Specific Responses of Salmonella enterica to Tomato Varieties and Fruit Ripeness Identified by In Vivo Expression Technology
CITATION SEARCH THUMBNAILS PDF VIEWER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00103212/00001
 Material Information
Title: Specific Responses of Salmonella enterica to Tomato Varieties and Fruit Ripeness Identified by In Vivo Expression Technology
Series Title: PLoS One
Physical Description: Archival
Language: English
Creator: Noel, Jason T.
Arrach, Nabil
Alagely, Ali
McClelland, Michael
Teplitski, Max
Publisher: Public Library of Science
Publication Date: August, 2010
 Subjects
Genre:
Spatial Coverage:
 Notes
Abstract: Background: Recent outbreaks of vegetable-associated gastroenteritis suggest that enteric pathogens colonize, multiply and persist in plants for extended periods of time, eventually infecting people. Genetic and physiological pathways, by which enterics colonize plants, are still poorly understood. Methodology/Principal Findings: To better understand interactions between Salmonella enterica sv. Typhimurium and tomatoes, a gfp-tagged Salmonella promoter library was screened inside red ripe fruits. Fifty-one unique constructs that were potentially differentially regulated in tomato relative to in vitro growth were identified. The expression of a subset of these promoters was tested in planta using recombinase-based in vivo expression technology (RIVET) and fitness of the corresponding mutants was tested. Gene expression in Salmonella was affected by fruit maturity and tomato cultivar. A putative fadH promoter was upregulated most strongly in immature tomatoes. Expression of the fadH construct depended on the presence of linoleic acid, which is consistent with the reduced accumulation of this compound in mature tomato fruits. The cysB construct was activated in the fruit of cv. Hawaii 7997 (resistant to a race of Ralstonia solanacearum) more strongly than in the universally susceptible tomato cv. Bonny Best. Known Salmonella motility and animal virulence genes (hilA, flhDC, fliF and those encoded on the pSLT virulence plasmid) did not contribute significantly to fitness of the bacteria inside tomatoes, even though deletions of sirA and motA modestly increased fitness of Salmonella inside tomatoes. Conclusions/Significance: This study reveals the genetic basis of the interactions of Salmonella with plant hosts. Salmonella relies on a distinct set of metabolic and regulatory genes, which are differentially regulated in planta in response to host genotype and fruit maturity. This enteric pathogen colonizes tissues of tomatoes differently than plant pathogens, and relies little on its animal virulence genes for persistence within the fruit.
Funding: Publication of this article was funded in part by the University of Florida Open-Access publishing Fund. In addition, requestors receiving funding through the UFOAP project are expected to submit a post-review, final draft of the article to UF's institutional repository, IR@UF, (www.uflib.ufl.edu/ufir) at the time of funding. The Institutional Repository at the University of Florida (IR@UF) is the digital archive for the intellectual output of the University of Florida community, with research, news, outreach and educational materials
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 10.1371/journal.pone.0012406
System ID: UF00103212:00001

Downloads

This item has the following downloads:

( PDF )


Full Text


OPEN ACCESS Freely available online


Specific Responses of Salmonella enterica to Tomato

Varieties and Fruit Ripeness Identified by In Vivo

Expression Technology

Jason T. Noel1, Nabil Arrach2'3, Ali Alagely', Michael McClelland2'4, Max Teplitski1*
1 Soil and Water Science Department, Genetics Institute, University of Florida, Gainesville, Florida, United States of America, 2 Vaccine Research Institute of San Diego, La
Jolla, California, United States of America, 3Sanford-Burnham Medical Research Institute, La Jolla, California, United States of America, 4 Department of Pathology and
Laboratory Medicine, University of California Irvine, Irvine, California, United States of America


Abstract

Background: Recent outbreaks of vegetable-associated gastroenteritis suggest that enteric pathogens colonize, multiply
and persist in plants for extended periods of time, eventually infecting people. Genetic and physiological pathways, by
which enterics colonize plants, are still poorly understood.

Methodology/Principal Findings: To better understand interactions between Salmonella enterica sv. Typhimurium and
tomatoes, a gfp-tagged Salmonella promoter library was screened inside red ripe fruits. Fifty-one unique constructs that
were potentially differentially regulated in tomato relative to in vitro growth were identified. The expression of a subset of
these promoters was tested in plant using recombinase-based in vivo expression technology (RIVET) and fitness of the
corresponding mutants was tested. Gene expression in Salmonella was affected by fruit maturity and tomato cultivar. A
putative fadH promoter was upregulated most strongly in immature tomatoes. Expression of the fadH construct depended
on the presence of linoleic acid, which is consistent with the reduced accumulation of this compound in mature tomato
fruits. The cysB construct was activated in the fruit of cv. Hawaii 7997 (resistant to a race of Ralstonia solanacearum) more
strongly than in the universally susceptible tomato cv. Bonny Best. Known Salmonella motility and animal virulence genes
(hilA, flhDC, fliF and those encoded on the pSLT virulence plasmid) did not contribute significantly to fitness of the bacteria
inside tomatoes, even though deletions of sirA and motA modestly increased fitness of Salmonella inside tomatoes.

Conclusions/Significance:This study reveals the genetic basis of the interactions of Salmonella with plant hosts. Salmonella
relies on a distinct set of metabolic and regulatory genes, which are differentially regulated in plant in response to host
genotype and fruit maturity. This enteric pathogen colonizes tissues of tomatoes differently than plant pathogens, and
relies little on its animal virulence genes for persistence within the fruit.

Citation: Noel JT, Arrach N, Alagely A, McClelland M, Teplitski M (2010) Specific Responses of Salmonella enterica to Tomato Varieties and Fruit Ripeness Identified
by In Vivo Expression Technology. PLoS ONE 5(8): e12406. doi:10.1371/journal.pone.0012406
Editor: Niyaz Ahmed, University of Hyderabad, India
Received March 20, 2010; Accepted July 26, 2010; Published August 31, 2010
Copyright: 2010 Noel et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was funded by Florida Tomato Council and by the Center for Produce Safety, University of California, Davis and the California Department
of Food and Agriculture through Specialty Crop Block Grant Program grant # SCB09052 under CRIS project # 0208876 of Florida Agricultural Experiment Station
(MT, JTN). MM was supported, in part, by BARD IS-4267-09 and National Institutes of Health grant R01 A1075093-01. NA was supported by Tobacco-Related Disease
Research Program grant 16KT-0045. AA acknowledges support from the University of Florida Scholars Program. The funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: This study was funded, in part, by Florida Tomato Council. However, the funders had no role in technical design of the study (beyond
selecting the proposal through a peer-review mechanism). The funders had no role and sought no role in the subsequent experimentation, manuscript
preparation or decision to submit.
SE-mail: maxtep@ufl.edu


Introduction

The increase in produce-associated gastroenteritis outbreaks
indicates that non-typhoidal serovars of Salmonella enterica and
enterovirulent E. coli can contaminate fruit, vegetables and sprouts
[1-3]. S. enterica (including sv. Typhimurium 14028) and
enterovirulent E. coli were found to colonize internal tissues of
tomato, lettuce, alfalfa, cilantro, where they reach population
levels as high as 105-107 cfu/g of plant tissue under field and/or
laboratory conditions [4-7]. These populations are approximately
two orders of magnitude lower than those reached by dedicated
l.1, I. .,1.._. .,. like Pectobactertum carotorovum [7]. These observa-
tions indicate that under most biologically relevant conditions,


PLoS ONE | www.plosone.org


Salmonella can efficiently utilize nutrients found within plant tissues.
However, behaviors and nutrition of enterics inside plants are not
well understood.
Unlike closely related plant-associated members of the Entero-
bacteriaceae, Salmonella enterica and E. coli do not produce enzymes
capable of degrading plant cell wall polymers and their genomes
do not encode homologs of known pectinases or hemicellulases
[8,9]. Therefore, to proliferate inside plant tissues, these enteric
]. ..1. ., may rely on their ability to acquire simpler carbon
sources. This hypothesis was recently supported by a transcrip-
tomic analysis of the E. coli 0157:H7 _1.. 11 on lettuce leaf lysates
[10]. Tomato fruits contain simple sugars, sugar alcohols, organic
and fatty acids, and amino acids, which could be utilized by


August 2010 1 Volume 5 1 Issue 8 | e12406


*0*PLoS one







Salmonella-Tomato Interactions


Salmonella. The amounts of these metabolites differ among cultivars
and also depend on the maturity stage of fruit [11,12].
To formulate a hypothesis about behavior of S. enterica within
plant tissues, we analyzed reports of gene regulation in plant
1]. 1,.._...., during their interactions with plant hosts. In plant gene
expression has been documented for Xanthomonas campestris pv
vesicatoria during formation of a leaf spot on tomato [13], Ralstonia
solanacearum during bacterial wilt of tomato [14], the fire blight
.il,1.... .. Erwinia amylovora during rot of immature pear fruits [15],
Pseudomonas syringae pv syringae during colonization of bean
phyllosphere [16] and P. sringae pv tomato during _,1. 11. inside
tissues of Arabidopsis thaliana [17]. Even 1 ... _1. screen conditions
differed, all of these .1i, ... ,11... .. were found to express a
common set of -5-15% of their genes during interactions with
plant hosts. These commonly expressed genes included those for
type III secretion systems and effectors translocated by them, as
well as genes involved in the degradation of plant polymers. A
significant number of differentially regulated genes were involved
in stress resistance and responses to antibiotics, uptake of iron,
amino and carboxylic acids and synthesis of amino acids and
proteins [13,14,17]. Therefore, if Salmonella behaves similarly to
plant-associated bacteria during colonization of plant tissues, it is
reasonable to expect differential regulation of genes corresponding
to similar functions.
A few of the genes and mechanisms that enteric bacteria use to
colonize external surfaces of host plants have been identified.
Bacterial polymers (cellulose, poly-P-1,6-N-acetyl-D-glucosamine,
colanic acid and O-antigen capsule) and aggregative fimbriae are
involved in the attachment of E. coli and/or Salmonella to plant
surfaces [18-23]. In their reliance on self-produced cellulose for
attachment to plant surfaces, enteric ]. ,1i..... are similar to the
plant symbiotic and ].111.... ,;. bacteria but differ from the
interactions of Salmonella with animal tissues, where cellulose
production is dispensable [24].
Understanding the behavior and nutrition of non-typhoidal
Salmonella enterica inside plants may reveal ecological strategies this
enteric ] ,11._. .. uses to persist outside its animal hosts. Defining
differences in the Salmonella survival in different tomato cultivars or
maturity stages may offer opportunities to modify pre- or post-
harvest practices for promoting safer produce.


Results

Salmonella enterica sv. Typhimurium 14028 as a model
In a preliminary study, we tested whether S. enterica sv.
Typhimurium ATCC 14028 would serve as a useful model for
this study. Several serovars of non-typhoidal S. enterica have been
linked to multi-state outbreaks of produce-borne salmonellosis,
however only one isolate of S. enterica sv. Typhimurium was linked
to a "tomato" outbreak [25]. S. enterica sv Typhimurium was the
prevalent (91%) serovar isolated from fruit rinses, irrigation and
drinking water in a comprehensive survey of multiple farms in
Mexico [26]. Collectively, these reports suggest that studies using
this common serovar may have ecological significance. Even
1 .... _1. genome sequences are becoming available for a number of
the outbreak strains, the genetic and genomic tools available for S.
enterica sv Typhimurium ATCC 14028 are unmatched [27].
To test whether ATCC 14028 colonizes tomato fruit tissues as
well as the strains isolated from the salmonellosis outbreaks linked
to the tomato consumption, proliferation of this strain inside red
ripe tomatoes (cv. Campari) was compared with that of the six-
strain cocktail containing of S. enterica sv. Newport, Braenderup,
Javiana and Montevideo, which were isolated from tomato-related


salmonellosis outbreaks or from tomato fields on the Eastern Shore
of Virginia (Table Sl).
As shown in Fig. 1, S. enterica sv. Typhimurium ATCC 14028,
the model for this research, reached approximately 106 c.f.u./g of
tomato tissue, and grew inside tomatoes to the same final densities
as the outbreak strains (Fig. 1). These observations are consistent
with the report that another isolate of S. Typhimurium colonized
tomato fruits similarly to the strains, which have been linked to the
gastroenteritis outbreaks [28]. While it is likely that there exist
strain-specific differences in the regulation of some Salmonella genes
during colonization of tomatoes, the ability of ATCC 14028 to
proliferate inside tomatoes as efficiently as the outbreak strains
indicates that the data obtained using this model will be useful in
understanding the role of non-host environments in the ecology of
Salmonella.

Survey of Salmonella genes induced in tomato fruit
To identify promoters that are strongly regulated in Salmonella
during colonization of tomato fruits, two promoter-gfp libraries
were screened and tomato-specific constructs were isolated. Two
hundred eighty-eight clones containing differentially regulated
promoters were identified and sequenced. Fifty-one unique
fragments differentially regulated during _1.. 11. inside tomato
fruit were identified (Fig. 2). Six concatemer constructs were
identified and discarded (data not shown).
To validate results of the FACS sort, promoter-tnpR RIVET
reporters in several representative genes were constructed (Fig. 2).
TnpR-based RIVET reporters have been shown to be sensitive
and reliable for quantifying Salmonella gene expression in animal
and plant hosts, including tomato [7,29]. Regulation of the
promoters of interest in tomato fruits of cultivars harvested at two
different maturity stages was tested. Below, we will consider only
those select reporters that were validated by RIVET assays.

The promoter adjacent to fadH is differentially expressed
in green vs red fruits
The resolution of the RIVET reporter (S. enterica sv. Typhimur-
ium JTN24) in the promoter adjacent to fadH was strongly
reduced in red ripe tomatoes compared to the soft agar LB
(Fig. 3A). Consistent with the FACS experiments, there was no
resolution of the construct in LB broth (data not shown). ThefadH
gene encodes 2,4-dienoyl-CoA reductase, an iron-sulfur flavoen-
zyme required for the metabolism of unsaturated fatty acids with
double bonds at even carbon positions [30]. In fruits that are green
when immature and turn red when mature (tomato cvs. Hawaii
7997, Bonny Best; jalapeno and red bell pepper), the resolution of
the fadH RIVET reporter was ;_i,;. ,,11 higher in green fruit
compared to red ripe fruit. Interestingly, in bell pepper cv.
Diamond (ivory when immature, red when mature), resolution of
the reporter was strong in mature and immature fruit (Fig. 3A). In
fruit of tomato and bell pepper that are green when immature and
yellow when mature, expression was low and not statistically
different between fruits of different cultivars (Fig. 3A). The
reporter was not resolved on surfaces of leaves of tomato cvs
Bonny Best or Hawaii 7997 (data not shown). These observations
suggested that the reporter responds either positively or negatively
to the presence of chemical cues in fruit at different maturity
stages.
To test the hypothesis that the expression of the promoter
upstream of fadH depended on the presence of carotenoid
pigments, which accumulate as the fruits ripen and turn red,
hydrophobic pigments were extracted with chloroform. Resolution
of the reporter was tested in soft LB agar (0.3%) supplemented
with the tomato extract. No difference in the activity of the


August 2010 1 Volume 5 1 Issue 8 | e12406


,.a ', PLoS ONE I www.plosone.org






Salmonella-Tomato Interactions


Growth of Salmonella in tomatoes


1 E+7

51E+6

S1E+5

S1E+4

S1E+3

S.j...


2 4 6 8 10
Days post-infection


Figure 1. Growth of Salmonella enterica strains inside red ripe tomatoes. S. enterica sv. Typhimurium ATCC 14028 or a six-strain cocktail
containing S. enterica sv. Newport C6.3, sv. Braenderup 04E01347, Braenderup 04E01556, Braenderup 04E00783, Montevideo LJH519 and sv. Javiana
ATCC BAA-1593 were seeded onto shallow wounds in red ripe tomatoes of cv. Campari. Tomatoes were incubated at 22C, 40-60% relative humidity.
At indicated time points, three tomatoes per treatment were harvested, blended for 1 min at 260 r.p.m. in Stomacher 400 Circulator (Seward, West
Sussex, U.K.), and aliquots were plated on XLD medium. Averages of the three samples are shown, error bars are standard errors.
doi:10.1371/journal.pone.0012406.g001


reporter was observed in response to the hydrophobic extract of
red ripe tomatoes: 80 7% (t standard error) in soft LB i_ ,ii I,.
tomato extract vs 74--6% in soft LB agar. This indicates that it
was not non-polar compounds found in ripe tomatoes, which
down-regulated the reporter. Rather, it is likely to be a substance
present in green tomatoes that activated the reporter.

Linoleic acid upregulates Salmonella promoter adjacent
to fadH
In red ripe fruit, sucrose content decreases while glucose
increases. Palmitic and linoleic acids, myo-innositol, mannitol,
glycine, arginine and arabinose are found in higher amounts in
immature fruit (compared to a red ripe fruit) [11]. To test whether
any of these compounds affect expression of thefadH+fadH::tnpR
RIVET reporter JTN24, bioassays were carried out with each of
these compounds spotted on glass fiber disks and placed on the
surface of soft agar based on M9 minimal medium seeded with the
reporter. As an additional control, one of the assays contained
glycerol (its concentration in the tomato fruit does not change as
the fruit matures [11]). As shown in Fig. 3B, the basal level of the
reporter resolution in M9 soft agar was approximately 20%. The
addition of sucrose (which is not metabolized by S. enterica [31]),
mannitol or glycerol did not affect regulation of the reporter
(Fig. 3B). Partial induction was observed in the presence of other
tomato metabolites (arabinose, arginine, glucose, glycine, myo-
innositol and palmitate) known to differentially accumulate in
green (and not ripe) tomatoes. The addition of linoleic acid fully
induced the reporter (Fig. 3B) indicating that this is likely the cue
that triggers the expression of thefadH RIVET reporter in green,
and not red ripe tomatoes.

Tomato genotype contributes to the regulation of
Salmonella cysB
A fragment spanning the cysB promoter [32] was identified as
differentially regulated inside tomatoes (Fig.2). Consistent with
this, resolution of the cysB RIVET reporter (S. enterica sv.
TyphimuriumJTN71) in LB broth was 2 2% and was increased


PLoS ONE | www.plosone.org


in tomato fruits (Fig. 4). No resolution of the reporter was observed
in the phyllosphere of cvs. Bonny Best or Hawaii 7997 (data not
shown). The cysB gene encodes a regulator of an operon involved
in cysteine acquisition [32,33]. Cysteine accumulates in red ripe
tomatoes [11], therefore we tested whether the expression of the
cysB RIVET reporter would depend on the maturity of the fruit.
As shown in Fig. 4, regulation of the cysB reporter was not affected
by maturity of the fruit; rather it depended on the cultivar in which
the reporter was tested. The expression of cysB was highest in the
fruit of cv. Hawaii 7997 which is resistant to a race of a vascular
1. ,1.... ., R. solanacearum [34]. Expression of the cysB reporter was
lower in the fruits of cvs. Campari and Bonny Best which are both
susceptible to a number of bacterial and fungal .i,11..._. ., The
correlation between the expression of cysB and tomato disease-
resistant genotype is intriguing because one of the ysB-dependent
phenotypes in S. enterica is resistance to antibiotics ,l...l. the
mechanism by which L-cysteine and cysB effect this behavior is not
entirely clear) [33]. Interestingly, genes involved in cysteine
synthesis were differentially regulated in P. syringae on bean leaf
surfaces [16] and in E. coli 0157:H7 in response to lettuce leaf
lysates [10]. These observations suggest a conserved function for
the cysteine synthesis genes in the interactions of enterics and plant
] ,11.... with their plant hosts.

Regulation of STM2006
STM2006 is predicted to encode a DN \-I.1..I .. prophage
protein, which is found in the genomes of some salmonellae, with a
close homologue in Pectobacterium carotovorum, a plant-associated
member of Enterobacteriacea. Expression of the STM2006 RIVET
reporter (S. enterica sv. Typhimurium JTN69) was 3-to-4 fold
higher inside red or green tomatoes of three cultivars compared to
the LB control (Fig. 5). No resolution of the reporter was observed
on tomato leaf surfaces (data not shown). Because .i11,.1.. of
STM2006 are annotated as amino acid transporters in some
Salmonella published genomes, we tested whether supplementation
of MAI',-i .... medium with any of the 20 amino acids or y-
aminobutyric acid would mimic the increased expression observed


August 2010 1 Volume 5 1 Issue 8 | e12406


- S. enterica ATCC 14028
-o- 6-strain cocktail








Salmonella-Tomato Interactions


Coordlnateseofthe
probe


3395984: 3396218
1053294:1052870
2719365:2719401
78613:78275
907104:306821
192906:193228
3459260:3458922
3470433:3470070
2042873:2043033
3822098 :3822438
2623544:2623107
3902673:39028431
1846824:19467711



4115289:4114917
1345464:1345774
1794777:1794415
4363851:4363539
3256081:3256392
3454323:3453897
2491412:2491626



1809738:1309569'
1349088:1349259'
216377:216623

3744733:3744305
399110939W916
3816358:3316185
1343730:1343380


Nearest Gene


Putative Function of the Nearest Ge


General Metabolis

,fadH(STM3219) NADI dehydmrenase
pfL (STMO97O) pyruvatc formatcelyase
STM2373 ddhydroanoatc rcducd w
carD (STM0067) phosphate synthua
sucD(STM0739) succinyl-CoA synthase
pdrA(STMOI63) 4-hydrxythrconine-4-phosphatw dehydrog.
ariG (STM3290) argininosuccinatc synthase
obRE (STM33OI) GTP-bindinR protein
yedO(STM 953) D-cystcinc dcultbydrase
)*jW(STM3635) phosphoewlunolamineg mransfra
gua. (STM251I) purine metabolism
yibD(STM3707) pulati glycosyl transferasc
tdk (STM4514) thymidine kinase
Protein synthesis andddegradtllo,.

.vA (STM3905) threonint dehydratase
STM1262 tRNA
rnb (STM 1702) exoribonucle It
rplA (STM4150) SOS ribosoma protein
)WI (STM3094) 16S ribosomal RNA methyIetransfieric
iWR (STM3286) ranslauion initiation factor
mnmC (STM2379) 5-methylaninonmthyl-2-thiouridine methyltans.
Regulation


cysB(STM 713)
S1M1266
dksA (STMOI86)

zWlA (STM3576)
uhpA (S[M3790)
dpPA (STM3630)
STMI259


regulator orcysteine operon ad antibiotic Mist
putative tanseriptional regulator


transcriptional rerulator
Tronaadr*
PbCdln/Hg transporter
sugar uptake r-gior
dipeplide transport
ABC-type transport system ATPase component


20953261085823 STM2006 amino acid transport
Klrukhmv
3634465:3634915 bi*4 (STM3478) pulutive surface-exposd virulence protein
anti-sensefliQ message contributes to regulation
2W6259:2065922 flip (STM 1980) vi'virulcrice and miotity genes
Attachment


3827327:3827837
379920:379567
3802370.3802035




1920125:1919777

3391851 :3392147

2452240:2452657
4769022:4769971'
4437196:4437420
2513789:2513620'



2756267:2756544
2737946:2737442'
27476162747165
1135727:1136132'
4697078 :4697376

1713839:1714224
2423218:2422856

4672486:4672970
116633:116225'
95774:95462
95126:95297'



PLoSONEI


lpfA (STM3640)
stbE (STM0336)
bcMA (STM3619)


Probe
Orientation
Comments

I1 1l RIVET reporter differentially regulated in
Tomatoes












-g

loos as 4-


RIVET reporter not resolved In LB soft
agar or tomatoes









RIVET reporter dilRerentially regulated in
tomato









SRIVET rpepter diffemretially riguiated in
t tomato



m I RIVET repoter resolved at -13% in soft
,:= agar LB and tomftos


I.


long polar flmbria precursor
chaperone. other for Type I fimbriac
cellulose synthase


Stress response

yoaA (STM 12 1) SOS repair hclicase

j*T(STM3224) tellurit resistance
Cel envweipr
3)EC (STM2339) putative inner membrane protein
)4iH (STM4514) putative inner membrane protein
STM4218 putative inner membrane protein
STM2400 putative inner membrane protein


SIM2608
STM25B9
STM2597
STM1049
ImB (STh4454)

STM1624
"bM(STM2315)


ExtMulwimrosomanled elments, seudogmei s

teniinase.-ike iarge protein
specificity protein.J.Iik
major tail-like protein
phape tail ritcr prokin
psuedogene
WIkiwwnwfwwlots
putative cytoplasmic protein
hypodieiical protein


STM4429 unknown


STMOO98
STMQOM2
STMMOOS


RlVEY reporter di Terentially regulated in
tomato


4r= RIVET reporter not resolved in soft agar
LB or tomatoes
RIVET repohcr flly resolved in soft agar
LB and tomatoes


RIVET reporter not rsolved in soft agar
LB or tomatoes









RIVEt'repoer flt resolved in soft agar
LB or tomatoes


a_-
as>"


unknown
unknown
unknown


August 2010 1 Volume 5 1 Issue 8 1 e12406


www.plosone.org






Salmonella-Tomato Interactions


Figure 2. Probes differentially regulated inside red ripe tomato. Fragments of Salmonella genomic DNA which effected differential
accumulation of the GFP reporter protein were identified by sequencing, their coordinates are listed in the far left column. Coordinates are based on
the S. enterica sv Typhimurium genome sequence, GenBank accession # NC_003197. The position and orientation of the probe are indicated as an
unfilled arrow. The nearest annotated ORFs are shown as grey or black arrows. Sequences indicated with '1' occur multiple times within the
Salmonella genome; sequences indicated with '2' occur twice in the Salmonella genome.
doi:10.1371/journal.pone.0012406.g002


in tomatoes. None of tested amino acids (at 0.5-1 mM) affected
expression of the reporter in STM2006 to an appreciable degree
(data not shown). Therefore, while STM2006 is clearly upregu-
lated inside tomatoes, the mechanism of this differential regulation
is not yet clear.

Cryptic genes
The analysis of the genomic context revealed that only eleven of
the clones identified by FACS included known promoters (Fig. 2).
The other constructs represent intergenic regions or "cryptic"
genes, in which a differentially regulated fragment overlaps the
ORF or is in the opposite orientation (Fig. 2). This was not
unexpected. Such cryptic constructs are commonly identified using
in vivo expression analyses and represent a sizable proportion of the
in vivo expressed genes [13,14,16,35]. We do not yet know whether
these sequences are simply "noise" inherent to l.;_1.,1, i ...,,n1
assays or whether these are indications of niche-specific gene
regulation 1.1... _1 alternate mechanisms.
To test whether fragments found within ORFs, including those
in "backwards" orientations, acted as inducible promoters,
chromosomal RIVET reporters were constructed in fliQ ilvA,
STM4429, STM2608 and IpfA (Fig. 2). Regulation of these single
copy reporters was tested in soft LB agar (0.3%) and in tomatoes.
The single copy chromosomal reporters in ilvA, STM4429 and
STM2608 did not resolve inside red ripe tomato or in soft LB agar
(Fig. 2). The RIVET reporter corresponding to the antisensefliQ
transcript was resolved at --13% in soft LB agar and in tomatoes
(Fig. 2). This is consistent with the report by Wang and Harshey
_'1111 which identified low level antisense transcription of the
i i.' cistron [36].
The "backwards" RIVET reporter in IpfA was expressed at
5 3% in LB broth and at 49% 55% in soft LB agar (0.3%) (data
not shown and Fig. 6). Expression of this reporter was reduced
inside tomatoes, the regulation of the cryptic promoter within IpfA
was not affected by the cultivar or maturity of the tomato. The
mechanism of this regulation inside tomatoes is not yet clear, nor is
it clear how important the anti-sense regulation is to the
interactions of Salmonella with its hosts.

Fitness of the mutants in the tomato-dependent genes
We constructed individual deletion mutants infadH, cysB and in
the first genes in the following putative cistrons: STM2006-
STM2008; and IpfABCDE-yhjWproK. We tested competitive fitness
of these and bcs [37] mutants within red ripe tomatoes. As
indicated in Table 1, with the exception of cysB, none of the single
mutants was _i;... ,11 less fit than the wild type in the red ripe
fruit of tomato cv. Campari. Competitive fitness of the AcysB
mutant in fruits of cv. Hawaii 7997 and Bonny Best was not
statistically different from the control (data not shown), and the
competitive fitness of the AcysB mutant was only modestly reduced
in red ripe tomatoes ofcv. Campari. A mutant, in which bcsA, cysB,
fadH, IpfA and STM2006 were all deleted, was modestly, but
statistically ;_..;. ,11 less fit than the wild type in tomatoes
(Table 1), but not in LB shake cultures (data not shown). It is
possible that the relevant genes being regulated by the promoters
were downstream of these proximal genes. Alternatively, or
possibly concurrently, persistence of Salmonella inside tomatoes is


PLoS ONE | www.plosone.org


a complex phenotype, controlled by multiple nutritional and
regulatory inputs that are redundant, making the expression of
individual genes less important.

Role of known virulence, motility and attachment genes
in persistence within fruit
In E. coli 0157:H7, known virulence genes were upregulated
within the first 15-30 minutes of exposure to lettuce leaf exudates
[10]. Salmonella virulence gene were also involved in the
colonization of alfalfa seedlings [1 I. ] however promoters of
the Salmonella virulence genes were largely absent from the list in
Fig. 2. This could be explained by the fact that animal virulence
genes were not differentially regulated inside tomato fruit.
Alternatively, the corresponding constructs may have been
eliminated during the pre-sorting step (after _1., 11, in LB broth
and prior to inoculation into tomatoes) because many virulence
genes are expressed in LB medium and the corresponding
promoter probes would have been eliminated during the pre-
sorting step.
The majority of Salmonella strains recovered from fruit- or
vegetable-related outbreaks produced curli [23]. Genes encoding
aggregative fimbriae (curli) were shown to be involved in binding
of Salmonella and some E. coli strains to alfalfa seedlings and lettuce
leaves [18-22]. However, none of the genes in the agfoperon were
identified in our screen. To test whether aggregative fimbriae are
involved in the persistence of Salmonella within tomatoes, we
constructed a mutant with deletions in the and agfB genes, and
tested the expression of the agf RIVET reporter (S. enterica sv.
Typhimurium JTN204) inside tomatoes. The resolution of the
JTN204 RIVET reporter in LB shake cultures was 46-9%, while
inside red ripe Campari tomatoes the resolution was ;_;i.. ,11
lower (5-2 %). This lack of up-regulation of the reporter inside the
tomato is consistent with the results of the FACS sort, which did
not identify agfpromoter as differentially regulated inside tomato.
Competitive fitness of the mutant lacking agfB and was
indistinguishable from that of the wild type (Table 2), consistent
with the lack of expression of the gene inside tomatoes.
Conversely, a RIVET reporter in the yihT gene (S. enterica sv.
Typhimurium JTN203) encoding an aldolase involved in capsule
synthesis was expressed in red ripe Campari tomatoes (27-I .
but not in LB. However, deletion ofyihTor of the entireyihT-ompL
region did not reduce competitive fitness of the strain in red ripe
tomatoes (Table 2). These results indicate that even 1...n_.
aggregative fimbriae and capsule gene were involved in attach-
ment to plant surfaces in other studies, they are not critical to
persistence within tomatoes.
To test whether known Salmonella virulence genes have a role in
colonization and persistence within tomato fruits, equal numbers of
the wild type Salmonella 14028 and isogenic mutants were inoculated
onto wounded tomato fruits (Table 2). Deletion of the entire
virulence plasmid pSLT had no effect on the competitive fitness of
Salmonella within red ripe tomatoes (Table 2). In addition to the tra
and spy genes, pSLT carries pefand rck operons that are controlled
by the Salmonella N-Acyl homoserine lactone receptor SdiA [39].
The lack of the phenotype for pSLT in these experiments is
consistent with the earlier report that the Salmonella SdiA regulon is
dispensable during persistence within tomato fruits [7].


August 2010 1 Volume 5 1 Issue 8 | e12406






Salmonella-Tomato Interactions


A
Activity of the fadH reporter in tomato and pepper fruits
80-
SLB, 0.3% agar
. 60o mm.aur frruit
2 6Mature fruit


Responses of the fadH reporter to metabolites that
accumulate in tomato fruits


100


80


0
: 60

0
20


20


Figure 3. fadH expression depends on fruit ripeness and a specific metabolite. (A) To test regulation of fadH inside fruits of tomatoes and
peppers, fadH+ fadH::tnpR res1-tet-res reporter was seeded onto shallow wounds in mature and immature fruits of tomatoes and peppers. Activity of
the reporter was scored as the loss of the tetracycline marker in the recovered bacteria ('resolution'). Averages of at least two biological and three
technical replications are shown, error bars are standard errors. Fruits of each cultivar at different maturity stages were collected from the same plant.
In LB broth, resolution of the fadH RIVET reporter was 0%. (B) To test which metabolite from red or green tomatoes affects the fadH RIVET reporter,
resolution of the reporter was measured in soft M9 agar (0.3%) supplemented with the metabolites present in tomato fruits. An upward white arrow
at the bottom of each bar indicates that the metabolite increases as tomato matures, a downward arrow indicates that the metabolite decreases with
the maturity of the fruit [11]. Glycerol is present at the same amounts in red and green fruit [11], and therefore is not marked with an arrow. The
control sample contained no carbon sources. The experiment was repeated twice, with two technical replications. Data from the second experiment
are shown, error bars are standard errors.
doi:10.1371/journal.pone.0012406.g003

In Salmonella, the global regulatory system BarA/SirA-Csr genes csrB and csrC [40-42]. The BarA/SirA-Csr system,
controls invasion genes located on Salmonella P 1i ..._. ,,;. ;1 Island therefore, plays a role in some animal models of virulence and
(SPI)-1, SPI-4 and SPI-5 by activating transcription of small RNA also is crucial for attachment to abiotic surfaces [40,41,43].


SPLoS ONE I www.plosone.org 6 August 2010 1 Volume 5 1 Issue 8 | e12406






Salmonella-Tomato Interactions


O LB, 02% agar
] Immature fruit
E Maturefruit


Activity of the STM2006 reporter in tomatoes
B BC

rc-L -I B_


30 c


A A
4111


A-
440


Figure 4. Tomato cultivar-dependent regulation of Salmonella
cysB. Activity of the chromosomal cysB+ cysB::tnpR res 1-tet-res I reporter
was scored as the loss ('resolution') of tetracycline marker flanked by res
sites in reporter cells recovered from red and green tomatoes of three
cultivars and soft LB agar. Resolution of the reporter in LB broth was
22%. Averages of at least two biological and three technical
replications are shown, error bars are standard errors. Data points that
are significantly different are indicated by different upper case letters;
statistical significance was established by 1-way ANOVA and Tukey's
Honestly Significant Difference (HSD) for post hoc comparison of
means. All statistical tests were performed using JMP 5.1.2 (SAS
Institute, Inc).
doi:10.1371/journal.pone.0012406.g004


Furthermore, the sirA ..111,. 1.._ of Erwinia amylovora was upregu-
lated ,1 ;._ 11.. infection of a pear fruit [15]. Our results indicated
that the sirA mutant was 1.8-fold more competitive than the wild
type in tomato (Table 2). Deletion of both regulatory csr sRNAs
(which are known to be controlled by SirA [41,43,44]) resulted in a
phenotype with a competitive index of 1.57, however this was not
;_ ;,. i,11 different from either the wild type or a sirA mutation.
In an LB broth co-culture with the wild type S. enterica sv
Typhimurium 14028, competitive fitness of the sirA mutant was
statistically indistinguishable from the fitness of a wild type strain
marked with an antibiotic resistance gene in a neutral site of the
genome (data not shown). Deletion of hilA tended to increase the
competitive fitness of the mutant, consistent with the earlier
reports for the function of SPI-1 effectors in endophytic fitness
within alfalfa seedlings [38]. However, the hilA phenotype was
highly variable between biological replicates and therefore was not
statistically ;_..i. ,11 different from the wild type overall
(Table 2).
As an additional control, competitive fitness was also tested for
the mutants in hns, a gene encoding a histone-like protein that is
known to be expressed in tomatoes [7]) and aroA, a gene encoding
5-enolpyruvylshikimate 3-phosphate synthase. Both mutants were
strongly outcompeted by the wild type in tomatoes, consistent with
their pleiotropic phenotypes. However, in LB broth shake cultures
under experimental conditions used in this study, competitive
fitness of the mutants was not statistically different from the
competitive fitness of the wild type marked with an antibiotic
resistance (data not shown).


PLoS ONE | www.plosone.org


9'NoS


4-


*A

0


Figure 5. Activity of the STM2006 RIVET reporter in tomatoes.
cis merodiploid chromosomal RIVET reporter in STM2006 was
upregulated in tomatoes, compared to soft LB agar. Averages of the
two biological and three technical replications are shown; tomatoes of
each cultivar at different maturity stages were harvested from the same
plant at the same time. Data points that are significantly different are
indicated by different upper case letters; statistical significance was
established by 1-way ANOVA and Tukey's Honestly Significant
Difference (HSD) for post hoc comparison of means. All statistical tests
were performed using JMP 5.1.2 (SAS Institute, Inc).
doi:10.1371/journal.pone.0012406.g005

In IVET screens with plant i....i i :1, II i genes have not
been detected as upregulated inside plant tissues [13-17].
Consistent with the reports that bacterial :1 .. II elicit plant
defense responses, :i .. II .1 genes reduced endophytic fitness of S.
enterica sv. Typhimurium 14028 in alfalfa seedlings [38]. To test
whether Salmonella :1 _. II have a function in competitive
persistence within tomato fruits, phenotypes of regulatory and
structural :i _. II mutants were compared. A mutation inflhDC,
the master regulator of the :i _. II 1 regulon had no effect on
competitive fitness within red ripe tomatoes (Table 2). Deletion of
fliFresults in a :.. .. i i_. II ,I. I I.-. ,I ,.I with a functional motor, this
mutant was as competitive as the wild type. Disruption of motA
results in a :1 i.. 1 . I non-motile mutant. The motA mutant
colonized tomato fruit tissues better than the wild type
(competitive index 2.57, Table 2) while in LB shake cultures,
motA mutant was as competitive as the wild type (data not shown).
The conclusion that the presence of a :1 ,.. I I. 1 even if it is not
functional (as in the motA mutant) reduces competitive fitness of
Salmonella in tomatoes is consistent with the reduced endophytic
fitness of :i II 1 ,. I Salmonella strains within alfalfa sprouts [38].

Discussion

Despite improvements in agricultural and management prac-
tices, leafy greens, tomatoes, cucurbits, peppers and nuts were
among the foods linked to outbreaks of gastrointestinal illnesses
caused by Escherichia coli 0157:H7 and non-typhoidal serovars of
Salmonella enterica, causing thousands of hospitalizations and multi-
million dollar damage to the produce industry. This is consistent
with an ability of enterics to colonize plants as alternate hosts [1].


August 2010 1 Volume 5 1 Issue 8 | e12406


Activity of the cysB reporter in tomatoes


100
A
80


o 60


40
12






Salmonella-Tomato Interactions


Activity of the antisense IpfA reporter
in tomatoes


SLB, 0.3% agar
O Immature fruit
SMaturefruit


40 4
,a^s ,^~


Figure 6. Regulation of the RIVET reporter in the internal anti-
sense fragment of IpfA in tomatoes. A chromosomal RIVET reporter
constructed within IpfA and in the antisense orientation was strongly
expressed in soft LB agar and down-regulated in tomatoes, regardless
of the maturity state or the cultivar. In LB broth, the reporter was
resolved at 543%. Averages of the two biological and three technical
replications are shown; tomatoes of each cultivar at different maturity
stages were harvested from the same plant at the same time. Data
points that are significantly different are indicated by different upper
case letters; statistical significance was established by 1-way ANOVA
and Tukey's Honestly Significant Difference (HSD) for post hoc
comparison of means.
doi:10.1371/journal.pone.0012406.g006

However, it is not yet clear whether, how and to what extent
enterics may have evolved to use plants as environmental
reservoirs or alternate hosts.
If non-typhoidal salmonellae have indeed been selected to
utilize plants as alternate hosts, then such evolution should result in
effective colonization of plant tissues by these bacteria and in
specifically altered patterns of bacterial gene expression in
response to chemical signals and nutritional cues encountered
during their colonization of plant tissues. The results presented
here identify a significant set of genes in S. enterica sv.
Typhimurium 14028 that are differentially regulated during
colonization of the internal tissues of the tomato fruit. There
appears to be no appreciable overlap of this set of "tomato-
:. _,i ,. .1" genes with the set of genes in Salmonella required for its
infection and colonization of animals, or with the set of common
genes in 1d. .ii ....._. ;. bacteria required for their infection and
colonization of plant tissues.
The differences in colonization of plant cultivars by different
Salmonella serovars [28,38,45,46] and the differences in responses
of Salmonella RIVET reporters to the tomatoes of different cultivars
(Fig. 4) raise the 1i1,,; _.,;,,_ I... ;1,;1; . i, 1,.. interactions are co-
evolved, with adaptations of each partner to the other. Previous
studies have indicated that plant hosts recognize human enteric
].1i-... ... and initiate defenses to limit colonization by these


PLoS ONE | www.plosone.org


bacteria. For example, PR1, a defense-related protein commonly
used as a reporter of inducible plant defenses, was upregulated
after inoculation with Salmonella in both Arabidopsis and in lettuce
[6,38]. Our results showed that the cysB reporter was differentially
regulated in the tomato cultivars Hawaii 7997 and Bonny Best.
The cysB gene contributes to the resistance of Salmonella swarms to
antibiotics [33]. However, the connections between cysB expres-
sion and mechanisms of resistance in plants are not yet known.
Further characterization of the genetic mechanisms that determine
the specificity of interactions between Salmonella and crop cultivars
may lead to the identification of plant genotypes that are less
conducive to the proliferation of Salmonella within plant tissues and
thus to enhanced food safety.

Materials and Methods

Strain construction
All strains are listed in Table Sl. Deletion mutants were
constructed i, ;-_ 11.. one-step mutagenesis procedure of Datsenko
and Wanner (2000) with primers listed in Tables S1 and S2.
Mutants were confirmed by PCR and then transduced into the
wild type S. enterica sv Typhimurium 14028 using phage P22-
mediated transduction. When necessary, frt-kan-frt cassettes were
removed as in [47], and confirmed by PCR with a set of outside
primers (Table S2). The mutation in cysB was also confirmed
phenotypically: the mutant was a cysteine auxotroph unable to
utilize methionine, consistent with previous reports [48]. To
construct TIM174, AlpfA30, AfadH15, ASTM2006 and AcysB22::
kan mutations were introduced sequentially into JSG1748 using
P22 phage transductions.
To construct RIVET reporters, putative promoters of interest
were PCR amplified using primers specified in Table S2 using
genomic DNA of S. enterica sv Typhimurium 14028 as a template.
PCR products were first cloned into pCR2.1, and then sub-cloned
into pGOA1193 [49]. The resulting constructs were sequenced at
the University of Florida Biotechnology Core facilities, and then
mated into S. enterica sv Typhimurium JS246. Integration was
validated by PCR, ;i._ 1.. "third" primer (see Tables S1, S2). As
indicated in Table S1, some RIVET reporters were constructed
using pCE70 or pCE71, which integrate into the frt scar generated
by the excision of the frt-kan-frt cassette [29].

Plant materials
All assays were initially conducted in unwaxed red ripe tomatoes
cv. Campari purchased at local supermarkets. Follow-up exper-
iments, as indicated in text, were carried out with green or red-ripe
tomatoes of cvs. Bonny Best and Hawaii 7997 grown in a roof-top
greenhouse. Plants were grown from seed in Miracle-Gro Potting
Soil, fertilized biweekly with Miracle-Gro Tomato Plant Food (18-
21-21) (Marysville, OH). Seeds of Hawaii 7997 were from Dr.
Jeffrey B. Jones (Department of Plant P i,.1 ...- University of
Florida). Seeds of cv. Bonny Best were purchased from Millington
Seed Co \Ilii..._i... MI). Peppers and other tomato cultivars
were grown in the field under organic-like conditions on a private
farm outside of Archer, FL, from locally purchased seeds. Fruits
were harvested individually from the field- and greenhouse-grown
plants, ensuring that ripe and unripe fruit are from the same
plants; they were inoculated with Salmonella within 3-4 hrs of
harvest.
For experiments on Salmonella gene regulation in tomato
phyllosphere, tomatoes cvs. Bonny Best and Hawaii 7997 were
grown on a private farm outside of Archer, FL. Two days prior to
the inoculation, secondary stems were excised from the field-
grown plants, transferred to the laboratory, and were partially


August 2010 1 Volume 5 1 Issue 8 | e12406







Salmonella-Tomato Interactions


Table 1. Fitness of Salmonella enterica mutants in red ripe
tomatoes, cv Campari.


Genotypes

AfadH15::kan
Abcs::kan
AcysB22::kan'
AlpfA30::kan
ASTM2006::kaon
Abcs AlpfA30 AfadH15
ASTM2006 A cysB22::kan


Mean of the log competitive
index standard error


0.17010.330
0.15610.145
-0.2760.086
0.010+0.280
0.30010.310
-0.5100.010


1 boldface shading indicates that the strain is less competitive than the wild
type, based on a two-tailed t-test (p<0.05) of log competitive indices as
described in Materials and Methods.
doi:1 0.1371/journal.pone.0012406.t001


submerged into sterile de-ionized water in light at room
temperature.

Promoter-probe screen
To harvest promoters that were differentially regulated inside
red ripe Campari tomatoes, two promoter probe-gfp libraries were
used [50]. One library was constructed with stable Turbo-GFP
i1. I.... Inc) and another library was constructed with Turbo-
GFP that was destabilized by an LVA C-terminal tag [50].
Libraries were transformed back into S. enterica sv Typhimurium
14028 by electroporation. Fluorescent activated cell sorting
(FACS) was done on FACS Aria-I cell sorter (BD Biosciences,
San Jose, CA) supported by Diva 6.2 software. Green fluorescent



Table 2. Competitive fitness of Salmonella enterica mutants
in virulence, motility and attachment genes within red ripe
tomatoes.


5. enterica genotype

sirA3::cam'
AcsrB20 AcsrC30::kan
flhDC::Tn10
motA::Tn10
fliF:kan
hilA1550::MudJ
14028 pSLT-
aroA::Tnl02
hns::kan
AagfB36::frt AagfC37::kan
AyihT27::kan
AyihT27-ompL19::frt


Mean of the log
competitive index
standard error

0.24410.044
0.183 0.043
0.133 0.094
0.33810.101
0.07010.060
0.33410.172
0.03810.122
-1.2510.183
N/A3
0.0810.15
0.05 0.07
0.002+0.13


'-underline indicates that the strain is more competitive than the wild type,
based on a two tailed t-test (p<0.05) of log competitive indices as described in
Materials and Methods.
-boldface indicates that the strain is less competitive than the wild type, based
on a two tailed t-test (p<0.05) of log competitive indices as described in
Materials and Methods.
3- the hns mutant was completely unfit, none were recovered from tomatoes.
doi:1 0.1371/journal.pone.0012406.t002


cells were detected at 488 nm with 525/10 band path filter.
Conditions for the cell sort were established first by defining the
parameters for pTURBO-gfp (non-fluorescent negative control),
blended filtered tomato pulp (negative control), a strongly
fluorescent construct carrying promoter for dppA (positive control)
and filtered tomato pulp spiked with the dppA-gfp construct
(Fig. 7).
For the fluorescence-activated cell sorts, libraries were grown in
LB broth or in M9 medium with ampicillin and glucose as the sole
carbon source. The libraries were then sorted to remove constructs
that were strongly expressed in vitro. "Dim" constructs were
collected, washed in PBS to remove the carrier buffer and then
inoculated into shallow wounds made in red ripe tomatoes cv.
Campari or field grown tomatoes as described in [7]. Infected
tomatoes were incubated for 2 days at room temperature; fruits
were then ground on ice in Phosphate-Buffered Saline (PBS) with
a mortar and pestle. Pulp was filtered 1,. ..1 1. Whatman #1 filter
paper. The resulting suspension was then centrifuged at 24,000 g
for 10 min, the supernatant was removed and the pellet was
resuspended in 7 ml of PBS. The resulting suspension was then
sorted into the P1 gate (Fig. 7) to isolate fluorescent constructs.
These were plated on LB with ampicillin. Six hundred individual
colonies resulting from three independent experiments were
picked and transferred into wells of microtiter 96-well plates.
Lack of strong fluorescence during _.. i1,i in LB broth or in M9
with glucose was confirmed with a multimode Victor-2 microtiter
plate reader. Colony PCR with primers InsertlR and Turbo4F
was carried out to identify unique constructs. Plasmids extracted
from 288 individual colonies were individually transformed into
chemically competent E. coli DH5a, from which they were
extracted using Qiagen Plasmid miniprep kit and then submitted
for sequencing with either InsertlR or Turbo4F.

Culture conditions and reporter assays
Strain construction is described above, strains are listed in Table
S1, additional constructs [51-53] were shared by colleagues. All
strains were maintained as frozen glycerol stocks, and were sub-
cultured into LB with appropriate antibiotics (50 gg/ml kanamy-
cin, 200 ig/ml ampicillin, and 10 gg/ml tetracycline) prior to the
experiments. For plate assays, bacteria were seeded into soft LB or
M9 agar (0.3% agar) with or without X-gal (40 gg/ml) as
indicated in text. Linoleic acid purchased from MP Biomedical,
Solon, OH, USA.
For the RIVET assays in tomatoes, Salmonella cultures were
grown at 37'C overnight in LB supplemented with tetracycline.
Bacterial cultures were then pelleted, washed three times in an
equal volume of sterile PBS and diluted to 108 cfu/ml.
Approximately 105 cfu (in 3 jl of PBS) were inoculated onto
superficial 1 mm wounds on surfaces of unwaxed immature or
mature fruits. At least three technical and six biological
replications were carried out for each experiment. All RIVET
assays were carried out for a week. To harvest samples, unless
otherwise indicated in text, cores 15 mmx0.5 mm were removed
from fruits, homogenized in PBS and aliquots were then plated on
xylose-lysine deoxycholate (XLD) agar (Oxoid). Individual colo-
nies were patched on LB i_ ,1 ;11, tetracycline to detect constructs
in which TnpR recombinase was active. Regulation of the
reporters in tomato phyllosphere was carried essentially as above,
except 10 jtl of the bacterial suspension in sterile water were
spotted onto the abaxial surface of tomato leaves and incubated in
light at room temperature for two days. Leaves were then _. ,11
blotted on XLD, and then assayed for the TnpR recombinase
activity.


August 2010 1 Volume 5 1 Issue 8 | e12406


,.* ', PLoS ONE I www.plosone.org






Salmonella-Tomato Interactions


A Un-infected tomato B "OFF"


10z 10" 104 10"
fluorescence intensity


E 10-' i 1io4 105
Sort of the gfp library
recovered from tomato
Q4


10' 10" 10' 10'
fluorescence intensity


1 F ..
F 10 10; io4 105
Sort of the gfp-LVA library
recovered from tomato
sn r ----









102 103 104 105
fluorescence intensity


Figure 7. FACS cell sort of Salmonella gfp libraries recovered from tomato fruits. To establish parameters for cell sorting, filtered tomato
pulp (A), "OFF" gfp construct (B) and the constitutive "ON" construct dppA-gfp (C) were injected into FACS-Aria. To confirm the selection of gates, P1
("ON") and P2 ("OFF"), a dppA-gfp reporter was mixed with tomato pulp and injected into FACS-Aria (D). Once gates were established, promoter
probe libraries recovered from tomatoes were sorted (E, F). Cells segregating into gates P1 ("ON") or P2 ("OFF") were collected for the identification
of promoters.
doi:10.1371/journal.pone.0012406.g007


Fitness of the Salmonella mutants
To calculate a competitive index, wild type S. enterica sv.
Typhimurium 14028 and isogenic mutants were seeded at
105 cfu/infection, roughly at al:1 ratio into tomatoes. In parallel,
S. enterica sv. Typhimurium 14028 and its isogenic tetracycline-
resistant derivative JS246 were similarly inoculated onto eight
tomato fruits, two wounds per fruit. To establish in vitro baselines,
individual mutants and the wild type (strain ATCC 14028) were
grown in LB to OD6oo ~ 0.3-0.9, washed in PBS, diluted to
OD6oo ~ 0.05 in LB, mixed 1:1 and incubated for 3 hrs at 22'C in
5 ml tubes on a shaker (160 rpm). All samples were incubated and
harvested using similar protocols. The relative ratios of the strains
in the inocula and in the recovered samples were calculated by
dilution plating and patching on antibiotic containing media.
Competitive indices were calculated for each treatment using the
formula \ N ,/WTot)/ \ I /WTin), where M is the proportion of
mutant cell and WT is the proportion of the wild type cells in the
inocula (in) or in the recovered samples (out). Statistical and
biological significance of each competitive index were established
by comparing log values of the competitive indices of each pair to
the log of competitive index similarly calculated for ATCC 14028
vs JS246 11 1 111 'i using a two-tailed t-test with unequal
variances (p<0.05).

Extraction of hydrophobic components from red
tomatoes
Tomatoes cv. Campari were blended and mixed with two
volumes of chloroform. The organic phase was washed with 50:50


References
1. Lynch MF, Tauxe RV, Hedberg CW (2009) The growing burden of foodborne
outbreaks due to contaminated fresh produce: risks and opportunities. Epidemiol
Infect 137: 307-315.


PLoS ONE | www.plosone.org


ethanol:water, rotary evaporated to dryness and re-dissolved in
chloroform. For bioassays, aliquots were dried down in a Spin/
Vac and then resuspended in canola oil. Approximately 0.4
tomato equivalents were added to the LB 0.3% agar plates for the
bioassays.

Supporting Information

Table Sl Strains and Plasmids used in the study.
Found at: doi:10.1371/journal.pone.0012406.s001 (0.11 MB
DOC)
Table S2 Primers used in the study.
Found at: doi:10.1371/journal.pone.0012406.s002 (0.07 MB
DOC)

Acknowledgments
We thank D. Kaganov for his contribution to the assays of Salmonella
mutants within tomatoes during initial phases of the project; R. Harshey
and K. Klose for sharing unpublished strains; J. B. Jones for seeds of
Hawaii 7997; W. D. Bauer for insightful comments during the preparation
of the manuscript.

Author Contributions
Conceived and designed the experiments: JTN MM MT. Performed the
experiments: JTN AA. Analyzed the data: JTN MM MT. Contributed
reagents/materials/analysis tools: NA MM MT. Wrote the paper: MT.


2. Brandl MT (2006) Fitness of human enteric pathogens on plants
and implications for food safety. Annu Rev Phytopathol 44: 367
392.


August 2010 1 Volume 5 1 Issue 8 | e12406


C "ON"







Salmonella-Tomato Interactions


3. Teplitski M, Barak JD, Schneider KR (2009) Human enteric pathogens in
produce: un-answered ecological questions with direct implications for food
safety. Curr Opin Biotechnol 20: 166-171.
4. Dong Y, Iniguez AL, Ahmer BM, Triplett EW (2003) Kinetics and strain
specificity of rhizosphere and endophytic colonization by enteric bacteria on
seedlings of Medicago sativa and Medicago truncatula. Appl Environ Microbiol 69:
1783-1790.
5. Cooley MB, Miller WG, Mandrell RE (2003) Colonization ofArabidopsis thaliana
with Salmonella enterica and enterohemorrhagic Escherchia coli 0157:H7 and
competition by Enterobacter asbunae. Appl Environ Microbiol 69: 4915-4926.
6. Klerks MM, van Gent-Pelzer M, Franz E, Zijlstra C, van Bruggen AH (2007)
Physiological and molecular responses of Lactuca sativa to colonization by
Salmonella enterica serovar Dublin. Appl Environ Microbiol 73: 4905-4914.
7. NoelJT,JoyJ, SmithJN, Fatica M, Schneider KR, et al. (2010) Salmonella SdiA
recognizes N-acyl homoserine lactone signals from Pectobactenum carotoorum in
vitro, but not in a bacterial soft rot. Mol Plant-Microbe Interact 23: 273-282.
8. Abbott DW, Boraston AB (2007) The structural basis for exopolygalacturonase
activity in a family 28 glycoside hydrolase. J Mol Biol 368: 1215-1222.
9. Rodionov DA, Gelfand MS, Hugouvieux-Cotte-Pattat N (2004) Comparative
genomics of the KdgR regulon in Erinia chysanthemi 3937 and other gamma-
proteobacteria. Microbiology 150: 35713590.
10. Kyle JL, Parker CT, G D, Brandl MT (2010) Transcriptome analysis of
Escherichia coli 0157:H7 exposed to lysates of lettuce leaves. Appl Environ
Microbiol 76: 1375-1387.
11. Carrari F, Baxter C, Usadel B, Urbanczyk-Wochniak E, Zanor MI, et al. (2006)
Integrated analysis of metabolite and transcript levels reveals the metabolic shifts
that underlie tomato fruit development and highlight regulatory aspects of
metabolic network behavior. Plant Physiol 142: 1380-1396.
12. Mathieu S, Cin VD, Fei ZJ, Li H, Bliss P, et al. (2009) Flavour compounds in
tomato fruits: identification of loci and potential pathways affecting volatile
composition. J Exper Bot 60: 325-337.
13. Tamir-Ariel D, Navon N, Burdman S (2007) Identification of genes in
Xanthomonas campestris pv. vesicatoria induced during its interaction with tomato.
J Bacteriol 189: 6359 6371.
14. Brown DG, Allen C (2004) Ralstonia solanacearum genes induced during growth in
tomato: an inside view of bacterial wilt. Mol Microbiol 53: 1641 1660.
15. Zhao Y, Blumer SE, Sundin GW (2005) Identification of Erinia amylovora genes
induced during infection of immature pear tissue. J Bacteriol 187: 8088
8103.
16. Marco ML, LegacJ, Lindow SE (2005) Pseudomonas sringae genes induced during
colonization of leaf surfaces. Environ Microbiol 7: 1379 1391.
17. BochJ,Joardar V, Gao L, Robertson TL, Lim M, et al. (2002) Identification of
Pseudomonas syringae pv. tomato genes induced during infection of Arabidopsis
thaliana. Mol Microbiol 44: 7388.
18. Jeter C, Matthysse AG (2005) Characterization of the binding of diarrheagenic
strains of E. coi to plant surfaces and the role of curli in the interaction of the
bacteria with alfalfa sprouts. Mol Plant Microbe Interact 18: 1235-1242.
19. BarakJD, Gorski L, Naraghi-Arani P, Charkowski AO (2005) Salmonella enterica
virulence genes are required for bacterial attachment to plant tissue. Appl
Environ Microbiol 71: 5685-5691.
20. BarakJD, Jahn CE, Gibson DL, Charkowski AO (2007) The role of cellulose
and O-antigen capsule in the colonization of plants by Salmonella enterica. Mol
Plant Microbe Interact 20: 1083-1091.
21. Lapidot A, Yaron S (2009) Transfer of Salmonella enteica Serovar Typhimurium
from contaminated irrigation water to parsley is dependent on curli and
cellulose, the biofilm matrix components. J Food Protect 72: 618-623.
22. Boyer RR, Sumner SS, Williams RC, Pierson MD, Popham DL, et al. (2007)
Influence of curli expression by Escheichia coli 0157:H7 on the cell's overall
hydrophobicity, charge, and ability to attach to lettuce. J Food Prot 70:
1339-1345.
23. Solomon EB, Niemira BA, Sapers GM, Annous BA (2005) Biofilm formation,
cellulose production, and curli biosynthesis by Salmonella originating from
produce, animal and clinical sources. J Food Protect 68: 906-912.
24. Solano C, Garcia B, ValleJ, Berasain C, GhigoJM, et al. (2002) Genetic analysis
of Salmonella ente tidis biofilm formation: critical role of cellulose. Mol Microbiol
43: 793-808.
25. Bidol SA, Daly ER, Ricker RE, Hill TA, Taylor TH, et al. (2007) Multistate
outbreaks of Salmonella infections associated with raw tomatoes eaten in
restaurants, United States, 2005-2006. CDC Morbid Mortal Week Rep 56:
909-911.
26. Gallegos-Robles MA, Morales-Loredo A, Alvarez-Ojeda G, Vega PA,
Chew MY, et al. (2008) Identification of Salmonella serotypes isolated from
cantaloupe and chile pepper production systems in Mexico by PCR-restriction
fragment length polymorphism. J Food Prot 71: 2217-2222.
27. Andrews-Polymenis HL, Santiviago CA, McClelland M (2009) Novel genetic
tools for studying food-borne Salmonella. Curr Opin Biotechnol 20: 149-157.
28. Shi X, Namvar A, Kostrzynska M, Hora R, Warriner K (2007) Persistence and
growth of different Salmonella serovars on pre- and postharvest tomatoes. J Food
Prot 70: 2725-2731.








PLoS ONE I www.plosone.org


29. Merighi M, Ellermeier CD, Slauch JM, Gunn JS (2005) Resolvase-in vivo
expression technology analysis of the Salmonella enterica serovar Typhimurium
PhoP and PmrA regulons in BALB/c mice. J Bacteriol 187: 7407-7416.
30. Hubbard PA, Liang XQ Schulz H, KimJJP (2003) The crystal structure and
reaction mechanism of Escherichia coli 2,4-dienoyl-CoA reductase. J Biol Chem
278: 3755337560.
31. Abuoun M, Suthers PF, Jones GI, Carter BR, Saunders MP, et al. (2009)
Genome scale reconstruction of a Salmonella metabolic model: comparison of
similarity and differences with a commensal Escherichia coli strain. J Biol Chem
284: 29480-29488.
32. Ostrowski J, Kredich NM (1991) Negative autoregulation of cysB in Salmonella
Typhimurium in vitro interactions of CysB protein with the cysB promoter.
J Bacteriol 173: 2212-2218.
33. Turnbull AL, Surette MG (2008) L-Cysteine is required for induced antibiotic
resistance in actively swarming Salmonella enterica serovar Typhimurium.
Microbiology 154: 3410-3419.
34. Chellemi DO, Dankers HA, Olson SM, Hodge NC, ScottJW (1994) Evaluating
bacterial wilt-resistant tomato genotypes using a regional approach. J Am Soc
Hort Science 119: 325-329.
35. Silby MW, Levy SB (2004) Use of in vivo expression technology to identify genes
important in growth and survival of Pseudomonas luorescens PfO-1 in soil: discovery
of expressed sequences with novel genetic organization. J Bacteriol 186:
7411-7419.
36. Wang QF, Harshey RM (2009) Res signalling-activated transcription of rcsA
induces strong anti-sense transcription of upstream fliPQR flagellar genes from a
weak intergenic promoter: regulatory roles for the anti-sense transcript in
virulence and motility. Mol Microbiol 74: 7184.
37. Prouty AM, Schwesinger WH, Gunn JS (2002) Biofilm formation and
interaction with the surfaces of gallstones by Salmonella spp. Infect Immun 70:
2640-2649.
38. Iniguez AL, Dong Y, Carter HD, Ahmer BM, StoneJM, et al. (2005) Regulation
of enteric endophytic bacterial colonization by plant defenses. Mol Plant
Microbe Interact 18: 169 178.
39. Ahmer BM, Tran M, Heffron F (1999) The virulence plasmid of Salmonella
typhimurium is self-transmissible. J Bacteriol 181: 1364 1368.
40. Ahmer BM, van ReeuwijkJ, Watson PR, Wallis TS, Heffron F (1999) Salmonella
SirA is a global regulator of genes mediating enteropathogenesis. Mol Microbiol
31: 971 982.
41. Fortune DR, Suyemoto M, Altier C (2006) Identification of csrC and
characterization of its role in epithelial cell invasion in Salmonella enterica serovar
Typhimurium. Infect Immun 74: 331 339.
42. Teplitski M, Ahmer BM (2004) The control of secondary metabolism, motility,
and virulence by the two-component regulatory system BarA/SirA of
Salmonella and other gamma-proteobacteria. In Pruss BM, ed. Global
Regulatory Networks in Enteric Bacteria Research Signpost.
43. Teplitski M, Al-Agely A, Ahmer BM (2006) Contribution of the SirA regulon to
biofilm formation in Salmonella enterica serovar Typhimurium. Microbiol 152:
3411 3424.
44. Teplitski M, Goodier RI, Ahmer BM (2003) Pathways leading from BarA/SirA
to motility and virulence gene expression in Salmonella. J Bacteriol 185:
7257-7265.
45. Klerks MM, Franz E, van Gent-Pelzer M, Zijlstra C, van Bruggen AH (2007)
Differential interaction of Salmonella enterica serovars with lettuce cultivars and
plant-microbe factors influencing the colonization efficiency. ISME J 1:
620631.
46. BarakJD, Liang A, Narm KE (2008) Differential attachment to and subsequent
contamination of agricultural crops by Salmonella enterica. Appl Environ Microbiol
74: 5568-5570.
47. Datsenko KA, Wanner BL (2000) One-step inactivation of chromosomal genes
in Escherichia coli K-12 using PCR products. Proc Nad Acad Sci U S A 97:
6640-6645.
48. Neidhardt FC, Curtiss R (1996) Escherichia coli and Salmonella: cellular and
molecular biology edn 2nd. Washington D.C.: ASM Press.
49. Osorio CG, Crawford JA, Michalski J, Martinez-Wilson H, Kaper JB, et al.
(2005) Second-generation recombination-based in vivo expression technology for
large-scale screening for Vibrio cholerae genes induced during infection of the
mouse small intestine. Infect Immun 73: 972-980.
50. Arrach N, Zhao M, Porwollik S, Hoffman RM, McClelland M (2008) Salmonella
promoters preferentially activated inside tumors. Cancer Res 68: 4827-4832.
51. Metcalf WW, Jiang W, Daniels LL, Kim SK, Haldimann A, et al. (1996)
Conditionally replicative and conjugative plasmids carrying lacI alpha for
cloning, mutagenesis, and allele replacement in bacteria. Plasmid 35: 1-13.
52. Miller VL, MekalanosJJ (1988) A novel suicide vector and its use in construction
of insertion mutations: osmoregulation of outer membrane proteins and
virulence determinants in Vibrio cholerae requires toxR. J Bacteriol 170:
2575-2583.
53. Santiviago CA, Reynolds MM, Porwollik S, Choi SH, Long F, et al. (2009)
Analysis of pools of targeted Salmonella deletion mutants identifies novel genes
affecting fitness during competitive infection in mice. PLoS Pathogens 5:
e1000477.


August 2010 1 Volume 5 1 Issue 8 | e12406




University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Powered by SobekCM