Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2022 Aug 13.
Published in final edited form as: FEMS Microbiol Lett. 2015 May 7;362(11):fnv077. doi: 10.1093/femsle/fnv077

Older leaves of lettuce (Lactuca spp.) support higher levels of Salmonella enterica ser. Senftenberg attachment and show greater variation between plant accessions than do younger leaves

Paul J Hunter 1,*, Robert K Shaw 2, Cedric N Berger 2, Gad Frankel 2, David Pink 1, Paul Hand 1
PMCID: PMC7613271  EMSID: EMS151450  PMID: 25953858

Abstract

Salmonella can bind to the leaves of salad crops including lettuce and survive for commercially relevant periods. Previous studies have shown that younger leaves are more susceptible to colonization than older leaves and that colonization levels are dependent on both the bacterial serovar and the lettuce cultivar. In this study, we investigated the ability of two Lactuca sativa cultivars (Saladin and Iceberg) and an accession of wild lettuce (L. serriola) to support attachment of Salmonella enterica serovar Senftenberg, to the first and fifth to sixth true leaves and the associations between cultivar-dependent variation in plant leaf surface characteristics and bacterial attachment. Attachment levels were higher on older leaves than on the younger ones and these differences were associated with leaf vein and stomatal densities, leaf surface hydrophobicity and leaf surface soluble protein concentrations. Vein density and leaf surface hydrophobicity were also associated with cultivar-specific differences in Salmonella attachment, although the latter was only observed in the older leaves and was also associated with level of epicuticular wax.

Keywords: Salmonella, attachment, food safety, lettuce, phenotypic interactions, zoonoses

Introduction

Between 1992 and 2006, 23% ‘food poisoning’ outbreaks in England and Wales were linked to prepared salad crops (Little and Gillespie 2008), with lettuce being one of the salads most frequently associated with illness caused by food-borne pathogens (Warriner et al. 2003). Over a similar timeframe, Salmonella enterica was associated with 30% of such outbreaks in the USA (second only to Escherichia coli) (Brandl 2006). Despite recent implementation of assured production protocols, improvements in quality assurance testing and the advent of statutory monitoring through non-governmental agencies (FSA, EFSA, etc.), which have reduced the incidence of food-poisoning outbreaks (Monaghan 2006), incidents related to consumption of fresh produce still occur, indicating continuing issues with contamination by human pathogens.

Human pathogenic bacteria including Salmonella can bind to the leaves of various salad crops including lettuce and can survive in the phyllosphere (the aerial part of plants) for periods in excess of 30 weeks (Islam et al. 2004; Stine et al. 2005). Standard post-harvest decontamination procedures typically involve short immersion of salad crops in solutions containing approximately 20–200 μg mL-1 of active chlorine. These treatments reduce bacterial numbers but do not completely eliminate either the natural microbial population or human pathogens (Beuchat and Brackett 1990; Seo and Frank 1999; Lang, Harris and Beuchat 2004). This is a particular issue with ‘minimally processed’ salad crops, which are normally eaten uncooked, as there is a limit to the stringency of decontamination that can be applied without ruining the saleability of the produce.

Previous work has identified the younger leaves of plants as being able to support higher levels of enteric human pathogens than older leaves (Brandl and Amundson 2008); however, cultivar-specific variation in the level of contamination of lettuce has also been demonstrated for a number of food poisoning-associated Salmonella strains (Barak, Liang and Narm 2008) and S. enterica serovars have also been shown to differ in their ability to colonize lettuce plants (Klerks et al. 2007) suggesting both plant cultivar and pathogen serovar as potential sources of variation. In previous work, we have shown that the composition of the naturally occuring bacterial population which developed on field-grown lettuce differed significantly with cultivar, particularly in respect of the Enterobacteriaceae (the bacterial family to which Salmonella belongs) (Hunter et al. 2010). This variation was found to be associated with plant morphotype (a composite of leaf and whole-plant morphological traits) and levels of soluble carbohydrate and phenolic compounds. In this work, we examined whether plant leaf age and/or cultivar had an effect on the ability of lettuce leaf tissue to support attachment of the same Salmonella isolate and if any such differences could be correlated with differences in leaf surface characteristics.

Materials and Methodology

Plant accessions and experimental design

Lettuce (Lactuca sativa) cultivars Saladin (iceberg type) and Iceberg (batavian type) and the wild lettuce relative L. serriola (US96UC23) were raised under glass (18–22°C) in modular arrays (P40; Plantpak, Maldon, UK) of Levington F2s compost (Levington Horticulture Limited, Ipswich, UK). The oldest or newest fully expanded true leaves (corresponding to the first and approximately fifth or sixth true leaves) were inoculated with S. enterica ser. Senftenberg strain 070885 or used for leaf characteristic measurements at 6 weeks post-sowing. Separate sets of 40 plants of each accession randomly distributed within the modular trays were raised consecutively and plants from each set tested for both bacterial adhesion and plant leaf characteristics. Tests comparing characteristics of younger and older leaves were made on leaves from the same plants.

Leaf-adhesion assays followed the method of Shaw et al. (2008), as modified by Berger et al. (2009): 10-mm wide strips were trimmed from six replicate freshly excised leaves of each age from each plant accession into 30-mm diameter petri dishes and immersed in 4 mL of a S. enterica ser. Senftenberg culture grown to an OD600 of 1.00 (6–7 × 108 CFU mL-1) at room temperature in Luria Bertani (LB) broth. Samples were incubated statically at room temperature for 1 h to allow bacterial adhesion, but minimize time for subsequent proliferation (thus minimizing the influence of confounding factors). This corresponds with Cevallos-Cevallos et al. (2012) who found that 100% of leaf adhesion in tomato had occurred by 1 h. Non-adherent bacteria were removed by three 5-min washes in 3 mL of sterile distilled water (SDW) at 80 rpm in an orbital shaker. A 6-mm diameter (approximately 28 mm2) leaf disc was excised from each sample and macerated in SDW. The macerates were serially diluted in SDW and incubated on MacConkey agar overnight at 37°C to recover leaf-associated bacteria. An additional disc was excised from each sample and fixed in 3% glutaraldehyde in 100 mM sodium phosphate buffer pH 7.3 and processed for scanning electron microscopy (SEM) (Knutton 1995) to confirm bacterial attachment.

Leaf characteristics

Leaf characteristics were determined from five replicate samples of each leaf age from individual plants of each accession, grown contemporaneously, under the same conditions as those used for bacterial inoculation. Data for physical characteristics and leaf surface hydrophobicity were extracted from captured images using the Image Tool software (freeware available from University of Texas Health Science Center San Antonio at http://ddsdx.uthscsa.edu/dig/download.html). Five detached leaves for each plant accession and leaf age category were weighed, and full-leaf images of all samples used to determine leaf area and to measure distances between veins, along the mid-rib of each leaf. These latter measurements were used to calculate inter-vein distance (an inverse estimator of vein density) following the method of Uhl and Mossberger (1999). To measure water droplet contact angle (an estimator of surface hydrophobicity), leaf strips from the broadest portion of leaves from each physiological age from five replicate plants of each accession were cut and mounted onto microscope slides. Two strips were cut from each leaf, one mounted adaxial side uppermost; the other abaxial side uppermost and five 2-μl droplets of sterile, ultrapure water were placed on the surface of the central portion of each leaf strip. Edge-on images of the droplets were captured for contact angle measurement. Leaf strips were only handled at the ends and the regions handled not used for subsequent testing. Values for both upper and lower leaf surfaces were acquired for all plant accessions, and the data for both surfaces combined.

Stomatal density and cell perimeter density were estimated from SEM images. Tissue was washed with 100 mM sodium phosphate buffer, pH 7.3, fixed for 1 h in 250 mM glutaraldehyde, then 1 h in 40 mM OsO4, ethanol dried and sputter-coated with platinum (Knutton 1995). Three random fields of view from each leaf surface (upper and lower) of a single leaf of each age from five replicate plants of each accession were examined. To estimate stomatal density, stomatal counts in each field of view were made from images taken at approximately 150× magnification. Estimates of cell perimeter density required higher magnification images (approximately 500×). The number of cells in each field of view (n) was determined, and the combined linear perimeter distance (d) of five individual whole cells from each field of view also measured. The cell perimeter density for each field for view (area = a) was then estimated as: [n × (d/5)]/a. The areas of the fields of view for both magnifications were calculated from scale bars incorporated into the digital image output from the SEM.

Water-soluble leaf surface components (protein, sugars and phenolic compounds) were extracted following a modification of the method of Mercier and Lindow (2000). Individual leaves were detached from plants, imaged to enable determination of leaf area and then immersed in 10 mL of sterile ultrapure water overnight at 4°C with care taken to avoid immersion of the cut ends of the leaf bases. Each of five replicate samples for each leaf age from each plant accession comprised combined extracts of five leaves from separate plants, filter sterilized though a 0.2-μm syringe filter (Nalgene, VWR International (UK), Lutterworth, UK) and lyophilized. The dried sample was resuspended in 500 μL of ultrapure water. Separate extractions were made for each component. Protein concentrations were determined by A595 measurements against a standard curve of BSA using the Coomassie (Bradford) protein assay kit (Fisher Scientific (UK), Loughborough, UK). Sugar concentrations were determined using a modification of the 3,5-dinitrosalycilic acid (DNS) procedure of Bernfeld (1955): 250 μl of sample was boiled for 10 min with 750 μl of [44 mM DNS; 500 mM NaOH; 700 mM potassium sodium tartrate], and then cooled to room temperature. A540 measurements were compared to a glucose standard curve. Soluble phenolic compounds were measured using a variation of the method of Zhang et al. (2006): 50 μL of sample was diluted 1:10 with SDW and incubated for 5 min at room temperature with 100 μL Folin Ciocalteu reagent (Sigma-Aldrich Company Ltd, Dorset, UK). A total of 2 mL of 600 mM Na2CO3 was then added and the supernatant cleared by centrifugation for 1 min at 13 000 rpm after 2 h incubation at room temperature. A765 measurements of the supernatant were made against an SDW blank and compared to a gallic acid standard curve to estimate phenolic compound concentrations.

Leaf wax content was estimated by a modification of the process of Bakker et al. (1998) from five replicate samples of each leaf age from each plant accession: for each replicate sample, five detached leaves were sequentially extracted by gentle agitation for 15 min at room temperature in the same 10 mL volume of chloroform. The chloroform extracts were filtered into pre-weighed vials and evaporated to constant weight at room temperature to determine the weight of extracted wax.

Statistical analyses

The significances of differences in the various characteristics measured were tested by ANOVA. Comparisons were made both within and between leaf age categories. Characteristics showing significant differences were tested for correlation with Salmonella levels using Spearman’s rank correlation test. The significance of any identified correlations was then determined using a permutation test that allowed for tied ranks and differences in sample size.

Results and Discussion

We investigated the attachment of S. enterica ser. Senftenberg to leaves of three lettuce accessions (Table 1). This revealed that overall levels of attachment were significantly higher (P < 0.0001) on older leaves than younger ones (Table 1) and that significant differences existed between accessions (P 0.0463). Post-hoc tests revealed no significant differences between Salmonella attachment to younger leaves. In older leaves, significantly greater attachment (P < 0.05) to L. serriola US96UC23 than to the other two accessions was observed. While Brandl and Amundson (2008) showed that Salmonella colonize younger plant leaves more heavily than older leaves, in this study, plants were tested at a much younger age (1–6 true leaves) than those in the previous work (10–12 true leaves and mature heads). Brandl and Amundson also measured combined attachment and persistence/proliferation over 72 h at 38°C, whilst the assay timeframe in this study (1 h) is too short for bacterial proliferation to be a major factor, suggesting that different parameters may influence initial bacterial attachment and subsequent growth.

Table 1.

Bacterial counts and leaf surface characteristics for first and fifth to sixth true leaves from three lettuce accessions, showing results of ANOVA tests between leaf age categories with subsequent Spearman rank correlation coefficients and associated probabilities for association of characteristics with Salmonella counts.

Salmonella
enterica
ser. Senftenberg
(cfu ml−1)		 Phenolic
content
(mg g−1) 		Surface
protein
(ng cm−2)		 Surface
wax
(μg cm−2)		 Surface
sugars
(μg cm−2)		 Contact
angle
(°)		 Inter-vein
distance
(mm)		 Stomatal
density
(N° mm−2)		 Cell perimeter
density
(mm mm−2)
Saladin (1st leaf) 21333a 81.8a 3.59b 409.1d 6.67d 54.8c 4.2b 165.2c 240.2b
Saladin (5–6th leaf) 54333c 97.1a 0.52a 28.6a 2.23b 55.4d 9.7d 79.8a 135.2a
Iceberg (1st leaf) 13166a 81.8a 1.43a 123.8b 2.19b 43.9a 6.6c 115.1b 199.4a,b
Iceberg (5–6th leaf) 34166b 95.7a 0.18a 8.1a 0.87a 50.2b 12.5e 72.4a 162.2a
Serriola (1st leaf) 10816a 88.7a 7.18c 190.6c 8.13e 134.6f 2.9a 293.1d 153.9a
Serriola (5–6th leaf) 32666b 118.3b 1.56a 8.1a 2.75c 129.0e 6.8c 89.0a 132.7a
No. of plants 6 5 5 5 5 5 5 5 5
ANOVA (age) <0.0001*** 0.0031** <0.0001*** <0.0001*** <0.0001*** <0.0001*** <0.0001*** <0.0001*** 0.0090**
l.s.d. (P 0.05) 4821 20.2 2.68 21.8 0.08 0.2 0.2 23.1 68.0
Spearman correlation R 0.328 −0.780 −0.597 −0.506 −0.731 0.739 −0.795 −0.322
P 0.116 0.044* 0.056 0.070 0.027* 0.035* 0.042* 0.172
Open in a new tab

*, **, *** Results significant at 0.05, 0.01 and 0.001 level, respectively (no asterisks indicate non-significant results), l.s.d. (P 0.05)—least significant difference between means representing a significant difference at the 0.05 level, R—Spearman rank correlation coefficient (vs S. enterica ser, Senftenberg counts), P—probability (significance of correlation), a–findicate significant differences between values in each column based on l.s.d.

Significant differences were observed in all tested leaf characteristics between leaf ages (Table 1). Nutrition levels, particularly soluble sugar concentrations, are reportedly important in determining bacterial load of leaves. Mercier and Lindow (2000) reported significant correlations between initial leaf sugar concentrations and bacterial (Pseudomonas fluorescens) populations developing over 48 h, whilst Brandl and Amundson suggested higher levels of available nutrition in the younger leaves as a potential driver for the higher levels of colonization they noted. This study showed significantly increased concentrations of soluble leaf-surface sugars in young leaves of all cultivars compared to the older leaves but no significant correlation (P 0.070) with Salmonella attachment levels indicating that soluble sugar levels do not significantly influence Salmonella attachment. Sugar levels may, however, result in a more favourable environment for bacterial growth on the surface of younger leaves, which could become the determining factor in the overall colonization level over time.

A significant positive correlation was observed between Salmonella attachment and inter-vein distance. Significant negative correlations were also identified with water-soluble protein concentration, water droplet contact angle (hydrophobicity) and stomatal density (Table 1). Plant cultivar may also be a factor: Brandl and Amundson used a romaine-type lettuce (cv. Paris Island), whilst iceberg and batavian types (cvs. Saladin and Iceberg, respectively) and the wild relative L. serriola, were used in this study. These plants have considerably different leaf morphologies compared to the romaine type and morphotype has been shown to influence bacterial colonization of lettuce, particularly by members of the Enterobacteriaceae (which includes the genus Salmonella) (Hunter et al. 2010; Cevallos-Cevallos et al. 2012). Differences in Salmonella attachment levels between the cultivars used in this work were identified, although these were only significant in older leaves (P 0.0122), with cv. Saladin supporting significantly (P < 0.05) higher levels of attachment than either cv. Iceberg or L. serriola. Cultivar-level differences in leaf surface-soluble protein, sugar and wax concentrations, water droplet contact angle and inter-vein distance were also detected in the corresponding older leaf samples. Of these, inter-vein distance and surface wax were positively correlated with Salmonella attachment, whilst hydrophobicity was negatively correlated (Table 2).

Table 2.

Bacterial counts and leaf surface characteristics for fifth to sixth true leaves from three lettuce accessions, showing results of ANOVA tests between accessions with subsequent Spearman rank correlation coefficients and associated probabilities for association of characteristics with Salmonella counts.

Salmonella
enterica
ser. Senftenberg
(cfu ml−1)		 Phenolic
content
(mg g−1) 		Surface
protein
(ng cm−2)		 Surface
wax
(μg cm−2)		 Surface
sugars
(μg cm−2)		 Contact
angle
(°)		 Inter-vein
distance
(mm)		 Stomatal
density
(N° mm−2)		 Cell perimeter
density
(mm mm−2)
Saladin (5–6th leaf) 54333b 97.1 0.52a 28.6b 2.23b 55.4b 9.7b 79.8 135.2
Iceberg (5–6th leaf) 34166a 95.7 0.18a 8.1a 0.87a 50.2a 12.5c 72.4 162.2
Serriola (5–6th leaf) 32666a 118.3 1.56b 8.1a 2.75c 129.0c 6.8a 89.0 132.7
No. of plants 6 5 5 5 5 5 5 5 5
ANOVA (cv) 0.0122* 0.6000 0.0165* <0.0001*** <0.0001*** <0.0001*** <0.0001*** 0.3022 0.3000
l.s.d. (P 0.05) 8962 nt 0.53 5.3 0.40 3.1 1.1 nt nt
Spearman correlation R nt −0.357 0.668 −0.151 −0.747 0.420 nt nt
P 0.063 <0.0001*** 0.187 <0.0001*** 0.0009***
Open in a new tab

*, **, *** results significant at 0.05, 0.01 and 0.001 level, respectively (no asterisks indicate non-significant results), l.s.d. (P 0.05)—least significant difference between means representing a significant difference at the 0.05 level, nt—not tested further since ANOVA revealed no significant differences in characteristic between accessions. R—Spearman rank correlation coefficient (vs S. enterica ser. Senftenberg counts). P—probability (significance of correlation), a–cindicate significant differences between the data in each column based on l.s.d.

Stomata and grooves along the length of leaf veins are known to be preferential bacterial attachment sites on leaf surfaces (Mariano and McCarter 1993). In this study, negative correlations were shown with stomatal density and leaf vein density. SEM images both from this work (Fig. 1) and previous studies of S. enterica ser. Senftenberg (Berger et al. 2009) revealed diffuse attachment patterns rather than an association with the typical preferential bacterial attachment sites (i.e. veins and stomata), although Cevallos-Cevallos et al. (2012) implicated trichome density as a factor in Salmonella attachment to tomato leaves (we were unable to obtain sufficiently robust trichome density measurements to permit analysis). It is interesting to note, however, that in contrast to the distribution of S. enterica ser. Senftenberg on lettuce, Salmonella serovar Thompson has been shown to be associated with vein structures on leaves of corriander (Coriandrum sativum) (Brandl and Mandrell 2002), suggesting a stainspecific component to attachment in Salmonella. This may reflect differences in attachment mechanisms: thin aggregative fimbriae (known as tafi or curli) have been shown to play a role in S. enterica ser. Newport attachment to alfalfa sprouts (Barak et al. 2005) and Salmonella attachment to tomato leaves (Cevallos-Cevallos et al. 2012), whilst flagellae have been shown to be strongly involved in attachment of S. enterica ser. Senftenberg to a range of salad leaves (Berger et al. 2009).

Figure 1.

Figure 1

Open in a new tab

Scanning electron micrographs of S. enterica serovar Senftenberg attachment to lettuce cultivars Saladin (a, b), Iceberg (c, d) and L. serriola US96UC23 (e, f) on first true leaves (right-hand panels) and fifth true leaves (left-hand panels). All lines supported a diffuse adhesion pattern at both growth stages, often associated with filamentous networks (panel a inset). Bar = 10 μm. All samples were taken rom from the centre of the leaf lamella.

Furthermore, expansion of the leaf surface as the leaves matured resulting in reduced densities of these characteristics in older leaves (Van Volkenburgh 1999) may have contributed to the observed negative correlations. Although leaf expansion rates were not measured in this study, a similar level of reduction in the mean density of both veins and stomata (2–3-fold) between younger and older leaves was noted in all accessions.

The negative correlation between Salmonella attachments and protein levels suggested the presence of proteins on the leaf surface that either favoured growth of antagonistic microorganisms or reduced either general bacterial adhesion or survival or that of Salmonella specifically. SEM of inoculated samples revealed no indication of non-target microorganisms and the time scale of the experiment (1 h) would have provided only limited opportunity for microbial growth and/or interaction, consequently microbial antagonism seems unlikely. Proteins and peptides with antimicrobial properties have been reported in leaf washes and guttation fluid from a number of species however (Young et al. 1995; Grunwald et al. 2003; Shepherd et al. 2005). The deposition of such peptides on the leaf surface is likely to be influenced by localized water distribution and flow and could result in channelling of leaf surface water along the areas of lower hydrophobicity, e.g. vein structures (Leben 1988) conceivably leading to increased localized concentrations of antimicrobial peptides in these regions potentially contributing to the observed negative correlation of Salmonella attachment with vein density.

Leaf wax levels have been shown to directly influence adherence of microorganisms to leaf surfaces (Beatie 2002). A negative correlation (P 0.027) was observed between Salmonella levels and hydrophobicity. There was, however, no significant correlation between overall levels of surface wax and hydrophobicity (P 0.479) as might be expected, suggesting that total wax content was not the factor determining the hydrophobicity of the leaf surface.

Successional differences in epicuticular leaf wax composition during leaf development have been shown in cherry laurel (Prunus laurocerasus) (Jetter and Schäffer 2001). These changes, detectable in leaves as young as 10 days old, could potentially result in different levels of surface hydrophobicity without significant impact on total epicuticular wax levels. Leaf surface SEM images of L. serriola and the two L. sativa cultivars appeared to show physically different wax microstructures (Fig. 2). Although certain wax components have been associated with specific wax crystalloid microstructures (Barthlott et al. 1998), the observed differences did not correspond with Salmonella attachment from older leaves: counts were significantly higher from cv. Saladin (L. sativa) than from cv. Iceberg, whilst counts from cv. Iceberg and L. serriola (showing different wax morphology) were not significantly different. Differences in wax composition between closely related species such as Thellungiella halophila and T. Parvula have been reported (Teusnik et al. 2002), which did not result in any observable differences in leaf waxiness however. It is possible that such differences could result in leaf surface environments with different hydrophobicity. Furthermore, the epicuticular wax layer of leaves erodes over time (Beatie 2002). Such erosion could produce different microenvironments on the leaf surface and eroded surfaces may also provide additional physical niches for bacterial attachment.

Figure 2.

Figure 2

Open in a new tab

Scanning electron micrographs of the upper surfaces of the fifth true leaf of lettuce cultivar Iceberg (a) and accession L. serriola US96UC23 (b), showing different wax layer structures in L. serriola compared to the other two cultivars as typified by cv. Iceberg. The white arrows indicate stomata. The dark lines are the periphery of individual cells.

In summary, different factors appear to influence the attachment of Salmonella to lettuce leaves and subsequent proliferation, although some factors (e.g. venation) may be common to both phases of colonization. There may also be considerable serovar-specific differences in attachment mechanisms and it may not be possible to generalize from one serovar to others. In respect of S. enterica ser. Senftenberg and lettuce leaves, the hydrophobic properties of the epicuticular wax layer appears to be significant, although wax layer composition or the level of wax erosion (or a combination of both) may be more important than the total amount of wax on the leaf surface.

All of these features will have some level of underlying genetic control. The fact that the lines used in this study are the parents of genetic mapping populations will facilitate detailed investigations of the underlying genetics of epicuticular wax and other leaf surface characteristics. Determination of quantitative trait loci (QTL) for these characteristics and the use of transcriptomics technology to identify up- or downregulated genes should allow specific targets to be identified. In parallel, QTL for attachment of Salmonella (or other pathogens of interest such as E. coli) may also be beneficial in targeting future genetic investigations. Since both the genetic mapping populations and a lettuce genome sequence are in the public domain, candidate genes underlying any QTL could be identified and the associated plant biochemical pathways investigated.

One sentence summary.

This paper investigates the effect of lettuce leaf morphology and biochemistry on the surface attachment of the zoonotic pathogen Salmonella enterica.

Acknowledgements

We thank Josie Brough for the technical support, Daniel Eastward for advice on sugar analysis and Cedric Berger for the preliminary investigation.

Funding

This work was supported by the Biotechnology and Biological Sciences Research Council (Grant number BB/G014175/2).

Footnotes

Conflict of interest. None declared.

References

  1. Bakker MI, Baas WJ, Sijm DTHM, et al. Leaf wax of Lactuca sativa and Plantago major. Phytochemistry. 1998;47:1489–93. [Google Scholar]
  2. Barak JD, Gorski L, Naraghi-Arani P, et al. Salmonella enterica virulence genes are required for bacterial attachment to plant tissue. Appl Environ Microb. 2005;7:5685–91. doi: 10.1128/AEM.71.10.5685-5691.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barak JD, Liang A, Narm K-E. Differential attachment to and subsequent contamination of agricultural crops by Salmonella enterica. Appl Environ Microb. 2008;74:5568–70. doi: 10.1128/AEM.01077-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barthlott W, Nienhuis C, Cutler D, et al. Classification and terminology of plant epicuticular waxes. Bot J Linn Soc. 1998;126:237–60. [Google Scholar]
  5. Beatie GA. In: Phyllosphere Microbiology. Lindow E, Hecht-Poinar EI, Elliot VJ, editors. APS Press; St. Paul MN: 2002. Leaf surface waxes and the process of leaf colonization by microorganisms; pp. 3–26. [Google Scholar]
  6. Berger CN, Shaw RK, Brown DJ, et al. Interaction of Salmonella enterica with basil and other salad leaves. ISME J. 2009;3:261–5. doi: 10.1038/ismej.2008.95. [DOI] [PubMed] [Google Scholar]
  7. Bernfeld P. Amylases α and β. Methods Enzymol. 1955;1:149–58. [Google Scholar]
  8. Beuchat LR, Brackett RE. Survival and growth of Listeria monocytogenes on lettuce as influenced by shredding, chlorine treatment, modified atmosphere packaging and temperature. J Food Sci. 1990;55:755–8. [Google Scholar]
  9. Brandl MT. Fitness of human enteric pathogens on plants and implications for food safety. Ann Rev Phytopathol. 2006;44:367–92. doi: 10.1146/annurev.phyto.44.070505.143359. [DOI] [PubMed] [Google Scholar]
  10. Brandl MT, Amundson R. Leaf age as a risk factor in contamination of lettuce with Escherichia coli O157:H7 and Salmonella enterica. Appl Environ Microb. 2008;74:2298–306. doi: 10.1128/AEM.02459-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Brandl MT, Mandrell RE. Fitness of Salmonella enterica serovar Thompson in the Cilantro phyllosphere. Appl Environ Microb. 2002;68:3614–21. doi: 10.1128/AEM.68.7.3614-3621.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cevallos-Cevallos JM, Gu G, Danyluk MD, et al. Adhesion and splash dispersal of Salmonella enterica Typhimurium on tomato leaflets: effects of rdar morphotype and trichome density. Int J Food Microbiol. 2012;160:58–64. doi: 10.1016/j.ijfoodmicro.2012.09.021. [DOI] [PubMed] [Google Scholar]
  13. Grunwald I, Rupprecht I, Schuster G, et al. Identification of guttation fluid proteins: the presence of pathogenesis-related proteins in non-infected barley plants. Physiol Plantarum. 2003;119:192–202. [Google Scholar]
  14. Hunter PJ, Hand P, Pink D, et al. Both leaf properties and microbe-microbe interactions influence within-species variation in bacterial population diversity and structure in the lettuce (Lactuca species) phyllosphere. Appl Environ Microb. 2010;76:8117–25. doi: 10.1128/AEM.01321-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Islam M, Morgan J, Doyle MP, et al. Persistence of Salmonella enterica serovar Typhimurium on lettuce and parsley and in soils on which they were grown in fields treated with contaminated manure composts or irrigation water. Foodborne Pathog Dis. 2004;1:27–35. doi: 10.1089/153531404772914437. [DOI] [PubMed] [Google Scholar]
  16. Jetter R, Schäffer S. Chemical composition of the Prunus laurocerasus leaf surface. Dynamic changes of the epicuticular wax film during leaf development. Plant Physiol. 2001;126:1725–73. doi: 10.1104/pp.126.4.1725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Klerks MM, Franz E, van Gent-Pelzer M, et al. Differential interaction of Salmonella enterica serovars with lettuce cultivars and plant-microbe factors influencing the colonization efficiency. ISME J. 2007;1:620–31. doi: 10.1038/ismej.2007.82. [DOI] [PubMed] [Google Scholar]
  18. Knutton S. Electron microscopical methods in adhesion. Methods Enzymol. 1995;253:145–58. doi: 10.1016/s0076-6879(95)53015-0. [DOI] [PubMed] [Google Scholar]
  19. Lang MM, Harris LJ, Beuchat LR. Survival and recovery of Escherichia coli O157:H7, Salmonella and Listeria monocytogenes on lettuce and parsley as affected by method of inoculation, time between inoculation and analysis and treatment with chlorinated water. J Food Protect. 2004;67:1092–103. doi: 10.4315/0362-028x-67.6.1092. [DOI] [PubMed] [Google Scholar]
  20. Leben C. Relative humidity and the survival of epiphytic bacteria with buds and leaves of cucumber plants. Phytopathology. 1988;78:179–85. [Google Scholar]
  21. Little CL, Gillespie IA. Prepared salads and public health. J Appl Microbiol. 2008;105:1729–43. doi: 10.1111/j.1365-2672.2008.03801.x. [DOI] [PubMed] [Google Scholar]
  22. Mariano RLR, McCarter SM. Epiphytic survival of Pseudomonas viridiflava on tomato and selected weed species. Microb Ecol. 1993;26:47–58. doi: 10.1007/BF00166029. [DOI] [PubMed] [Google Scholar]
  23. Mercier J, Lindow SE. Role of leaf surface sugars in colonization of plants by bacterial epiphytes. Appl Environ Microb. 2000;66:369–74. doi: 10.1128/aem.66.1.369-374.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Monaghan JM. United Kingdom and European approach to fresh produce food safety and security. HortTechnology. 2006;16:559–62. [Google Scholar]
  25. Seo KH, Frank JF. Attachment of Escherichia coli O157:H7 to lettuce leaf surface and bacterial viability in response to chlorine treatment as demonstrated by using confocal scanning laser microscopy. J Food Protect. 1999;62:3–9. doi: 10.4315/0362-028x-62.1.3. [DOI] [PubMed] [Google Scholar]
  26. Shaw RK, Berger CN, Feys B, et al. Enterohemorrhagic Escherichia coli exploits EspA filaments for attachment to salad leaves. Appl Environ Microb. 2008;74:2908–14. doi: 10.1128/AEM.02704-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Shepherd RW, Bass WT, Houtz RL, et al. Phylloplanins of tobacco are defensive proteins deployed on aerial surfaces by short glandular trichomes. Plant Cell. 2005;17:1851–61. doi: 10.1105/tpc.105.031559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Stine SW, Song I, Choi C, et al. Effect of relative humidity on preharvest survival of bacterial and viral pathogens on the surface of cantaloupe, lettuce, and bell peppers. J Food Protect. 2005;68:1352–8. doi: 10.4315/0362-028x-68.7.1352. [DOI] [PubMed] [Google Scholar]
  29. Teusnik R, Rahman M, Bressan RA, et al. Cuticular waxes on Arabidopsis thaliana close relatives Thellungiella halophila and Thellungiella parvula. Int J Plant Sci. 2002;163:309–15. [Google Scholar]
  30. Uhl D, Mossberger V. Leaf venation density as a climate and environmental proxy: a critical review and new data. Palaeogeogr Palaeocl. 1999;149:15–26. [Google Scholar]
  31. Van Volkenburgh E. Leaf expansion—an integrating plant behaviour. Plant Cell Environ. 1999;22:1463–73. [Google Scholar]
  32. Warriner K, Ibrahim F, Dickinson M, et al. Internalization of human pathogens within growing salad vegetables. Biotechnol Genet Eng. 2003;20:117–34. doi: 10.1080/02648725.2003.10648040. [DOI] [PubMed] [Google Scholar]
  33. Young SA, Guo A, Guikema JA, et al. Rice cationic peroxidase accumulates in xylem vessels during incompatible reactions with Xanthomonas oryzae pv. oryzae. Plant Physiol. 1995;107:1333–41. doi: 10.1104/pp.107.4.1333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Zhang Q, Zhang J, Shen J, et al. A simple 96-well microplate method for estimation of total polyphenol content in seaweeds. J Appl Phycology. 2006;18:445–50. [Google Scholar]

RESOURCES