M.G. 1993. Escherichia coli O157:H7 infection from a manured garden. Lancet

342: 367

220

Claflin, L.E., Stuteille, D.L., Armbrust, D.V. 1973. Wind-blown soil in epidemiology of bacterial leaf spot of alfalfa and common blight of bean. Phytopathology 63: 1417-1419

Clement, B.G., Kehl, L.E., DeBord, K.L., Kitts, C.L. 1998. Terminal restriction fragment patterns (TRFPs), a rapid, PCR-based methods for the comparison of complex bacterial communities. Journal of Microbiological Methods 31: 135-142

Cocolin, L., Innocente, N., Biasutti, M., Comi, G. 2004. The late blowing in cheese: a new molecular approach based on PCR and DGGE to study the microbial ecology of the alteration process. International Journal of Food Microbiology 90: 83-91

Cocolin, L., Manzano, M., Aggio, D., Cantoni, C., Comi, G. 2001a. A novel polymerase chain reaction (PCR) – denaturing gradient gel electrophoresis (DGGE) for the identification of Micrococcaceae strains involved in meat fermentations. Its application to naturally fermented Italian sausages. Meat Science 57: 59-64

Cocolin, L., Manzano, M., Cantoni, C., Comi, G. 2001b. Denaturing gradient gel electrophoresis analysis of the 16S rRNA gene V1 region to monitor dynamic changes in the bacterial population during fermentation of Italian sausages. Applied and

Environmental Microbiology 67: 5113-5121

Coghlan, A.C. 2000. Food poisoning bugs thrive in crop sprays. New Scientist 7th Oct:

221

Cohen, D., Green, M., Block, C., Slepon, R., Ambar, R., Wasserman, S.S., Levine,

M.M. 1991. Reduction of transmission of shigellosis by control of houseflies (Musca domestica). Lancet 337: 993-997

Collins, M.D., Jones, D. 1983. Reclassification of Corynebacterium flaccumfaciens,

Corynebacterium betae, Corynebacterium oortii and Corynebacterium poinsettiae in the genus Curtobacterium, as Curtobacterium flaccumfaciens comb. nov. Journal of

General Microbiology 129: 3545-3548

Connor, A.J. 1983. The comparative toxicology of vineyard pesticide to wine yeasts.

American Journal of Enology and Viticulture 34: 278-279

Conner, D.E., Brackett, R.E., Beuchat, L.R. 1986. Effect of temperature, sodium chloride and pH on growth of Listeria monocytogenes in cabbage juice. Applied and

Environmental Microbiology 52: 59-63

Coventry, M.J., Gordon, J.B., Wilcock, A., Harmark, K., Davidson, B.E., Hickey,

M.W., Hillier, A.J., Wan, J. 1997. Detection of bacteriocins of lactic acid bacteria isolated from foods and comparison with pediocin and nisin. Journal of Applied

Microbiology 83: 248-258

Cowan, M.M. 1999. Plant products as antimicrobial agents. Clinical Microbiological

Reviews 12: 564-582

222

Cox, J.M., Fleet, G.H. 2003. New directions in the microbiological analysis of foods. In

Hocking, A.D. (ed.), Foodborne microorganisms of public health significance, 6th

Edition. Southwood Press, Sydney: 103-161

Cox, N.A., Mercuri, A.J., Thomson, J.E., Chew, V. 1976. Swab and excised tissue sampling for total and Enterobacteriaceae counts of fresh and surface-frozen broiler skin. Poultry Science 55: 2405-2408

Croci, L., De Medici, D., Scalfaro, C., Fiore, A., Toti, L. 2002. The survival of hepatitis

A in fresh produce. International Journal of Food Microbiology 73: 29-34

Crosby, D.G. 1963. The organic constituents of food. I. Lettuce. Journal of Food

Science 28: 347-355

Crowdy, S.H. 1971. The control of leaf pathogens using conventional and systemic fungicides. In: Preece, T.F., Dickinson, C.H. (Eds.), Ecology of leaf surface microorganisms. Academic Press, London, 395-407

Cullison, M.A., Jaykus, L-A. 2002. Magnetized carbonyl iron and insoluble Zirconium

Hydroxide mixture facilitates bacterial concentration and separation from nonfat dry milk. Journal of Food Protection 65: 1806-1810

223

Davidson, P.M., Taylor, T.M. 2007. Chemical preservatives and natural antimicrobial compounds. In Doyle, M.P., Beuchat, L.R. (eds). Food Microbiology: Fundamentals and Frontiers 3rd edition. ASM Press, Washington DC: 713-745

Davidson, R., Proctor, P., Preston, M., Borzyk, A., Goldman, C., Harris, S., Bertolin, J.,

Thususka, J., Willey, B., Low, D.E., McGeer, A. 1996. Investigation of a lettuce-borne

Escherichia coli O157:H7 outbreak in a hospital. In Program and abstracts of the 36th interscience conference on Antimicrobial agents and chemotherapy Sept. 15-18, 1996,

New Orleans Ca. Abstract J106: 238

Davis, H., Taylor, J.P., Perdue, J.N., Stelma Jr, G.N., Humphreys Jr, J.M., Rowntree III,

R., Greene, K.D. 1988. A shigellosis outbreak traced to commercially distributed shredded lettuce. American Journal of Epidemiology 128: 1312-1321

Deans, S.G. 1991. Evaluation of antimicrobial activity of essential (volatile) oils. In

Linskens, H.F., Jackson, J.F. (eds). Essential oils and waxes. Springer-Verlag, Berlin:

309-320

Deans, S.G., Ritchie, G. 1987. Antibacterial properties of plant essential oils.

International Journal of Food Microbiology 5: 165-180

Delaquis, P. 2006. Fresh-cut vegetables. In Sapers, G.M., Gorny, J.R., Yousef, A.E.

(eds). Microbiology of Fruits and Vegetables. CRC Press, Boca Raton, FL: 253-265

224

Dent, K.C., Stephen, J.R., Finch-Savage, W.E. 2004. Molecular profiling of microbial communities associated with seeds of Beta vulgaris subsp. Vulgaris (sugar beet).

Journal of Microbiological Methods 56: 17-26

de Simon, M., Tarrago, C., Ferrer, M.D. 1992. Incidence of Listeria monocytogenes in fresh foods in Barcelona (Spain). International Journal of Food Microbiology 16: 153-

156

Desmarchelier, P. 1996. Foodborne disease: emerging problems and solutions. Medical

Journal of Australia 165: 668-671

De Vero, L., Gala, E., Gullo, M., Solieri, L., Landi, S., Giudici, P. 2006. Application of denaturing gradient gel electrophoresis (DGGE) analysis to evaluate acetic acid bacteria in traditional balsamic vinegar. Food Microbiology 23: 809-813

Dewettinck, T., Hulsbosch, W., van Hege, K., Top, E.M., Verstraete, W. 2001.

Molecular fingerprinting of bacterial populations in groundwater and bottled mineral water. Applied Microbiology and Biotechnology 57: 412-418

Dijkshoorn, L., van Aken, E., Shunburne, L., van der Reijden, T.J.K., Bernards, A.T.,

Nemec, A., Towner, K.J. 2005. Prevalence of Acinetobacter baumannii and other

Acinetobacter spp. in faecal samples from non-hospitalised individuals. Clinical and

Microbiological Infection 11: 329-332

225

do Socorro Rocha Bastos, M., de Fátima Ferreira Soares, N., de Andrade, N.J., Arruda,

A.C., Alves, R.E. 2005. The effect of the association of sanitizers and surfactant in the microbiota of the cantaloupe (Cucumis melo L.) melon surface. Food Control 16: 369-

373

Dowe, M.J., Jackson, E.D., Mori, J.G., Bell, C.R. 1997. Listeria monocytogenes survival in soil and incidence in agricultural soils. Journal of Food Protection 60: 1201-

1207

Doyle, M.P. 1990. Fruit and vegetable safety- microbiological considerations.

Hortscience 25: 1478-1482

Duffy, E.A., Lucia, L.M., Kells, J.M., Castillo, A., Pillai, S.D., Acuff, G.R. 2005.

Concentrations of Escherichia coli and genetic diversity and antibiotic resistance profiling of Salmonella isolated from irrigation water, packing shed equipment, and fresh produce in Texas. Journal of Food Protection 68: 70-79

Duffy, G. 2003. Verocytoxigenic Escherichia coli in animal faeces, manures and slurries. Journal of Applied Microbiology 94: 94S-103S

Egert, M., Friedrich, M.W. 2003. Formation of pseudo-terminal restriction fragments, a

PCR-related bias affecting terminal restriction fragment length polymorphism analysis of microbial community structure. Applied and Environmental Microbiology 69: 2555-

2562

226

Eginton, P.J., Holah, J., Allison, D.G., Handley, P.S., Gilbert, P. 1998. Changes in the strength of attachment of microorganism to surfaces following treatment with disinfectants and cleansing agents. Letters in Applied Microbiology 27: 101-105

El-Baradei, G., Delacroix-Buchet, A., Ogier, J-C. 2007. Biodiversity of bacterial ecosystems in traditional Egyptian domiati cheese. Applied and Environmental

Microbiology 73: 1248-1255

Eloff, J.N. 1998. Which extractant should be used for the screening and isolation of antimicrobial components from plants? Journal of Ethnopharmocology 60: 1-8

Ercolani, G.L. 1976. Bacteriological quality assessment of fresh marketed lettuce and fennel. Applied and Environmental Microbiology 31: 847-852

Ercolini, D. 2004. PCR-DGGE fingerprinting: novel strategies for detection of microbes in food. Journal of Microbiological Methods 56: 297-314

Ercolini, D., Hill, P.J., Hodd, C.E.R. 2003a. Bacterial community structure and location in Stilton chese. Applied and Environmental Microbiology 69: 3540-3548

Ercolini, D., Mauriello, G., Blaiotta, G., Moschetti, G., Coppola, S. 2003b. PCR-DGGE fingerprints of microbial succession during a manufacture of traditional water buffalo mozzarella cheese. Journal of Applied Microbiology 96: 263-270

227

ES ISO 6887-1:2001. Microbiology of foods and animal feeding stuffs- Preparation of test samples, initial suspension and decimal dilutions for microbiological examination,

Part 1: General rules for the preparation of the initial suspension and decimal dilutions

Estrada, P., Mavingui, P., Cournoyer, B., Fontaine, F., Balandreau, J., Caballero-

Mellado, J. 2002. A N2-fixing endophytic Burkholderia sp. associated with maize plants cultivated in Mexico. Canadian Journal of Microbiology 48: 285-294

EU (European Commission). 2007. Maximum Residue Limits (MRLs) database.

Available at http://europa.eu.int/comm/food/plant/protection/pesticides/index_en.htm.

Accessed 21 July, 2007.

Fain, A.R. 1996. A review of the microbiological safety of fresh salads. Dairy Food and

Environmental Sanitation 16: 146-149

Falcão, D.P, Valentini, S.R., Leite, C.Q.F. 1993. Pathogenic or potentially pathogenic bacteria as contaminants of fresh water from different source in Araraquara, Brazil.

Water Research 27: 1737-1741

Fasoli, S., Marzotto, M., Rizzotti, L., Rossi, F., Dellaglio, F., Torriani, S. 2003.

Bacterial composition of commercial probiotic products as evaluated by PCR-DGGE analysis. International Journal of Food Microbiology 82: 59-70

228

Fenlon, D.R. 1981. Seagulls (Larus spp.) as vectors of salmonellae: an investigation into the range of serotypes and numbers of salmonellae in gull faeces. Journal of

Hygiene Cambridge 86: 195-202

Fenlon, D.R. 1983. A comparison of salmonella serotypes found in the faeces of gulls feeding at a sewage works with serotypes present in the sewage. Journal of Hygiene

Cambridge 91: 47-52

Fenlon, D.R. 1985. Wild birds and silage as reservoirs of Listeria in the agricultural environment. Journal of Applied Bacteriology 59: 537-543

Filichkin, S.A., Gelvin, S.B. 1992. Effect of dimethyl sulfoxide concentration on specificity of primer matching in PCR. Biotechniques 12: 828, 830

Fleet, G.H. 1999. Microorganisms in food ecosystems. International Journal of Food

Microbiology 50: 101-117

Francis, G.A., Thomas, C., O’Beirne, D. 1999. The microbiological safety of minimally processed vegetables. International Journal of Food Science and Technology 34: 1-22

Frank, J.F. 2001. Microbial attachment to food and food contact surfaces. Advances in

Food and Nutrition Research 43: 319-370

229

Freire, J.R., Robbs, C.F. 2000. Isolation and identification of pathogenic bacteria in minimally processed hydroponic lettuce. Alimentaria 37: 55-60

Frost, J.A., McEvoy, B. Bentley, C.A., Andersson, Y., Rowe, A. 1995. An outbreak of

Shigella sonnei infection associated with consumption of iceberg lettuce. Emerging

Infectious Diseases 1: 26-29

FSANZ (Food Standards Australia New Zealand). 2007. Food Standards Code, Section

1.4.2, Maximum Residue Limits (Australia). Available at: http://www.foodstandards.gov.au/foodstandardscode/index.cfm#_FSCchapter1.

Accessed 22 July, 2007.

Gallois, A., Samson, R., Ageron, E., Grimont, P.A.D. 1992. Erwinia carotovora subsp. nov. associated with odorous soft rot of chicory (Cichorium intybus L.). International

Journal of Systematic Bacteriology 42: 582-588

Garbeva, P., van Overbeek, L.S., van Vuurde, J.W.L., van Elsas, J.D. 2001. Analysis of endophytic bacterial communities of potato by plating and denaturing gradient gel electrophoresis (DGGE) of 16S rDNA based PCR fragments. Microbial Ecology 41:

369-383

Garcia-Gimeno, R.M., Zurera-Cosano, G. 1997. Determination of ready-to-eat vegetable shelf-life. International Journal of Food Microbiology 36: 31-38

230

Garcia-Gimeno, R.M., Zurera-Cosano, G., Amaro-Lopez, M. 1996. Incidence, survival and growth of Listeria monocytogenes in ready-to-use mixed vegetable salads in Spain.

Journal of Food Safety 16: 75-86

Garcia-Villanova Ruiz, B., Cueto Espinar, A., Bolanos Carmona, M.J. 1987. A comparative study of strains of salmonella isolated from irrigation waters, vegetables and human infections. Epidemiology and Infection 98: 271-276

Garg, N. Churey, J.J., Splittstoesser, D.F. 1990. Effect of processing conditions on the microflora of fresh-cut vegetables. Journal of Food Protection 53: 701-703

Garrett, E.H., Gorny, J.R., Beuchat, L.R., Farber, J.N., Harris, L.J., Parish, M.E.,

Suslow, T.V., Busta, F.F. 2003. Microbiological safety of fresh and fresh-cut produce: description of the situation and economic impact. Comprehensive Reviews in Food

Science and Food Safety 2(Supplement): 13-37

Gavini, F., Mergaert, J., Beji, A., Mielcarek, C., Izard, D., Kersters, K., De Ley, J. 1989.

Transfer of Enterobacter agglomerans (Beijerinck 1888) Ewing and Fife 1972 to

Pantoea gen. nov. as Panteoa agglomerans comb. nov. and description of Pantoea dispersa sp. nov. International Journal of Systematic Bacteriology 39: 337-345

Geldreich, E.E., Bordner, R.H. 1971. Fecal contamination of fruits and vegetables during cultivation and processing for market. A review. Journal of Milk and Food

Technology 34: 184-195

231

Gennari, M., Lombardi, P. 1993. Comparative characterization of Acinetobacter strains isolated from different foods and clinical sources. Zentralblatt Bakterologie 279: 553-

564

Gennari, M., Parini, M., Volpon, D., Serio, M. 1992. Isolation and characterization by conventional methods and genetic transformation of Psychrobacter and Acinetobacter from fresh and spoiled meat, milk and cheese. International Journal of Food

Microbiology 15: 61-75

Giraffa, G., 2004. Studying the dynamics of microbial populations during food fermentation. FEMS Microbiology Reviews 28: 251-260

Giraffa, G., Neviani, E. 2001. DNA-based, culture-independent strategies for evaluating microbial communities in food-associated ecosystems. International Journal of Food

Microbiology 67: 19-34

Goldsmid, J.M., Speare, R. 1997. The parasitology of foods. In: Foodborne microorganisms of public health significance, 5th Edition, Hocking, A.D., Arnold, G.,

Jenson, I., Newton, K., Sutherland, P. (eds.). Trenear Printing: pp. 583-602

Gombas, D.E., Chen, Y., Clavero, R.S., Scott, V.N. 2003. Survey of Listeria monocytogenes in ready-to-eat foods. Journal of Food Protection 66: 559-569

232

Gómez, R., Muñoz, M., de Ancos, B., Cano, M.P. 2002. New procedure for the detection of lactic acid bacteria in vegetables producing antibacterial substances.

Lebesmittel-Wissenschaft und Technologie 35: 284-288

González, C.F., Pettit, E.A., Valadez, V.A., Provin, E.A. 1997. Mobilisation, cloning, and sequence determination of a plasmid encoded polygalacturonase from a phytopathogenic Burkholderia (Pseudomonas) cepacia. Molecular and Plant Microbe

Interactions 10: 840-851

Gorny, J.R. 2006. Microbial contamination of fresh fruits and vegetables. In Sapers,

G.M., Gorny, J.R., Yousef, A.E. (eds). Microbiology of Fruits and Vegetables. CRC

Press, Boca Raton, FL: 3-32

Gras, M-H., Druet-Michaud, C., Cerf, O. 1994. La flore bacterienne des feuilles de salade fraiche. Sciences des Aliments. 14: 173-188

Grehn, M., von Graevenitz, A. 1978. Search for Acinetobacter calcoaceticus subsp. anitratus by enrichment of fecal samples. Journal of Clinical Microbiology 8: 342-343

Greig, J.D., Todd, E.C.D., Bartleson, C.A., Michaels, B.S. 2007. Outbreaks where food workers have been implicated in the spread of foodborne disease. Part 1. Description of the problem, methods, and agents involved. Journal of Food Protection 70: 1752-1761

233

Griffiths, R.I., Whiteley, A.S., O’Donnell, A.G., Bailey, M.J. 2000. Rapid method for coextraction of DNA and RNA from natural environments for analysis of ribosomal

DNA- and rRNA-based microbial community composition. Applied and Environmental

Microbiology 66: 5488-5491

Grimont, P.A.D., Grimont, F., Starr, M.P. 1981. Serratia species isolated from plants.

Current Microbiology 5: 317-322

Grohmann, G.S. 1997. Viruses, food and environment. In Hocking, A.D., Arnold, G.,

Jenson, I., Newton, K., Sutherland, P. (eds). Foodborne Microorganism of Public Health

Significance. Trenear Printing, Australia: 603-620

Gromek, D., Kisiel, W., Klodzinska, A., Chojnacka-Woejcik, E. 1992. Biologically active preparations from Lactuca virosa L. Phytotherapy Research 6: 285-287

Gross, D.C., Cody, Y.S. 1985. Mechanisms of plant pathogensis by Pseudomonas species. Canadian Journal of Microbiology 31: 403-410

Guan, T.T.Y., Blank, G., Holley, R.A. 2005. Survival of pathogenic bacteria in pesticide solutions and on treated tomato plants. Journal of Food Protection 68: 292-304

Guan, T.Y., Blank, G., Ismond, A., Van Acker, R. 2001. Fate of foodborne bacterial pathogens in pesticide products. Journal of the Science, Food and Agriculture 81: 503-

512

234

Guardabassi, L., Dalsgaard, A., Olsen, J.E. 1999. Phenotypic characterisation and antibiotic resistance of Acinetobacter spp. isolated from aquatic sources. Journal of

Applied Microbiology 87: 659-667

Guo, X., Chen, J., Brackett, R.E., Beuchat, L.R. 2002. Survival of Salmonella on tomatoes stored at high relative humidity, in soil, and on tomatoes in contact with soil.

Journal of Food Protection 65: 274-279

Hamilton-Miller, J.M.T., Shah, S. 2001. Identity and antibiotic susceptibility of enterobacterial flora of salad vegetables. International Journal of Antimicrobial Agents

18: 81-83

Han, D.C., Kyung, K.H. 1995. Antimicrobial activity of autoclaved cabbage juice.

Korean Journal of Food Science and Technology 27: 74-79

Han, Y., Linton, R.H., Nelson, P.E. 2004. Effects of recovery, plating, and inoculation methods on quantification of Escherichia coli O157:H7 and Listeria monocytogenes from strawberries. Journal of Food Protection 67: 2436-2442

Handschur, M., Pinar, G., Gallist, B., Lubitz, W., Haslberger, A.G. 2005. Culture free

DGGE and cloning based monitoring of changes in bacterial communities of salad due to processing. Food and Chemical Toxicology 43: 1595-1605

235

Hanlon, G.W. 2005. The emergence of multidrug resistant Acinetobacter species: a major concern in the hospital setting. Letters in Applied Microbiology 41: 375-378

Hansen, M.C., Tolker-Nielsen, T., Givskov, M., Molin, S. 1998. Biased 16S rDNA PCR amplification caused by interference from DNA flanking the template region. FEMS

Microbiology Ecology 26: 141-149

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(Supplement): 78-141

Harvey, J.M. 1978. Reduction of losses in fresh market fruits and vegetables. Annual

Reviews in Phytopathology 16: 321

Hauben, L., Moore, E.R.B., Vauterin, L., Steenackers, M., Mergaert, J., Verdonck, K.,

Swings, J. 1998. Phylogenetic position of phytopathogens within the

Enterobacteriaceae. Systematic and Applied Microbiology 21: 384-397

Heard, G.M. 1999. Microbial safety of ready-to-eat salads and minimally processed vegetables and fruits. Food Australia 51: 414-420

Heard, G.M. 2002. Microbiology of fresh-cut produce. In: Lamikanra, O. (Ed.), Fresh- cut fruits and vegetables. CRC Press, Boca Raton, Florida, 187-248

236

Heisich, J.E., Wagner, D.E., Nierman, M.L., Peeler, J.T. 1989. Listeria spp. found on fresh market produce. Applied and Environmental Microbiology 55: 1925-1927

Henke, W., Herdel K., Jung, K., Schnorr, D., Loening, S.A. 1997. Betaine improves the

PCR amplification of GC-rich DNA sequences. Nucleic Acids Research 25: 3957-3958

Hernández, F., Monge, R., Jiménez, C., Taylor, L. 1997. Rotavirus and Hepatitis A virus in market lettuce (Lactuca sativa) in Costa Rica. International Journal of Food

Microbiology 37: 221-223

Herwaldt, B.L. 2000. Cyclospora cayetanensis: a review, focusing on outbreaks of cyclosporiasis in the 1990s. Clinical Infectious Diseases 31: 1040-1057

Herwaldt, B.L., Ackers, M-A. 1997. An outbreak in 1996 of cyclosporiasis associated with imported raspberries. New England Journal of Medicine 336: 1548-1556

Heuer, H., Smalla, K. 1997. Application of denaturing gradient gel electrophoresis and temperature gradient gel electrophoresis for studying soil microbial communities. In van

Elsas, J.P., Trevors, J.T., Wellington, E.M.H. (eds). Modern Soil Microbiology. Marcel

Dekker, New York: 353-373

Hilborn, E.D., Mermin, J.H., Mshar, P.A., Hadler, J.L., Voetsch, A., Wojtkunski, C.,

Swartz, M., Mshar, R., Lambert-Fair, M-A., Farrar, J.A., Glynn, M.K., Slutsker, L.

237

1999. A multistate outbreak of Escherichia coli O157:H7 infections associated with consumption of mesclun lettuce. Archives of Internal Microbiology 159: 1758-1764

Himathongkham, S., Bahari, S., Riemann, H., Cliver, D. 1999. Survival of Escherichia coli O157:H7 and Salmonella typhimurium in cow manure and cow manure slurry.

FEMS Microbiology Letters 178: 251-257

Hislop, E.C. 1976. Some effects of fungicides and other agrochemicals on the microbiology of the aerial surfaces of plants. In: Dickinson, C.H., Preece, T.F. (Eds.),

Microbiology of aerial plant surfaces. Academic Press, London, 41-74

Ho, J.L., Shands, K.N., Friedland, G., Eckind, P., Fraser, D.W. 1986. An outbreak of

Type 4b Listeria monocytogenes infection involving patients from Eight Boston hospitals. Archives for Internal Medicine 146: 520-524

Hochberg, E.G. 1996. The market for agricultural pesticide inert ingredients. In: Foy,

C.L., Pritchard, D.W. (Eds.), Pesticide formulation and adjuvant technology. CRC

Press, Boca Raton, Florida, 203-208

Holbrook, R. 2000. Detection of microorganisms in foods- principles of culture methods. In Lund, B.M., Baird-Parker, T.C., Gould, C.W. (eds). The microbiological safety and quality of food. Aspen Publishers, Gaithersburg, Maryland: 1761-1790

238

Holton, J. 1983. A note on the preparation and use of a selective differential medium for the isolation of Acinetobacter spp. from clinical sources. Journal of Applied

Bacteriology 54: 141-142

Horby, P.W., O’Brien, S.J., Adak, G.K., Graham, C., Hawker, J.I., Hunter, P., Lane, C.,

Lawson, A.J., Mitchell, R.T., Reacher, M.H., Threlfall, E.J., Ward, L.R. 2003. A national outbreak of multi-resistant Salmonella enterica serovar Typhimurium definitive phage type (DT) 104 associated with consumption of lettuce. Epidemiology and

Infection 130: 169-178

Hornitzky, M.A., Bettelheim, K.A., Djordjevic, S.P. 2000. The isolation of enterohaemorrhagic Escherichia coli O111:H- from Australian cattle. Australian

Veterinary Journal 78: 636-637

Houang, E.T.S., Chu, Y.W., Leung, C.M., Chu, K.Y., Berlau, J., Ng, K.C., Cheng,

A.F.B. 2001. Epidemiology and infection control implications of Acinetobacter spp. in

Hong Kong. Journal of Clinical Microbiology 39: 228-234

Hsieh, T.F., Huang, H.C., Erickson, R.S. 2005. Biological control of bacterial wilt of bean using endophyte, Pantoea agglomerans. Journal of Phytopathology 153: 606-614

Huang, Y., Deverall, B.J., Morris, S.C. 1993. Effect of Pseudomonas cepacia on postharvest biocontrol of infection by Penicillium digitatum and on wound responses of citrus fruit. Australasian Plant Pathology 22: 84-93

239

Hutchison, M.L., Walters, L.D., Moore, A., Crookes, K.M., Avery, S.M. 2004. Effect of length of time before incorporation on survival of pathogenic bacteria present in livestock wastes applied to agricultural soil. Applied and Environmental Microbiology

70: 5111-5118

Hutchinson, M.L., Walters, L.D., Moore, T., Thomas, D.J.I., Avery, S.M. 2005. Fate of pathogens present in livestock wastes spread onto fescue plots. Applied and

Environmental Microbiology 71: 691-696

Hwang, C-A., Beuchat, L.R. 1995. Efficacy of selected chemicals for killing pathogenic and spoilage microorganisms on chicken skin. Journal of Food Protection 58: 19-23

ICMSF. 1998. Vegetables and vegetable products. In Microbial Ecology of Food

Commodities. London: Blackie Academic and Professional: 215-251

Ingham, S.C., Losinski, J.A., Andrews, M.P., Breuer, J.E., Breuer, J.R., Wood, T.M.,

Wright, T.H. 2004. Escherichia coli contamination of vegetables grown in soils fertilised with noncomposted bovine manure: garden scale studies. Applied and

Environmental Microbiology 70: 6420-6427

Institute of Food Technologists (IFT). 2003. Analysis and evaluation of preventative control measures for the control and reduction/elimination of microbial hazards on fresh

240

and fresh-cut produce. Comprehensive Reviews in Food Science and Food Safety

2(Supplement): 1-204

Iwasa, M., Makino, S-I., Asakura, H., Kobori, H., Morimoto, Y. 1999. Detection of

Escherichia coli O157:H7 from Musca domestica at a cattle farm in Japan. Journal of

Medical Entomology 36: 108-112

Jabar, M.A., Al-Mossawi, A. 2007. Susceptibility of some multiple resistant bacteria to garlic extract. African Journal of Biotechnology 6: 771-776

Jackson, S.G., Goodbrand, R.B., Johnson, R.P., Odorico, V.G., Alves, D., Rahn, K.,

Wilson, J.B., Welch, M.K., Khakhria, R. 1998. Escherichia coli O157:H7 diarrhoea associated with well water and infected cattle on an Ontario farm. Epidemiology and

Infection 120: 17-20

Jacobs, D., Angles, M.L., Goodman, A.E., Neilan, B.A. 1997. Improved methods for in situ enzymatic amplification and detection of low copy number genes in bacteria. FEMS

Microbiological Letters 132: 65-73

Jacques, M-A., Morris, C.E. 1995. Bacterial population dynamics and decay on leaves of different ages of ready-to-use broad-leaved endive. International Journal of Food

Microbiology 30: 221-236

241

Janisiewicz, W.J., Conway, W.S., Brown, M.W., Sapers, G.M., Fratamico, P.,

Buchanan, R.L. 1999. Fate of Escherichia coli O157:H7 on fresh-cut apple tissue and its potential for transmission by fruit flies. Applied and Environmental Microbiology

65: 1-5

Jawad, A., Hawkey, P.M., Heritage, J., Snelling, A.M. 1994. Description of Leeds

Acinetobacter medium, a new selective and differential medium for isolation of clinically important Acinetobacter spp., and comparison with Herellea agar and

Holton’s agar. Journal of Clinical Microbiology 32: 2353-2358

Jay, J.M., Margitic, S. 1979. Comparison of homogenising, shaking and blending on the recovery of microorganisms and endotoxins from fresh and frozen ground beef as assessed by plate counts and the Limulus amoebocyte lysate test. Applied and

Environmental Microbiology 38: 879-884

Jiang, X., Morgan, J., Doyle, M.P. 2002. Fate of Escherichia coli O157:H7 in manure- amended soil. Applied and Environmental Microbiology 68: 2605-2609

Jiwa, S.F.H., Krovacek, K., Wadström, T. 1981. Enterotoxigenic bacteria in food and water from an Ethiopian community. Applied and Environmental Microbiology 41:

1010-1019

Jobling, J.J., Richardson, K.C., Patterson, B.D. 1998. Freshness in convenience fruit and vegetables. Food Australia 50: 443-446

242

Johannessen, G.S., Bengtsson, G.B., Heier, B.T, Bredholt, S., Wasteson, Y., Rørvik,

L.M. 2005. Potential uptake of Escherichia coli O157:H7 from organic manure into crisphead lettuce. Applied and Environmental Microbiology 71: 2221-2225

Johannessen, G.S., Frøseth, R.B., Solemdal, L., Jarp, J., Wasteson, Y., Rørvik, L.M.

2004. Influence of bovine manure as fertiliser on the bacteriological quality of organic iceberg lettuce. Journal of Applied Microbiology 96: 787-794

Johannessen, G.S., Loncarevic, S., Kruse, H. 2002. Bacteriological analysis of fresh produce in Norway. International Journal of Food Microbiology 77: 199-204

Johnston, L.M., Elhanafi, D., Drake, M., Jaykus, L-A. 2005. A simple method for the direct detection of Salmonella and Escherichia coli O157:H7 from raw alfalfa sprouts and spent irrigation water using PCR. Journal of Food Protection 68: 2256-2263

Johnston, L.M., Jaykus, L-A., Moll, D., Anciso, J., Mora, B., Moe, C.L. 2006. A field study of the microbiological quality of fresh produce of domestic and Mexican origin.

International Journal of Food Microbiology 112:83-95

Jones, F., Smith, P., Watson, D.C. 1978. Pollution of a water supply catchment by breeding gulls and the potential environmental health implication. Journal of the

Institute for Water Engineering Science 32: 469-482

243

Juni, E. 1972. Interspecies transformation of Acinetobacter: genetic evidence for a ubiquitous genus. Journal of Bacteriology 112: 917-931

Kämpfer, P. 2000. Acinetobacter. In Encyclopedia of Food Microbiology, ed. Robinson,

R.K., Batt, A, Patel, P.D. pp 7-15. Academic Press, London

Kapperud, G., Rorvik, L.M., Hasseltvedt, V., Hoiby, E.A., Iverson, B.G., Staveland, K.,

Johnsen, G., Leitao, J., Herikstad, H., Andersson, Y., Langeland, G., Gondrosen, B.,

Lassen, J. 1995. Outbreak of Shigella sonnei infection traced to imported iceberg lettuce. Journal of Clinical Microbiology 33: 609-614

Kapperud, G., Rosef, O. 1983. Avian reservoir of Campylobacter fetus subsp. jejuni,

Yersinia spp., and Salmonella spp. in Norway. Applied and Environmental

Microbiology 45: 375-380

Karlin, S., Altschul, F. 1990. Methods for assessing the statistical significance of molecular sequencing features by using general scoring schemes. Proceedings of the

National Academy of Science USA 87: 2264-2268

Katzenelson, E., Teltch, B. 1976. Dispersion of enteric bacteria by spray irrigation.

Journal of the Water Control Federation 48: 710-716

244

Katzenelson, E., Teltch, B., Shuval, H.I. 1977. Spray irrigation with wastewater: the problem of aerosolisation and dispersion of enteric microorganisms. Progress in Water

Technology 9: 1-11

Kedersha, N.L., Rome, L.H. 1986. Preparative agarose gel electrophoresis for the purification of small organelles and particles. Analytical Biochemistry 156: 161-170

Kennedy, B.W., Alcorn, S.M. 1980. Estimates of U.S. crop losses to prokaryote plant pathogens. Plant Disease 64: 674-676

Kenney, S.J., Anderson, G.L., Williams, P.L., Millner, P.D., Beuchat, L.R. 2006.

Migration of Caenorhabditis elegans to manure and manure compost and potential for transport of Salmonella newport to fruits and vegetables. International Journal of Food

Microbiology 106: 61-68

Kersters, K., Ludwig, W., Vancanneyt, M., De Vos, P., Gillis, M., Schleifer, K-H. 1996.

Recent changes in the classification of the Pseudomonads: an overview. Systematic and

Applied Microbiology 19: 465-477

Khan, M.R., Saha, M.L., Kibria, A.H.M.G. 1992. A bacteriological profile of salad vegetables in Bangladesh with special reference to coliforms. Letters in Applied

Microbiology 14: 88-90

245

Kim, J-G., Yousef, A.E., Chism, G.W. 1999. Use of ozone to inactivate microorganism on lettuce. Journal of Food Safety 19: 17-34

Kim, M-K., Kim, M-J., Shin, D-H., Song, C-G., Lee, H-S. 2001. Growth inhibiting effects of vegetable extracts on beneficial and harmful intestinal bacteria. Agricultural

Chemistry and Biotechnology 44: 65-70

King, Jr, A.D., Bolin, H.R. 1989. Physiological and microbiological storage stability of minimally processed fruits and vegetables. Food Technology 43: 132-139

King Jr, A.D., Magnusson, J.A., Török, T., Goodman, N. 1991. Microbial flora and storage quality of partially processed lettuce. Journal of Food Science 56: 459-461

Kisand, V., Wikner, Z. 2003. Limitied resolution of 16S rDNA DGGE caused by melting properties and closely related DNA sequences. Journal of Microbiological

Methods 54: 183-191

Klein, B.P. 1987. Nutritional consequences of minimal processing of fruits and vegetables. Journal of Food Quality 10: 179-193

Koek, P.C., de Witte, Y., de Maaker, J. 1983. The microbial ecology of prepared raw vegetables. In Roberts, T.A., Skinner, F.A. (eds). Food Microbiology: advances and prospects. Academic Press, London: 231-240

246

Korhonen, L.K., Martikainen, P.J. 1991. Survival of Escherichia coli and

Campylobacter jejuni in untreated and filtered lake water. Journal of Applied

Bacteriology 71: 379-382

Koseki, S., Isobe, S., Itoh, K. 2004. Efficacy of acidic electrolysed water ice for pathogen control on lettuce. Journal of Food Protection 67: 2544-2549

Koseki, S., Yoshida, K., Isobe, S., Itoh, K. 2001. Decontamination of lettuce using acidic electrolysed water. Journal of Food Protection 64: 652-658

Kowalchuk, G.A., Stephen, J.R., De Boer, W., Prosser, J.I., Embley, T.M., Woldendorp,

J.W. 1997. Analysis of ammonia-oxidizing bacteria of the β subdivision of the class

Proteobacteria in coastal sand dunes by denaturing gradient gel electrophoresis and sequencing of PCR-amplified 16S ribosomal DNA fragments. Applied and

Environmental Microbiology 63: 1489-1497

Krechel, A., Faupel, A., Hallmann, J., Ulrich, A., Berg, G. 2002. Potato-associated bacteria and their antagonistic potential towards plant-pathogenic fungi and the plant- parasitic nematode Meloidogyne incognita (Kofoid & White) Chitwood. Canadian

Journal of Microbiology 48: 772-786

Kudva, I.T., Blanch, K., Hovde, C.J. 1998. Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Applied and Environmental

Microbiology 64: 3166-3174

247

Kurdziel, A.S., Wilkinson, N., Langton, S., Cook, N. 2001. Survival of poliovirus on soft fruit and salad vegetables. Journal of Food Protection 64: 706-709

Kyung, K.H., Fleming, H.P. 1994. Antibacterial activity of cabbage juice against lactic acid bacteria. Journal of Food Science 59: 125-129

Lacava, P.T., Andreote, F.D., Araújo, W.L., Azevedo, J.L. 2006. Characterisation of the endophytic bacterial community from citrus by isolation, specific PCR and DGGE.

Pesquisa Agropecuaria Brasileira 41: 637-642

Lantz, P-G., Hahn-Hägerdal, B., Rådström, P. 1994. Sample preparation methods in

PCR-based detection of food pathogens. Trends in Food Science and Technology 5:

384-389

Lautrop, H. 1974. Genus IV. Acinetobacter Brisou and Prevot 1954. 727. In Bergey’s

Manual of Determinative Bacteriology ed Buchanan R.E. and Gibbons, N.E. pp. 436-

438. Baltimore, MD: Williams and Wilkins Co.

LeChevallier, M.W., Seidler, R.J., Evans, T.M. 1980. Enumeration and characterisation of standard plate count bacteria in chlorinated and raw water supplies. Applied and

Environmental Microbiology 40: 922-930

248

Lee, J-S., Heo, G-Y., Lee, J.W., Oh, Y-J., Park, J.A., Park, Y-H., Pyun, Y-R., Ahn, J.S.

2005. Analysis of kimchi microflora using denaturing gradient gel electrophoresis.

International Journal of Food Microbiology 102: 143-150

Lee, Y-L., Cesario, T., Wang, Y., Shanbrom, E., Thrupp, L. 2003. Antibacterial activity of vegetables and juices. Nutrition 19: 994-996

Leff, L.G., Dana, J.R., McArthur, J.V., Shimkets, L.J. 1995. Comparison of methods of

DNA extraction from stream sediments. Applied and Environmental Microbiology 61:

1141-1143

Li, M.Y., Zhou, G.H., Xu, X.L., Li, C.B., Zhu, W.Y. 2006. Changes of bacterial diversity and main flora in chilled pork during storage using PCR-DGGE. Food

Microbiology 23: 607-611

Liao, C-H. 2006. Bacterial soft rot. In Sapers, G.M., Gorny, J.R., Yousef, A.E. (eds),

Microbiology of Fruits and Vegetables. CRC Press, Boca Raton, FL: 117-134

Liao, C-H., Fett, W.F. 2001. Analysis of native microflora and selection of strains antagonistic to human pathogens on fresh produce. Journal of Food Protection 64: 1110-

1115

249

Liao, C.H., McCallus, D.E., Wells, J.M. 1993. Calcium-dependent pectate lysase production in the soft rotting bacterium Pseudomonas fluorescens. Phytopathology 83:

813-818

Liao, C.H., Sapers, G.M. 1999. Influence of soft rot bacteria on growth of Listeria monocytogenes on potato tuber slices. Journal of Food Protection 62: 343

Liao, C.H., Wells, J.M. 1987. Diversity of pectolytic, fluorescent pseudomonads causing soft rots of fresh vegetables at produce markets. Phytopathology 77: 673-677

Liesack, W., Stackebrandt, E. 1992. Occurrence of novel groups of the domain Bacteria as revealed by analysis of genetic material isolated from an Australian terrestrial environment. Journal of Bacteriology 174: 5072-5078

Liesack, W., Weyland, H., Stackebrandt, E. 1991. Potential risks of gene amplification by PCR as determined by 16S rDNA analysis of a mixed culture of strict barophilic bacteria. Microbial Ecology 21: 191-198

Lillard, H.S. 1988a. Effect of surfactant or changes in ionic strength on the attachment of Salmonella typhimurium to poultry skin and muscle. Journal of Food Science 53:

727-730

250

Lillard, H.S. 1988b. Comparison of sampling methods and implications for bacterial decontamination of poultry carcasses by rinsing. Journal of Food Protection 51: 405-

408

Lin, C-M., Fernando, S.Y., Wei, C-I. 1996. Occurrence of Listeria monocytogenes,

Salmonella spp., Escherichia coli and Escherichia coli O157:H7 in vegetable salads.

Food Control 7: 135-140

Lin, C-M., Kim, J., Du, W-X., Wei, C-I. 2000. Bactericidal activity of isothiocyanate against pathogens on fresh product. Journal of Food Protection 63: 25-30

Little, C. Roberts, D., Youngs, E., de Louvois, J. 1999. Microbiological quality of retail imported unprepared whole lettuce: A PHLS Food Working Group Study. Journal of

Food Protection 62: 325-328

Loncarevic, S., Johannessen, G.S., Rørvik, L.M. 2005. Bacteriological quality of organically grown leaf lettuce in Norway. Letters in Applied Microbiology 41: 186-189

Lopez, I., Ruiz-Larrea, F., Cocolin, L., Orr, E., Phister, T., Marshall, M.,

VanderGheynst, J., Mills, D.A. 2003. Design and evaluation of PCR primers for analysis of bacterial populations in wine by denaturing gradient gel electrophoresis.

Applied and Environmental Microbiology 69: 6801-6807

251

Lucien, F.P., Foster, N.R. 2000. Solubilities of solid mixtures in supercritical carbon dioxide: a review. Journal of Supercritical Fluids 17: 111-134

Lucier, G. 1998. Leafy greens: foundation of the vegetable industry. Agricultural

Outlook 248: 5-8

Lucore, L.A., Cullison, M.A., Jaykus, L-A. 2000. Immobilisation with metal hydroxides as a means to concentrate food-borne bacteria for detection by cultural and molecular methods. Applied and Environmental Microbiology 66: 1769-1776

Luechtefeld, N.A.W., Blaser, M.J., Reller, L.B., Wang, W-L.L. 1980. Isolation of

Campylobacter fetus subsp. jejuni from migratory waterfowl. Journal of Clinical

Microbiology 12: 406-408

Lueders, T., Friedrich, M.W. 2003. Evaluation of PCR amplification bias by terminal restriction fragment length polymorphism analysis of small-subunit rRNA and mcrA genes by using defined template mixtures of methanogenic pure cultures and soil DNA extracts. Applied and Environmental Microbiology 69: 320-326

Lund, B.M. 1983. Bacterial Spoilage. In Dennis, C. (ed). Postharvest Pathology of

Fruits and Vegetables. Academic Press, London: 219-257

Lund, B.M. 1992. Ecosystems in vegetable foods. Journal of Applied Bacteriology

Symposium Supplement 73: 115S-126S

252

Lund, B.M. 1993. The microbiological safety of prepared salad vegetables. Food

Technology International 1993: 196-200

MacGowan, A.P., Bowker, K., McLauchlin, J., Bennett, P.M., Reeves, D.S. 1994. The occurrence and seasonal changes in the isolation of Listeria spp. in shop bought food stuffs, human faeces, sewage and soil from urban sources. International Journal of Food

Microbiology 21: 325-334

Madden, J.M. 1992. Microbial pathogens in fresh produce – the regulatory perspective.

Journal of Food Protection 55: 821-823

Magnuson, J.A., King, Jr, A.D., Torok, T. 1990. Microflora of partially processed lettuce. Applied and Environmental Microbiology 56: 3851-3854

Mahmoud, Z.F., Kassem, F.F., Abdel-Salam, N.A., Zdero, C. 1986. Sesquiterpene lactones from Lactuca sativa. Phytochemistry 25: 747-748

Maidak, B.L., Cole, J.R., Parker Jr, C.T., Garrity, G.M., Larsen, N., Li, B., Lilburn,

T.G., McCaughey, M.J>, Olsen, G.J., Overbeek, R., Pramanik, S., Schmidt, T.M.,

Tiedje, J.M., Woese, C.R. 1999. A new version of the RDP (Ribosomal Database

Project). Nucleic Acids Research 27: 171-173

253

Mandel, A.D., Wright, K., McKinnon, J.M. 1964. Selective medium for isolation of

Mima and Herellea organisms. Journal of Bacteriology 88: 1524-1525

Marchetti, R., Cassedei, M.A., Guerzoni, M.E. 1992. Microbial population dynamics in ready-to-use vegetable salads. Italian Journal of Food Science 2: 97-108

Marongiu, B., Piras, A., Pani, F., Porcedda, S., Ballero, M. 2003. Extraction, separation and isolation of essential oils from natural matrices by supercritical CO2. Flavour and

Fragrance Journal 18: 505-509

Martin, D.L., Gustafson, T.L., Pelosi, J.W., Suarez, L., Pierce, G.V. 1986.

Contaminated produce- a common source for two outbreaks of Shigella gastroenteritis.

American Journal of Epidemiology 124: 299-305

Maszenan, A.M., Seviour, R.J., Patel, B.K.C., Wanner, J. 2000. A fluorescently labelled rRNA targeted oligonucleotide probe for the in situ detection of G-bacteria of the genus

Amaricoccus in activated sludge. Journal of Applied Microbiology 88: 826-835

Mathieu-Daudé, F., Welsh, J., Vogt, T., McClelland, M. 1996. DNA rehybridisation during PCR: the ‘C0t effect’ and its consequences. Nucleic Acids Research 24: 2080-

2086

254

Maukon et al. 2003. Methodologies for the characterisation of microbes in industrial environments: a review. Journal of Industrial Microbiology and Biotechnology 30: 327-

356

Maxcy, R.B. 1982. Lettuce salad as a carrier of microorganisms of public health significance. Journal of Food Protection 41: 435-438

McArthur, J.V., Kovacic, D.A., Smith, M.H. 1988. Genetic diversity in natural populations of a soil bacterium across a landscape gradient. Proceedings of the National

Academy of Science USA 85: 9621-9624

McDougall, S., Napier, T., Valenzisi, J., Watson, A., Duff, J., Geitz, G., Franklin, T.

2002. Integrated pest management in lettuce: information guide. NSW Agriculture,

NSW, Australia

McKillip, J.L., Jaykus, L.A., Drake, M.A. 2000. A comparison of methods for the detection of Escherichia coli O157:H7 from artificially-contaminated dairy products using PCR. Journal of Applied Microbiology 89: 49-55

Medlin, L., Ellwood, H.J., Stickel, S., Sogin, M.L. 1988. The characterisation of enzymatically amplified eukaryotic 16S-like rRNA coding regions. Gene 71: 491-499

255

Mehrotra, I. 2004. A perspective on developing and marketing food products to meet individual needs of population segments. Comprehensive Reviews in Food Science and

Food Safety 3: 142-144

Melick, C.O. 1917. The possibility of typhoid infection through vegetables. Journal of

Infectious Diseases 21: 28-38

Mercier, J., Marrone, P.G. 2006. Biological control of microbial spoilage of fresh produce. In Sapers, G.M., Gorny, J.R., Yousef, A.E. (eds). Microbiology of fruits and vegetables. CRC Press, Boca Raton, FL: 535-539

Mercier, J., Reeleder, R.D. 1986. Effects of the pesticides maneb and carbaryl on the phylloplane microflora of lettuce. Canadian Journal of Microbiology 33: 212-216

Meunier, S., Meyer, J.A. 1985. Modification de la flore epiphytique de feuilles de froment d’hiver suite a des traitements fungicides. Med. Fac. Landvouww. Rijksuniv.

Gent. 50: 1039-1043

Meyer, R., Luthy, J., Candrian, U. 1991. Direct detection by polymerase chain reaction

(PCR) of Escherichia coli in water and soft cheese and identification of enterotoxigenic strains. Letters in Applied Microbiology 13: 268

Miambi, E., Guyot, J-P., Ampe, F. 2003. Identification, isolation and quantification of representative bacteria from fermented cassava dough using an integrated approach of

256

culture-dependent and culture-independent methods. International Journal of Food

Microbiology 82: 111-120

Millard, P.S., Gensheimer, K.F., Addiss, D.G., Sosin, D.M., Beckett, G.A., Houck-

Jankoski, A., Hudson, A. 1994. An outbreak of cryptosporidiosis from fresh-pressed apple cider. Journal of the American Medical Association 272: 1592-1596

Modey, W.K., Mulholland, D.A., Raynor, M.W. 1996. Analytical supercritical fluid extraction of natural products. Phytochemical Analysis 7: 1-15

Monge, R., Chinchilla, M. 1996. Presence of Cryptosporidium oocysts in fresh vegetables. Journal of Food Protection 59: 202-203

Mukherjee, A., Speh, D., Jones, A.T., Buesing, K.M., Diez-Gonzalez, F. 2006.

Longitudinal microbiological survey of fresh produce grown by farmers in the Upper

Midwest. Journal of Food Protection 69: 1928-1936

Murray, C. 2001. Gram negative facultative rods. In Hocking, A.D., Arnold, G., Jenson,

I., Newton, K., Sutherland, P. (eds.). Foodborne microorganisms of public health significance, 5th Edition. Trenear Printing, Sydney: 319-328

Muyzer, G. 1999. DGGE/TGGE a method for identifying genes from natural ecosystems. Current Opinions in Microbiology 2: 317-322

257

Muyzer, G., De Waal, E.C., Uitterlinden, A.G. 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental

Microbiology 59: 695-700

Muyzer, G., Smalla, K. 1998. Application of denaturing gradient gel electrophoresis

(DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology.

Antonie van Leeuwenhoek 73: 127-141

Myers, R.M., Fischer, S.G., Lerman, L.S., Maniatis, T. 1985. Nearly all single base substitutions in DNA fragments joined to a GC-clamp can be detected by denaturing gradient gel electrophoresis. Nucleic Acids Research 9: 3131-3145

Naimi, T.S., Wicklund, J.H., Olsen, S.J., Krause, G., Wells, J.G., Bartkus, J.M., Boxrud,

D.J., Sullivan, M., Kassenborg, H., Besser, J.M., Mintz, E.D., Osterholm, M.T.,

Hedberg, C.W. 2003. Concurrent outbreaks of Shigella sonnei and enterotoxigenic

Escherichia coli infections associated with parsley: implications for surveillance and control of foodborne illness. Journal of Food Protection 66: 535-541

Nascimento, M.S., Silva, N., Catanozi, M.P.L.M., Silva, K.C. 2003. Effects of different disinfection treatments on the natural microbiota of lettuce. Journal of Food Protection

66: 1697-1700

258

National Food Industry Strategy (NFIS). 2003. Quantitative analysis, Food and

Beverage Sector, Key Performance Indicators and External Trade. BIS Shrapnel,

Sydney

Natvig, E.E., Ingham, S.C., Ingham, B.H., Cooperband, L.R., Rpoer, T.R. 2002.

Salmonella enterica serovar Typhimurium and Escherichia coli contamination of root and leaf vegetables grown in soils incorporated bovine manure. Applied and

Environmental Microbiology 68: 2737-2744

Nedoluha, P.C., Owens, S., Russek-Cohen, E., Westhoff, D.C. 2001. Effect of sampling method on the representative recovery of microorganisms from the surfaces of aquacultured finfish. Journal of Food Protection 64: 1515-1520

Neiderhauser, C., Candrian, U., Hofelein, C., Jermini, M., Buhler, H.P., Luthy, J. 1992.

Use of polymerase chain reaction for detection of Listeria monocytogenes in food.

Applied and Environmental Microbiology 58: 1564-1568

Nemec, A., De Baere, T., Tjernberg, I., Vaneechoutte, M., van der Reijden, T.J.K.,

Dijkshoorn, L. 2001. Acinetobacter ursingii sp. nov. and Acinetobacter schindleri sp. nov., isolated from human clinical specimens. International Journal of Systematic and

Evolutionary Microbiology 51: 1891-1899

259

Neyts, K., Huys, G., Uyttendaele, M., Swings, J., Debevere, J. 2000. Incidence and identification of mesophilic Aeromonas spp. from retail foods. Letters in Applied

Microbiology 31: 359-363

Nguyen-the, C., Carlin, F. 1994. The microbiology of minimally processed fresh fruit and vegetables. Critical Reviews in Food Science and Nutrition 34: 371-401

Nguyen-the, C., Carlin, F. 2000. Fresh and processed vegetables. In The microbiological safety and quality of food ed. Lund, B.M., Baird-Parker, T.C. and

Gould, C.W. pp. 620-684. Gaithersburg, Maryland: Aspen Publishers

Nguyen-the, C., Lund, B.M. 1991. The lethal effect of carrot of Listeria species. Journal of Applied Bacteriology 70: 479-488

Nguyen-the, C., Lund, B.M. 1992. An investigation of the antibacterial effect of carrot on Listeria monocytogenes. Journal of Applied Bacteriology 73: 23-30

Nguyen-the, C., Prunier, J.P. 1989. Involvement of pseudomonads in deterioration of

‘ready-to-use’ salads. International Journal of Food Science and Technology 24: 47-58

Nielsen, D.S., Teniola, O.D., Ban-Koffi, L., Owusu, M., Andersson, T.S., Holzapfel,

W.H. 2007. The microbiology of Ghanaian cocoa fermentations analysed using culture- dependent and culture-independent methods. International Journal of Food

Microbiology 114: 168-186

260

Normander, B., Prosser, J.I. 2000. Bacterial origin and community composition in the barley phytosphere as a function of habitat and presowing conditions. Applied and

Environmental Microbiology 66: 4372-4377

Nostro, A., Germanò, M.P., D’Angelo, V., Marino, A., Cannatelli, M.A. 2000.

Extraction methods and bioautography for evaluation of medicinal plant antimicrobial activity. Letters in Applied Microbiology 30: 379-384

Nübel, U., Engelen, B., Felske, A., Snaidr, J., Wieshuber, A., Amann, R.I., Ludiwg, W.,

Backhaus, H. 1996. Sequence heterogeneities of genes encoding 16S rRNAs in

Paenibacillus polymyxa detected by temperature gradient gel electrophoresis. Journal of

Bacteriology 178: 5636-5643

Nuorti, J.P., Niskanen, T., Hallanvuo, S., Mikkola, J., Kela, E., Hatakka, M.,

Fredriksson-Ahomaa, M., Lyytikäinen, O., Siitonen, A., Korkeala, H., Ruutu, P. 2004.

A widespread outbreak of Yersinia pseudotuberculosis O:3 infection from iceberg lettuce. Journal of Infectious Diseases 189: 766-774

Ogden, I.D., Fenlon, D.R., Vinten, A.J.A., Lewis, D. 2001. The fate of Escherichia coli

O157 in soil and its potential to contaminate drinking water. International Journal of

Food Microbiology 66: 111-117

261

Ogden, I.D., Hepburn, N.F., MacRae, M., Strachan, N.J.C., Fenlon, D.R., Rusbridge,

S.M., Pennington, T.H. 2002. Long-term survival of Escherichia coli O157 on pasture following an outbreak associated with sheep at a scout camp. Letters in Applied

Microbiology 34: 100-104

Ogram, A., Sayler, G.S., Barkay, T. 1987. The extraction and purification of microbial

DNA from sediments. Journal of Microbiological Methods 7: 57-66

Okafo, C.N., Umoh, V.J., Galadima, M. 2003. Occurrence of pathogens on vegetable harvested from soils irrigated with contaminated streams. Science of the Total

Environment 311: 49-56

Olsen, A.R., Hammack, T.S. 2000. Isolation of Salmonella spp. from the housefly,

Musca domestica L., and the dump fly, Hydrotaea aenescens (Wiedemann) (Diptera:

Muscidae), at a caged-layer houses. Journal of Food Protection 63: 958-960

Olsen, B.H., Nagy, L.A. 1984. Microbiology of potable water. Advances in Applied

Microbiology 30: 73-132

Osborn, A.M., Moore, E.R.B., Timmis, K.N. 2000. An evaluation of terminal restriction fragment length polymorphism (T-RFLP) analysis for the study of microbial community structure and dynamics. Environmental Microbiology 2: 39-50

262

Padaga, M., Heard, G.M., Paton, J.E., Fleet, G.H. 2000. Microbial species associated with different sections of broccoli harvested from three regions in Australia.

International Journal of Food Microbiology 60: 15-24

Palleroni, N.J., Holmes, B. 1981. Pseudomonas cepacia sp. nov., nom. rev.

International Journal of Systematic Bacteriology 31: 479-481

Palma, M., Taylor, L.T. 1999. Fractional extraction of compounds from grape seeds by supercritical fluid extraction and analysis for antimicrobial and agrochemical activities.

Journal of Agricultural and Food Chemistry 47: 5044-5048

Palombo, E.A., Semple, S.J. 2001. Antibacterial activity of traditional Australian medicinal plants. Journal of Ethnopharmocology 77: 151-157

Palumbo, J.D., O’Keeffe, T.L., Abbas, H.K. 2007. Isolation of maize soil and rhizospere bacteria with antagonistic activity against Aspergillus flavus and Fusarium verticillioides. Journal of Food Protection 70: 1615-1621

Parish, M.E., Beuchat, L.R., Suslow, T.V., Harris, L.J., Garrett, E.H., Farber, J.N.,

Busta, F.F. 2003. Methods to reduce/eliminate pathogens from fresh and fresh-cit produce. Comprehensive Reviews in Food Science and Food Safety 2(Supplement):

161-173

263

Park, S.J., Cho, J.C., Kim, S-J. 2003. Bacterial distribution and variation in water supply systems. Korean Journal of Microbiology 31: 245-254

Park, C.E., Sanders, G.W. 1992. Occurrence of thermotolerant campylobacters in fresh vegetables sold at farmers’ outdoor markets and supermarkets. Canadian Journal of

Microbiology 38: 313-316

Pederson, C.S., Fisher, P. 1944a. Bactericidal activity of vegetable juice. Journal of

Bacteriology 47: 421-422

Pederson, C.S., Fisher, P. 1944b. The bactericidal action of cabbage and other vegetable juices. NY State Agricultural Experimental Station Technical Bulletin 273: 3-32

Perombelon, M.C.M., Fox, R.A., Lowe, R. 1979. Dispersion of Erwinia carotovora in aerosols produced by the pulverisation of potato haulm prior to harvest. Phytopathology

Z. 94: 249-260

Perva-Uzulnalić, A., Škerget, M., Weinreich, B., Knez, Ž. 2004. Extraction of chilli pepper (var. Byedige) with supercritical CO2: effect of pressure and temperature on capsaicinoid and colour extraction efficiency. Food Chemistry 87: 51-58

Peusch, M., Müller-Seitz, E., Petz, M., Müller, A., Anklam, E. 1997. Extraction of capsaicinoids from chillies (Capsicum frutescens L.) and paprika (Capsicum annuum

264

L.) using supercritical fluids and organic solvents. Zeitschrift fur Lebensmittel

Unterschulung Forsch A 204: 351-355

Pianetti, A., Sabatini, L., Bruscolini, F., Chiaverini, F., Cecchetti, G. 2004. Faecal contamination indicators, salmonella, vibrio and aeromonas in water used for the irrigation of agricultural products. Epidemiology and Infection 132: 231-238

Picman, A.K., Towers, G.H.N. 1983. Antibacterial activity of sesquiterpene lactones.

Biochemical and Systematic Ecology 11: 321-327

Pingulkar, K., Kamat, A., Bongirwar, D. 2001. Microbiological quality of fresh leafy vegetables, salad components and ready-to-eat salads: an evidence of inhibition of

Listeria monocytogenes in tomatoes. International Journal of Food Science and

Nutrition 52: 15-23

Pivarnik, L.F., Donath, H., Patnoad, M.S., Roheim, C. 2005. New England consumers’ willingness to pay for fresh fruits and vegetables grown on CAP-certified farms. Food

Protection Trends 25: 256-266

Polz, M.F., Cavanaugh, C.M. 1998. Bias in template-to-product ratios in multitemplate

PCR. Applied and Environmental Microbiology 64: 3724-3730

Pomp, D., Medrano, J.F. 1991. Organic solvents as facilitators of polymerase chain reaction. Biotechniques 10: 58-59

265

Prakitchaiwattana, C.J., Fleet, G.H., Heard, G.M. 2004. Application and evaluation of denaturing gradient gel electrophoresis to analyse the yeast ecology of wine grapes.

FEMS Yeast Research 4: 865-877

Preston, M., Borczyk, A., Davidson, R., McGeer, A., Bertoli, J., Harris, S., Thususka,

J., Goldman, C., Green, K., Low, D., Proctor, P., Johnson, W., Khakhria, R. 1997.

Hospital outbreak of Escherichia coli O157:H7 associated with rare phage type –

Ontario. Canadian Communicable Disease Report 23: 33-37

Priest, F.G. 1989. Isolation and identification of aerobic endospore-forming bacteria. In:

Harwood, C.R. (Ed.), Bacillus. Plenum Press, New York, 27-56

Pulawska, J., Sobiczewski, P. (2005). Development of a semi-nested PCR based method for sensitive detection of tumorigenic Agrobacterium in soil. Journal of Applied

Microbiology 98: 710-721

Puttalingamma, V., Manja, K.S. 1998. Incidence of Aeromonas spp. in market vegetables. Journal of Mycology and Plant Pathology 28: 53-55

Ragaert, P., Devlieghere, F., Debevere, J. 2007. Role of microbiological and physiological spoilage mechanisms during storage of minimally processed vegetables.

Postharvest Biology and Technology 44: 184-194

266

Raiden, R.M., Sumner, S.S., Eifert, J.D., Pierson, M.D. 2003. Efficacy of detergents in removing Salmonella and Shigella spp. from the surface of fresh produce. Journal of

Food Protection 66: 2210-2215

Rauha, J-P., Remes, S., Heinonen, M., Hopia, A., Kähkönen, M., Kujala, T., Pihlaja, K.,

Vuorela, H., Vuorela, P. 2000. Antimicrobial effects of Finnish plant extracts containing flavonoids and other phenolic compounds. International Journal of Food Microbiology

56: 3-12

Raventós, M., Duarte, S., Alarcón, R. 2002. Application and possibilities of supercritical CO2 extraction in food processing industry: an overview. Food Science and

Technology International 8: 269-284

Rees, W.A., Yager, T.D., Korte, J., von Hippel, P.H. 1993. Betaine can eliminate the base pair composition dependence of DNA melting. Biochemistry 32: 137-144

Reverchon, E. 1997. Supercritical fluid extraction and fractionation of essential oils and related products. Journal of Supercritical Fluids 10: 1-37

Reysenbach, A.L., Giver, L.J., Wickham, G.S., Pace, N.R. 1992. Differential amplification of rRNA genes by polymerase chain reaction. Applied and Environmental

Microbiology 58: 3417-3418

267

Ricker, D.H., Taylor, S.R., Gartner, S.C., Kurland, G. 1991. Fatal pulmonary aspergillosis presenting as acute eosinophilic pneumonia in a previously healthy child.

Chest 100: 875-877

Riordan, D.C.R., Sapers, G.M., Hankinson, T.R., Magee, M., Mattrazzo, A.M., Annous,

B.A. 2001. A study of US orchards to identify sources of Escherichia coli O157:H7.

Journal of Food Protection 64: 1320-1327

Riser, E.C., Grabowski, J., Glenn, E.P. 1984. Microbiology of hydroponically grown lettuce. Journal of Food Protection 47: 765-769

Roberts, D., Watson, G.N., Gilbert, R.J. 1982. Contamination of food plants and plant products with bacteria of public health significance. In Rhodes-Roberts, M.E., Skinner,

F.A. (eds). Bacteria and Plants. Academic Press, London: 169-195

Robertson, L.J., Gjerde, B. 2001. Occurrence of parasites on fruits and vegetables in

Norway. Journal of Food Protection 64: 1793-1798

Robertson, L.J., Gjerde, B., Campbell, A.T. 2000. Isolation of Cyclospora oocysts from fruits and vegetables using lectin-coated paramagnetic beads. Journal of Food

Protection 63: 1410-1414

268

Robinson, I., Adams, R.P. 1978. Ultra-violet treatment of contaminated irrigation water and its effect on the bacteriological quality of celery at harvest. Journal of Applied

Bacteriology 45: 83-90

Rojas, R., Bustamante, B., Bauer, J., Fernández, I., Albán, J., Lock, O. 2003.

Antimicrobial activity of selected Peruvian medicinal plants. Journal of

Ethnopharmocology 88: 199-204

Rosas, I., Baez, A., Coutino, M. 1984. Bacteriological quality of crops irrigated with wastewater in the Xochimilco plots, Mexico City, Mexico. Applied and Environmental

Microbiology 47: 1074-1079

Rosenblum, L.S., Mirkin, I.R., Allen, D.T., Safford, S., Hadler, S.C. 1990. A multifocal outbreak of Hepatitis A traced to commercially distributed lettuce. American Journal of

Public Health. 80: 1075-1080

Rudi, K., Flateland, S.L., Hanssen, J.F., Bengtsson, G., Nissen, H. 2002. Development and evaluation of a 16S ribosomal DNA array-based approach for describing complex microbial communities in ready-to-eat vegetable salads packed in a modified atmosphere. Applied and Environmental Microbiology 68: 1146-1156

Rudolfs, W., Falk, L.L., Ragotzkie, R.A. 1951. Contamination of vegetables grown in polluted soil. I. Bacterial contamination. Sewage and Industrial Waste 23: 253-268

269

Saad, S.M.I., Iaria, S.T., Furlanetto, S.M.P. 1995. Motile Aeromonas spp. in retail vegetables from São Paulo, Brazil. Reviews in Microbiology 26: 22-27

Saddik, M.F., El-Sherbeeny, M.R., Bryan, F.L. 1985. Microbiological profiles of

Egyptian raw vegetables and salads. Journal of Food Protection 48: 883-886

Sadovski, A.Y., Fattal, B., Goldberg, D. 1978a. Microbial contamination of vegetables irrigated with sewage effluent by the drip method. Journal of Food Protection 41: 336-

340

Sadovski, A.Y., Fattal, B. Goldberg, D., Katzenelson, E., Shuval, H.I. 1978b. High levels of microbial contamination of vegetables irrigated with wastewater by the drip method. Applied and Environmental Microbiology 36: 824-830

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: 434-439

Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis,

K.B., Erlich, H.A. 1988. Primer-directed enzymatic amplification of DNA with thermostable DNA polymerase. Science 239: 487-491

Saltveit, M.E. 1999. Effects of ethylene on quality of fresh fruits and vegetables.

Postharvest Biology and Technology 15: 279-292

270

Sambrook, J., Fritsch, E.F., Maniatis, T. 1989. Molecular cloning- a laboratory manual,

2nd Edition. Cold Spring Harbor Laboratory Press, USA

Sambrook, J., Russell, D.W. 2001. Polymerase chain reaction. In Molecular cloning: a laboratory manual, the third edition. Cold Spring Harbor Laboratory Press, Cold Spring

Harbor, New York: 8.18-8.50

Sapers, G.M. 2001. Efficacy of washing and sanitizing methods for disinfection of fresh fruit and vegetable products. Food Technology and Biotechnology 39: 305-311

Sapers, G.M. 2006. Washing and sanitising treatments for fruits and vegetables. In

Sapers, G.M., Gorny, J.R., Yousef, A.E. (eds). Microbiology of Fruit and Vegetables.

CRC Press, Boca Raton, FL: 375-400

Sapers, G.M., Gorny, J.R., Yousef, A.E. (eds). 2006. Microbiology of Fruits and

Vegetables. CRC Press, Boca Raton, FL

Sarkar, G., Kapelner, S., Sommer, S.S. 1990. Formamide can dramatically improve the specificity of PCR. Nucleic Acids Research 18: 7465

Sato, K., Sasaki, S.S., Goda, Y., Yamada, T., Nunomura, O., Ishikawa, K., Maitani, T.

1999. Direct connection of supercritical fluid extraction and supercritical fluid

271

chromatography as a rapid quantitative method for capsaicinoids in placentas of

Capsicum. Journal of Agricultural Food Chemistry 47: 4665-4668

Schabereiter-Gurtner, C., Maca, S., Rölleke, S., Nigl, K., Lukas, J., Hirschl, A., Lubitz,

W., Barisani-Asenbauer, T. 2001. 16S rDNA-based identification of bacteria from conjunctival swabs by PCR and DGGE fingerprinting. IOVS 42: 1164-1171

Schlech, W.F., Lavigne, P.M., Bortolussi, R.A., Allen, A.C., Haldane, E.V., Wort, A.J.,

Hightower, A.W., Johnson, S.E., King, S.H., Nicholls, E.S., Broome, C.V. 1983.

Epidemic listeriosis – evidence for transmission by food. New England Journal of

Medicine 308: 203-206

Sekiguchi, H., Tomioka, N., Nakahara, T., Uchiyama, H. 2001. A single band does not always represent single bacterial strains in denaturing gradient gel electrophoresis.

Biotechnology Letters 23: 1205-1208

Sela, S., Nestel, D., Pinto, R., Nemny-Lavy, E., Bar-Joseph, M. 2005. Mediterranean fruit fly as a potential vector of bacterial pathogens. Applied and Environmental

Microbiology 71: 4052-4056

Seymour, I.J., Appleton, H. 2001. Foodborne viruses and fresh produce. Journal of

Applied Microbiology 91: 759-773

272

Shukla, A.K., Tiwari, B.K., Mishra, R.R. 1988. Effect of foliar application of herbicides, Bethiocarb, 2,4-D and Fluchloralin on phyllosphere microflora of potato.

Plant and Soil 106: 277-280

Singaglia, M., Albenzio, M., Corbo, M.R. 1999. Influence of process operations on shelf-life and microbial population of fresh-cut vegetables. Journal of Industrial

Microbiology and Biotechnology 23: 484-488

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: 2342-2353

Skovgaard, N., Morgen, C-A. 1988. Detection of Listeria spp. in faeces from animals, in feeds, and in raw foods of animal origin. International Journal of Food Microbiology

6: 229-242

Sloan, A.E. 2000. At the (fresh) cutting edge. Food Technology 54: 22-23

Sloan, A.E. 2005. Top 10 global food trends. Food Technology 59: 20-32

Sloan, A.E. 2007. Top 10 food trends. Food Technology 61: 23-39

Smith, K.T., Long, C.M., Bowman, B., Manos, M.M. 1990. Using cosolvents to enhance PCR amplification. Amplifications 5: 16-17

273

Solomon, E.B., Potenski, C.J., Matthews, K.R. 2002a. Effect of the irrigation method on transmission to and persistence of Escherichia coli O157:H7 on lettuce. Journal of Food

Protection 65: 673-676

Solomon, E.B., Yaron, S., Matthews, K.R. 2002b. Transmission of Escherichia coli

O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalisation. Applied and Environmental Microbiology 68: 397-400

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: 1243-1248

Speksnijder, A.G.C.L., Kowalchuk, G.A., De Jong, S., Kline, E., Stephen, J.R.,

Laanbroek, H.J. 2001. Microvariation artifacts introduced by PCR and cloning of closely related 16S rRNA gene sequences. Applied and Environmental Microbiology

67: 469-472

Stafford, R.J., McCall, B.J., Neill, A.S., Leon, D.S., Dorricott, G.J., Towner, C.D.,

Micalizzi, G.R. 2002. A statewide outbreak of Salmonella Bovismorbificans phage type

32 infection in Queensland. Communicable Diseases Intelligence 26: 568-573

Stevens, K.A., Jaykus, L-A. 2004. Bacterial separation and concentration from complex sample matrices: a review. Critical Reviews in Microbiology 30: 7-24

274

Steele, M., Mahdi, A., Odumeru, J. 2005. Microbial assessment of irrigation water used for production of fruit and vegetables in Ontario, Canada. Journal of Food Protection

68: 1388-1392

Steele, M., Odumeru, J. 2004. Irrigation water as a source of foodborne pathogens on fruit and vegetables. Journal of Food Protection 67: 2839-2849

Stevens, K.A., Jaykus, L-A. 2004. Bacterial separation and concentration from complex sample matrices: a review. Critical Reviews in Microbiology 30: 7-24

Stine, S.W., Song, I., Choi, C.Y., Gerba, C.P. 2005. Application of microbial risk assessment to the development of standards for enteric pathogens in water used to irrigate fresh produce. Journal of Food Protection 68: 913-918

Stockwell, V.O., McLaughlin, R.J., Henkels, M.D., Loper, J.E., Sugar, D., Roberts,

R.G. 1999. Epiphytic colonisation of pear stigmas and hypanthia by bacteria during primary bloom. Phytopathology 89: 1162-1168

Strauch, D. 1991. Survival of pathogenic microorganisms and parasites in excreta, manure and sewage sludge. Rev. Sci. Tech. Off. Int. Epiz. 10: 813-846

Surjadinata, B.B., Cisneros-Zevallos, L. 2003. Modeling wound-induced respiration of fresh-cut carrots (Daucus carota L.). Journal of Food Science 68: 2735-2740

275

Suslow, T.V., Oria, M.P., Beuchat, L.R., Garrett, E.H., Parish, M.E., Harris, L.J.,

Farber, J.N., Busta, F.F. 2003. Production practices s risk factors in microbial food safety of fresh and fresh-cut produce. Comprehensive Reviews in Food Science and

Food Safety 2(Supplement): 38-77

Suzuki, M.T., Giovannoni, S.J. 1996. Bias caused by template annealing in the amplification of mixtures of 16S rRNA genes by PCR. Applied and Environmental

Microbiology 62: 625-630

Suzuki, M., Rappe, M.S., Giovannoni, S.J. 1998. Kinetic bias in estimates of coastal picoplankton community structure obtained by measurements of small-subunit rRNA gene PCR amplicon length heterogeneity. Applied and Environmental Microbiology 64:

4522-5649

Svedhem, A., Kaijser, B. 1981. Isolation of Campylobacter jejuni from domestic animals and pets: probable origin of human infection. Journal of Infection 3: 37-40

Szabo, E.A., Coventry, M.J. 2001. Vegetable and vegetable products. In Moir, C.J.,

Andrew-Kabilafkas, C., Arnold, G., Cox, B.M., Hocking, A.D., Jenson, I. (eds).

Spoilage and processed foods: causes and diagnosis. Southwood Press, Marrickville,

NSW: 217-223

276

Szabo, E.A., Scurrah, K.J., Burrows, J.M. 2000. Survey for psychrotrophic bacterial pathogens in minimally processed lettuce. Letters in Applied Microbiology 30: 456-460

Szewzyk, U., Szewzyk, R., Manz, W., Schleifer, K-H. 2000. Microbiological safety of drinking water. Annual Reviews in Microbiology 54: 81-127

Takasugi, M., Okinaka, S., Katsui, N., Masamune, T., Shirata, A., Ohuchi, M. 1985.

Isolation and structure of Lettucenin A, a novel guaianolide phytoalexin from Lactuca sativa var. capitata (Compositae). Journal of the Chemical Society for Chemical

Communications 10: 621-622

Tamminga, S.K., Beumer, R.R., Kampelmacher, E.H. 1978. The hygienic quality of vegetables grown in or imported into the Netherlands: a tentative study. Journal of

Hygiene Cambridge 80: 143-154

Tang, M.Y., Cheong, Y.M., Zainuldin, T. 1994. Incidence of Listeria spp. in vegetables in Kuala Lumpur. Medical Journal of Malaysia 49: 217-222

Tassinari, A., Franco, B.D.G., Landgraf, M. 1994. Incidence of Yersinia spp. in food in

Sao Paulo, Brazil. International Journal of Food Microbiology 21: 263-270

Tekoriené, R. 2003. Bacteria of the genus Pseudomonas on rotting vegetables, fruit and seeds. Botanica Lithuanica 9: 203-206

277

Temmerman, R., Scheirlinck, I., Huys, G., Swings, J. 2003. Culture-independent analysis of probiotic products by denaturing gradient gel electrophoresis. Applied and

Environmental Microbiology 69: 220-226

Theunissen, J., Britz T.J., Torriani, S., Witthuhn, R.C. 2005. Identification of probiotic microorganisms in South African products using PCR-based DGGE analysis.

International Journal of Food Microbiology 98: 11-21

Thomas, D.S. 1988. Electropositively charged filters for the recovery of yeasts and bacteria from beverages. Journal of Applied Bacteriology 65: 35

Thompson, J.R., Marcelino, L.A., Polz, M.F. 2002. Heteroduplexes in mixed-template amplifications: formation, consequences and elimination by reconditioning PCR.

Nucleic Acids Research 9: 2083-2088

Thunberg, R.L., Tran, T.T., Bennett, R.W., Matthews, R.N., Belay, N. 2002. Microbial evaluation of selected fresh produce obtained at retail markets. Journal of Food

Protection 65: 677-682

Thurman, R., Faulkner, B., Veal, D., Cramer, G., Meiklejohn, M. 1998. Water quality in rural Australia. Journal of Applied Microbiology 84: 627-632

Tournas, V.H. 2005. Moulds and yeasts in fresh and minimally processed vegetables, and sprouts. International Journal of Food Microbiology 99: 71-75

278

Tyrrel, S.F., Knox, J.W., Weatherhead, E.K. 2006. Microbiological water quality requirements for salad irrigation in the United Kingdom. Journal of Food Protection 69:

2029-2035

Ukuku, D.O., Fett, W. 2002. Relationship of cell surface charge and hydrophobicity to strength of attachment of bacteria to cantaloupe rind. Journal of Food Protection 63:

1093-1099

Ultee, A., Souvatzi, N., Maniadi, K., König, H. 2004. Identification of the culturable and nonculturable bacteria population in ground water of a municipal water supply in

Germany. Journal of Applied Microbiology 96: 560-568

US EPA (Environmental Protection Agency). 2007. Tolerances and exemptions from tolerances for pesticide chemicals in foods. Available at: http://www.epa.gov/pesticides/regulating/laws.htm. Accessed 21 July, 2007.

Vakili, N.G. 1967. Importance of wounds in bacterial spot (Xanthomonas vesicatoria) of tomatoes in the field. Phytopathology 57: 1099-1103

van Antwerpen, T., Rutherford, R.S., Vogel, J.L. 2002. Assessment of sugarcane endophytic bacteria and rhizospheric Burkholderia species as antifungal agents.

Proceedings of the Annual Congress of the South African Sugar Technology

Association 76: 301-304

279

Vanbroekhoven, K., Ryngaert, A., Wattiau, P., De Mot, R., Springael, D. 2004.

Acinetobacter diversity in environmental samples assessed by 16S rRNA gene PCR-

DGGE fingerprinting. FEMS Microbiological Ecology 50: 37-50

van Hannen, E.J., van Agterveld, M.P., Gons, H.J., Laanbroek, H.J. 1998. Revealing genetic diversity of eukaryotic microorganism in aquatic environments by denaturing gradient gel electrophoresis. Journal of Phycology 34: 206-213

Van Looveren, M., Goossens, H. and the ARPAC Steering Group (2004) Antimicrobial resistance of Acinetobacter spp. in Europe. Clinical and Microbiology Infection 10:

684-704

Van Renterghem, B., Huysman, F., Rygole, R., Verstraete, W. 1991. Detection and prevalence of Listeria monocytogenes in the agricultural ecosystem. Journal of Applied

Bacteriology 71: 211-217

Varadaraj, K., Skinner, D.M. 1994. Denaturants or cosolvents improve the specificity of

PCR amplification of a G+C–rich DNA using genetically engineered DNA polymerases. Gene 140: 1-5

Vaz da Costa-Vargas, S.M., Mara, D.D., Vargas-Lopez, C.E. 1991. Residual faecal contamination on effluent-irrigated lettuces. Water Science and Technology 24: 89-94

280

Villari, P., Crispino, M., Montuori, P., Stanzione, S. 2000. Prevalence and molecular characterization of Aeromonas spp. in ready-to-eat foods in Italy. Journal of Food

Protection 63: 1754-1757

Viswanathan, P., Kaur, R. 2001. Prevalence and growth of pathogens on salad vegetables, fruits and sprouts. International Journal of Hygiene and Environmental

Health 203: 205-213

Volcani, Z. 1969. The effect of mode of irrigation and wind direction on disease severity caused by Xanthomonas vesicatoria on tomato in Israel. Plant Disease Reporter

53: 459-461

von Wintzingerode, F.V., Göbel, U.B., Stackebrandt, E. 1997. Determination of microbial diversity in environmental samples: pitfalls of PCR-based rRNA analysis.

FEMS Microbiological Reviews 21: 213-229

Wachtel, M.R., Whitehand, L.C., Mandrell, R.E. 2002a. Association of Escherichia coli

O157:H7 with preharvest leaf lettuce upon exposure to contaminated irrigation water.

Journal of Food Protection 65: 18-25

Wachtel, M.R., Whitehand, L.C., Mandrell, R.E. 2002b. Prevalence of Escherichia coli associated with a cabbage crop inadvertently irrigated with partially treated sewage wastewater. Journal of Food Protection 65: 471-475

281

Wagner, A., Blackstone, N., Cartwright, P., Dick, M., Misof, B., Snow, P., Wagner,

G.P., Bartels, J., Murtha, M., Pendleton, J. 1994. Surveys of gene families with polymerase chain reaction-PCE selection and PCR drift. Systematic Biology 43: 250-

261

Walker, J.R.L. 1994. Antimicrobial compounds in food plants. In Dillon, V.M., Board,

R.G. (eds). Natural antimicrobial systems and food preservation. CAB International,

UK: 181

Wallace, J.S., Cheasty, T., Jones, K. 1997. Isolation of vero cytotoxin-producing

Escherichia coli O157 from wild birds. Journal of Applied Microbiology 82: 399-404

Wang, G., Doyle, M.P. 1998. Survival of enterohemorrhagic Escherichia coli O157:H7 in water. Journal of Food Protection 61: 662-667

Wang, G., Zhao, T., Doyle, M.P. 1996. Fate of enterohemorrhagic Escherichia coli

O157:H7 in bovine feces. Applied and Environmental Microbiology 62: 2567-2570

Watada, A.E., Ko, N.P., Minott, D.A. 1996. Factors affecting quality of fresh-cut horticultural products. Postharvest Biology and Technology 9: 115-125

Watkins, J., Sleath, K.P. 1981. Isolation and enumeration of Listeria monocytogenes from sewage, sewage sludge and river water. Journal of Applied Bacteriology 50: 1-9

282

Weil, A., Hampel, A. 1973. Preparative agarose gel electrophoresis of ribonucleic acid.

Biochemistry 12: 4361-4367

Welinder-Olsson, C., Stenqvist, K., Badenfors, M., Brandberg, Å, Florén, K., Holm, M.,

Holmberg, L., Kjellin, E., Mårild, S., Studahl, A., Kaijser, B. 2003. EHEC outbreak among staff at a children’s hospital- use of PCR for verocytotoxin detection and PFGE for epidemiological investigation. Epidemiology and Infection 132: 43-49

Wells, J.M., Butterfield, J.E. 1999. Incidence of Salmonella on fresh fruits and vegetables affected by fungal rots or physical injury. Plant Disease 83: 722

Wells, P.A., Foster, N.R. 1986. Extraction processes using supercritical fluids.

Chemical Engineering in Australia 11: 10-14

Welshimer, H.J. 1960. Survival of Listeria monocytogenes in soil. Journal of

Bacteriology 80: 316-320

Wilson, I.G. 1997. Inhibition and facilitation of nucleic acid amplification. Applied and

Environmental Microbiology 63: 3741-3751

Wright, C., Kominos, S.D., Yee, R.B. 1976. Enterobacteriaceae and Pseudomonas aeruginosa recovered from vegetables salads. Applied and Environmental Microbiology

31: 453-454

283

Yajima, M., Nozaki, K., Takayanagi, T., Yokotsuka, K. 1997. An antimicrobial fraction from the residue obtained by supercritical carbon dioxide extraction of Capsicum spp. for use in food preservation. Journal of Antibacterial and Antifungal Agents 25: 131-

137

Yamamoto, S., Kasai, H., Arnold, D.L., Jackson, R.W., Vivian, A., Harayama, S. 2000.

Phylogeny of the genus Pseudomonas: intrageneric structure reconstructed from the nucleotide sequences of gyrB and rpoD genes. Microbiology 146: 2385-2394

Yang, C-H., Crowley, D.E. 2000. Rhizosphere microbial community structure in relation to root location and plant iron nutritional status. Applied and Environmental

Microbiology 66: 345-351

Yao, J., Nair, M.G., Chandra, A. 1994. Supercritical carbon dioxide extraction of Scotch

Bonnet (Capsicum annuum) and quantification of capsaicin and dihydrocapsaicin.

Journal of Agricultural Food Chemistry 42: 1303-1305

Young, J.M., Kuykendall, L.D., Martinez-Romero, E., Kerr, A., Sawada, H. 2001. A revision of Rhizobium Frank 1889, with an emended description of the genus, and the inclusion of all species of Allorhizobium undicola de Lajudie et al. 1998 as new combinations: Rhizobium radiobacter, R. rhizogenes, R. rubi, R. undicola and R. vitis.

International Journal of Systematic and Evolutionary Microbiology 51: 89-103

284

Yu, K., Newman, M.C., Archbold, D.D., Hamilton-Kemp, T.R. 2001. Survival of

Escherichia coli O157:H7 on strawberry fruit and reduction of the pathogen population by chemical agents. Journal of Food Protection 64: 1334-1340

Zhang, J. 30 November 2005. When eating your vegetables makes you sick. Wall Street

Journal, Section D: 1

Zhou, J., Bruns, M.A., Tiedje, J.M. 1996. DNA recovery from soils of diverse composition. Applied and Environmental Microbiology 62: 316-322

Zhuang, H., Barth, M.M., Hildebrand, D.F. 1994. Packaging influenced total chlorophyll, soluble protein, fatty acid composition and lipoxgenase activity in broccoli florets. Journal of Food Science 59: 1171-1174

285