Comparative genome analysis of pathogenic and non-pathogenic Clavibacter strains reveals ...

14 Pages·2014·0.5 MB·English

Załugaetal.BMCGenomics2014,15:392
http://www.biomedcentral.com/1471-2164/15/392
RESEARCH ARTICLE Open Access
Comparative genome analysis of pathogenic and
non-pathogenic Clavibacter strains reveals
adaptations to their lifestyle
Joanna Załuga1*, Pieter Stragier1, Steve Baeyen2, Annelies Haegeman2, Johan Van Vaerenbergh2,
Martine Maes2 and Paul De Vos1,3
Abstract
Background: The genus Clavibacter harbors economically important plant pathogens infecting agricultural crops
such as potato and tomato. Although thevast majorityof Clavibacter strains are pathogenic,thereis an increasing
number ofnon-pathogenic isolates reported. Non-pathogenic Clavibacter strains isolated from tomato seeds are
particularly problematic becausetheyaffectthecurrentdetectionandidentificationtestsforClavibactermichiganensis
subsp.michiganensis(Cmm),which is regulated with a zero tolerance in tomato seed. Their misidentification as
pathogenic Cmm hampers a clear judgment on the seed quality and health.
Results: Togetmoreinsightinthegeneticfeatureslinkedtothelifestyleofthesebacteria,awhole-genomesequenceof
thetomatoseed-bornenon-pathogenicClavibacterLMG26808wasdetermined.Togainabetter understanding of the
molecular determinantsofpathogenicity,thegenomesequenceofLMG26808wascomparedwiththatofthepathogenic
Cmmstrain(NCPPB382).ThecomparativeanalysisrevealedthatLMG26808doesnotcontainplasmidspCM1andpCM2
andalsolacksthemajorityofimportantvirulencefactorsdescribedsofarforpathogenicCmm.Thisexplainsitsapparent
non-pathogenicnatureintomatoplants.Moreover,thegenomeanalysisofLMG26808detectedsequencesfromaplasmid
originatingfromamemberofEnterobacteriaceae/Klebsiellarelative. Genes received that way and coding for antibiotic
resistancemayprovideacompetitiveadvantageforsurvivalofLMG26808initsecologicalniche.Genetically,LMG
26808wasthemostsimilartothepathogenicCmmNCPPB382butcontainedmoremobilegeneticelements.The
genomeofthisnon-pathogenicClavibacterstraincontainedalsoahighnumberoftransportersandregulatorygenes.
Conclusions:Thegenomesequenceofthenon-pathogenicClavibacterstrainLMG26808andthecomparativeanalyses
withotherpathogenicClavibacterstrains provided a better understanding of the genetic bases of virulence and
adaptation mechanisms present in the genus Clavibacter.
Keywords: Non-pathogenicClavibacter,Bacterialwiltandcanker,Tomatoseeds,Genomesequencing,Quarantine,Plant
pathogen
Background initiatethecharacterizationofthesenon-pathogenicisolates
Clavibacter isgenerallyconsideredagenus ofplantpatho- [3,4]. The main objective of this study was to investi-
gens, but ecological surveys suggest that environmental, gate key genomic features of non-pathogenic Clavibac-
non-pathogenic Clavibacter occur more commonly than ter isolated from tomato seeds. These strains tend to
previously thought [1,2]. Generally these non-pathogenic be misidentified as Cmm in serological and molecular
isolatesareoverlookedsincediagnosticproceduresfocuson tests used in seed assays. Their high genetic and pheno-
pathogenicstrains.Justrecently,studieswereundertakento typic similarity to pathogenic Cmm strains hampers a
clearjudgmentonseedhealth.
The majority of non-pathogenic Clavibacter strains
*Correspondence:[email protected] isolated from tomato seeds exhibit similar cell and col-
1LaboratoryofMicrobiology,DepartmentofBiochemistryandMicrobiology,
ony morphology to the genuine Cmm [5]. Because of
GhentUniversity,K.L.Ledeganckstraat35,GentB-9000,Belgium
Fulllistofauthorinformationisavailableattheendofthearticle
©2014Załugaetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycredited.TheCreativeCommonsPublicDomain
Dedicationwaiver(http://creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,
unlessotherwisestated.
Załugaetal.BMCGenomics2014,15:392 Page2of14
http://www.biomedcentral.com/1471-2164/15/392
the commonbiologicalorigin(tomatoseed),highsequence strains,which could deliversomeinformative insightsinto
similarities and similar physiological characteristics, the the differences in virulence determinants, genetic content
non-pathogenicClavibacter strains are suggestedto be the and adaptation to a lifestyle in their natural ecological
most related to Cmm. Initial in planta experiments niche(s). Genome comparison between pathogenic and
demonstrated that this group of isolates is not patho- non-pathogenicstrainsbelongingtothesamespeciesisan
genic to the tomato plant and they do not colonize the important and valuable approach to identify genes that
vascular tissues of tomato [4]. Non-pathogenic clavi- may contribute to virulence and general fitness of the
bacters neither induce a hypersensitive reaction (HR) organism.
after infiltration of Nicotiana tabacum and Nicotiana In this report we present the genome analysis of non-
benthamiana leaves[3],norwheninoculatedtoMirabilis pathogenic Clavibacter LMG 26808 isolated from to-
jalapa (J. Van Vaerenbergh, personal communication). mato seed. The specific purposes of this study were a) to
Furthermore, a majority of these strains is lacking one or generate a draft genome sequence of this strain, b) to
bothCmmplasmidscarryingimportantvirulencefactors. analyze it for virulence-related gene content by compar-
So far there is very little information available on non- ing it to the available genome of the pathogenic Clavibac-
pathogenicClavibacterstrainsisolatedfromtomatoseeds. ter michiganensis subsp. michiganensis (Cmm) NCPPB
Reports concerning the ecological niche, survival abilities 382, c) to perform a comparative analysis with the ge-
or nutritional requirements are lacking. Knowledge about nomes of Cmm (NCPPB 382) [8], Clavibacter michiga-
the biology of these strains is limited, not only because nensis subsp. nebraskensis (Cmn) (NCPPB 2581, released
they were only recently identified as constituting a separ- withoutpublication)andClavibactermichiganensissubsp.
ate Clavibacter group but also because their significance sepedonicus (Cms) (ATCC 33113) [9], pathogenic to to-
in the Cmm identification procedure has not been evalu- mato, maize and potato, respectively, d) to search for ad-
ated previously. Their ecological niches remain unknown; aptationstoanon-pathogeniclifestyle.
routes of transmission and possible sources of these
strainshavenotyetbeenrecognized.
Methods
High genetic and phenotypic similarities of non-
StrainsandDNAextraction
pathogenic Clavibacter and pathogenic Cmm strains
Non-pathogenicClavibactersp. LMG26808wasreceived
are the main reasons for their misidentifications as
as isolate PD 5684 from Naktuinbouw, The Netherlands.
Cmm in the currently recommended detection/identi-
Itwasobtainedindilutionplatingonsemi-selectivemedia
fication testsforCmmintomatoseeds[6].Cross-reactions
accordingtothe currentmethodfordetectionofCmmin
with antisera specific for Cmm and/or positive PCR reac-
tomato seeds recommended by the International Seed
tions with primers used for identification of Cmm illustrate
Federation(ISF) [6]. LMG26808 isphenotypically similar
theproximityofsurfaceantigensandgenomicsequencesof
to Cmm on SCMF and CMM1T and was identified as
non-pathogenic seed-borne Clavibacter to the pathogenic
Cmm in commonly practiced PCR tests but showed no
Cmm[3].RecentstudiesdemonstratedthatneitherPCRas-
pathogenicityintomatoplants[3,4].LMG26808wasaer-
saysbasedoncommonlyused16SrRNA genes or ITS re-
obically grown on MTNA (mannitol, trimethoprim, nali-
gion, nor those designed for the detection of known
dixic acid, amphotericin) medium without antibiotics
virulence factors are specific to only pathogenic Cmm
at 25°C for 24 h-48 h [10]. Stock cultures were stored
[3,7]. Furthermore, some non-pathogenic Clavibacter
at −80˚C in MicrobankTM beads (Pro-Lab Diagnostics,
strains showed fainter PCR amplicons on the gel im-
Canada).Total genomic DNA was extracted according
peding the correct interpretation of the results [4].
to the guanidium-thiocyanate-EDTA-sarkosyl method
Taxonomically, these non-pathogenic clavibacters from
described by Pitcher [11], which was adapted for Gram-
tomato seeds are distinct from all Clavibacter subspe-
positive bacteria by a pre-treatment with lysozyme
cies (based on the analysis of housekeeping genes gyrB
(5 mg/μl lysozyme in TE buffer) and incubation for
and dnaA) [4].
40 minutes at 37°C.
Recent developments in the field of molecular biology
and sequencingallowed generating complete genome se-
quences andsubsequentlydeterminingmetabolictraitsfor Plasmidextraction
many organisms. Complete genome sequences of Cmm Isolation of plasmid DNA was based on the alkaline
NCPPB 382 [8], Cms ATCC 33113 [9] and Cmn NCPPB method of Anderson and McKay [12]. Agarose gel elec-
2581 (released without publication) provide genetic infor- trophoresis was performed in a Tris acetate buffer con-
mation that allows for comparative studies and helps to taining 40 mM Tris, 20 mM acetic acid, and 2 mM
better understand their pathogenicity characteristics and Na EDTA (pH 8.1). Gels contained 0.8% agarose and
2
host adaptation. However, no information is available electrophoresis was performed at 55 Vfor 16 hrs at 4˚C.
about the genome content of non-pathogenic Clavibacter Gelswerestainedwith ethidium bromide0.5μg/ml.
Załugaetal.BMCGenomics2014,15:392 Page3of14
http://www.biomedcentral.com/1471-2164/15/392
Genomesequencing Comparativegenomicanalysis
Library preparation and genome sequencing was per- Artemis software was used for data management and
formed by BaseClear (Leiden, The Netherlands). High- DNAPlotter was used for genome visualization [20]. The
molecular weight genomic DNA was used as input for MAUVE alignment tool was used for multiple genomes
library preparation using the Illumina TruSeq DNA li- sequence alignment and visualization. IslandViewer was
brary preparation kit (Illumina). Briefly, the gDNA was used to analyze possible genomic islands (GI) on the
fragmentedandsubjectedtoend-repair,A-tailing,ligation draft genome [21]. IslandViewer integrates two sequence
of adaptors including sample-specific barcodes and size- composition GIs prediction methods, namely IslandPath-
selection to obtain a library with median insert-size DIMOB [22] and SIGI-HMM [23] and one single com-
around 300 bp. After PCR enrichment, the resultant parativeGIpredictionmethod,namelyIslandPick[21]for
library was checked on a Bioanalyzer (Agilent) and genomicislandprediction.
quantified. The libraries were multiplexed, clustered, ISsaga application from ISfinder server [24] was used
and sequenced on an Illumina HiSeq 2000 with paired- to identify insertion sequences (IS) and transposons in
end 50 cycles protocol. The sequencing run was analyzed the draft genome of LMG 26808. Sequences exhibiting
with the Illumina CASAVA pipeline (v1.8.2). The raw se- homology to IS and transposon sequences were verified
quencing data produced was processed removing the with the Mobilomics software [25]. The core genome
sequence reads which were of too low quality (only was estimated using the Phylogenetic profiler tool that is
“passing filter” reads were selected) and discarding reads part of the IMG system (https://img.jgi.doe.gov/cgi-bin/
containingadaptor sequences orPhiXcontrol withanin- er/main.cgi)at a similaritycutoffof50%.
housefilteringprotocol. The presence of possible virulence-related genes and
Apaired-end(PE) DNAlibrarywith amean insert size genes expressed during tomato infection in the draft
of 300 bp was sequenced with average reads of 101 bp genome of the non-pathogenic Clavibacter was analyzed
onanIlluminaGenomeHiSeq2000(IlluminaInc.).Next, by comparing it with tomato pathogen Cmm NCPPB
a mate-paired (MP) DNA library with a mean insert size 382. The comparative screening of the gene content was
of 3800 bp was sequenced with average reads of 51 bp performed in RAST, IMG-ER and EDGAR [26]. Absence
on an Illumina Hiseq2000 (Illumina Inc.). Automatic or presence of coding sequences in each genome, as re-
trimming (based on a threshold of Q=20) and assembly ported by RAST, IMG-ER and EDGAR were independ-
was performed using CLC Genomics Workbench v5.0. ently confirmed by performing protein and nucleotide
An initial de novo assembly was performed in CLC Gen- BLASTqueriesinthetargetgenomes.Proteinswithamino
omics Workbench v5.0 using the quality trimmed and acid sequence similarities higher than 50% and with a
paired readsfromthePEandMPreads.Allcontigsshorter coverage higher than 70% were considered homologs.
than 200 bp were discarded. Remaining N-nucleotides in Based on the RAST, IMG-ER and EDGAR annotation re-
the scaffolds, introduced during scaffolding, were removed sults,the presence of known and putative virulence fac-
from the final sequence by breaking up the scaffolds tors, pathogenicity related genes and genes uniquely
back into contigs where they were encountered. The present in the non-pathogenic Clavibacter LMG 26808
quality of the final draft genome sequence was com- were investigated. Identification of orthologous groups
pared to the initial PE-based de novo assembly through between fouravailableClavibactergenomeswasachieved
comparative read-mapping in CLC Genomics Work- by OrthoMCL analyses [27]. OrthoMCL clustering ana-
bench v5.0 using the trimmed read sets. Contigs were lyseswere performed using default parameters with the
ordered automatically with MAUVE [13] and manually P-value Cut-off=1×10−5.
with Artemis [14] by comparing with Cmm NCPPB
382. Deposition
The current draft genome sequence was deposited at
Genomeannotation Genbank under accession number AZQZ00000000 after
Functional annotation and metabolic reconstruction were automatic annotation by the PGAAP online annotation
performed with (1) the Rapid Annotation Subsystem pipeline.
Technology (RAST) server [15], using Glimmer [16] for
genecallingandallowingframeshiftcorrection,backfilling Results and discussion
of gaps and automatic fixing of errors, with (2) the Inte- Generalfeaturesofnon-pathogenicClavibacterLMG
grated Microbial Genomes Expert Review (IMG-ER) an- 26808
notation pipeline (https://img.jgi.doe.gov/cgi-bin/er/main. Genome assembly using paired-end and mate-paired
cgi) [17]. Assigned functions were checked with BLAST reads resulted in a 3.47 Mb sequence represented in 70
[18].Alignmentandphylogeneticanalysiswereperformed contigs from which the longest covered more than one
withMEGA5.0[19]. million bp (Table 1). The initial PE de novo assembly
Załugaetal.BMCGenomics2014,15:392 Page4of14
http://www.biomedcentral.com/1471-2164/15/392
Table1Genomecharacteristicsofthenon-pathogenic couldbedetected.TheGCcontentofthedraftgenomeav-
ClavibacterLMG26808 erages 72%. There are 46 tRNA genes and two complete
Genomecharacteristics Non-pathogenic rRNA operons. A total of 3218 protein-coding genes are
ClavibacterLMG26808 predicted in non-pathogenic Clavibacter strain (in IMG-
No.contigs(>200bp) 70 ER),whichissimilartotheCmmgenomeNCPPB382that
Totalcontigsize(bp) 3,476,455 contains 3107 protein-coding genes. The genome of the
non-pathogenicClavibacterstraincontains685(21.3%)pro-
N50(bp)afterscaffolding 383,456
teins without predicted function being either annotated as
Largestcontigsize(bp) 1,028,177
conserved hypothetical proteins or proteins with unknown
GCcontent(%) 72.01
function.
No.RNAcalls 7rRNA The number of genes detected in the draft genome of
46tRNA LMG 26808was higher thaninthe otherthree complete
No.CDScalls 3218 Clavibacter genomes (Table 2). The average nucleotide
identity (ANI) between the draft genome of the non-
NCBIaccessionno. AZQZ00000000
pathogenic Clavibacter and the three published Clavi-
Numberofinsertionelements 10
bacter genomes Cmm NCPPB 382 (NC_009480.1), Cms
ATCC 33113 (NC_010407.1) and Cmn NCPPB 2581
wasusedforscaffoldingwiththeMPdataset.Inthefinal (NC_020891.1) was determined using the in silico DNA-
consensus sequence each base matched at least Phred DNA hybridization (DDH) method included in the soft-
quality scoreof35.LMG26808containsonechromosome ware JSpecies [28]. The results indicated that LMG
andevidenceofapresenceofaplasmidthatshowedahigh 26808 is genetically most related to Cmm NCPPB 382
similarity to a Klebsiella pneumoniae Kp11978 plasmid (94.96% ANI), followed by Cmn NCPPB 2581 (92.75%
pOXA-48 (JN626286.1). The genesofKp11978were found ANI) and Cms ATCC 33113 (92.48% ANI). Although
on 15 contigs in a draft genome of LMG 26808 (estimated based on the ANI values the LMG 26808 genome is the
sizeofthesecontigs~48kbp,%GC~50%)(Additionalfile1: most similar tothatofpathogenicCmmNCPPB382,the
Table S1). No sequences of known Clavibacter plasmids synteny plots of LMG 26808 and Cmn NCPPB 2581 and
Table2Comparisonofgenomecharacteristics(basedonIMG-ERserver)
GenomeName Clavibactermichiganensis Clavibactermichiganensis Clavibactermichiganensis Non-pathogenic
subps.michiganensis subsp.sepedonicus subsp.nebraskensis ClavibacterLMG26808
NCPPB382 ATCC33113 NCPPB2581
Accessionnumbera NC_009480.1 NC_010407.1 NC_020891.1 AZQZ00000000
Host tomato potato maize *
Disease bacterialwiltandcanker potatoringrot wiltandblight non-pathogenic
Size 3395237 3403786 3063596 3476455
Genes 3169 3168 2936 3282
CDS 3107 3117 2890 3218
CDS(%) 98.04 98.39 98.43 98.05
RNA 62 51 46 64
rRNA 6 6 6 7
tRNA 45 45 30 46
Enzymes 759 712 740 750
CRISPR 1 0 0 1
GC% 72 72 73 72
Codingbases 3041059 2955244 2823671 3074588
Signalpb 281 234 219 140
Signalp(%) 8.87 7.39 7.46 4.27
HomologstoLMG26808(%)c 2716(87.4) 2457(78.8) 2531(87.5) -
aOnlytheGenbankrecordsofthechromosomesaregiven.
bNumberofgenescodingsignalpeptides.
cCalculatedusingaGenomeGeneBestHomologstoolincludedinIMG-ERserver.
*Isolatedfromtomatoseeds,hostunknown.
Załugaetal.BMCGenomics2014,15:392 Page5of14
http://www.biomedcentral.com/1471-2164/15/392
the percentage of homologous genes shared by LMG amongClavibacter strains most likely contribute to the
26808 and Cmn NCPPB 2581 are also considerably high differences inthebacteriallifestyle.Phage-relatedrecom-
(Table 2, Figure 1). The genomes of LMG 26808, NCPPB binases (e.g. Cl_00892, Cl_03056), integrase/resolvase
382 and NCPPB 2581 are collinear with less than 5 re- (e.g. Cl_02713) and other insertion elements (transpo-
combinationalbreakpoints. sases, e.g. Cl_03190) associated with a phage were found
Comparison on a genomic scale revealed a high con- in higher numbers in the genome of LMG 26808 than in
servation in the gene sequence among genomes of LMG the Cmm NCPPB 382 genome (Table 3). The genome of
26808, NCPPB 382 and NCPPB 2581 (Figure 1). There LMG 26808 contained sixteen genes coding for transpo-
are 299 genes (~10%) present in the LMG 26808 draft sasesandrecombinases(Table3).Thisnumberwasmuch
genome that were not detected in the Cmm NCPPB 382 lower in comparison to more than 100 genes found in
genome. Forty eight of them were detected in Cmn and/ Cms (ATCC 33113)[9]. Detected IS belonged to IS3,IS4,
or Cms genomes (Additional file 1: Table S2). 37 unique IS5, IS6 and IS1380 families. Transposases were repre-
genes of LMG 26808 were associated with the plasmid sented by Tn3 (20%). No pseudogenes among transpo-
and/or low GC regions. 214 of the unique genes were sases and recombinases were detected suggesting that
foundinthecorechromosomeofLMG26808(Additional these elements may encode functional genes.None of the
file 1: Table S2). Almost half of the genes specific for IS elements found in LMG 26808 has homologs in the
LMG26808belonged to hypothetical or unknown pro- other Clavibacter strains. The most common IS element
teins (120). Remaining sequences were coding for add- in Cms ATCC 33113 is IS1121 [9]. Cmm NCPPB 382
itional ABC transporters, antibiotic resistance genes, containsonlyafewIS,whicharemostprobablynonfunc-
acetyltransferases and several enzymes that in majority tional [8]. Cmn NCPPB 2581 contained only two types of
couldnotbeassignedtoanyKEGG(KyotoEncyclopediaof IS,namelyILSre2andISNCY(predictedbyISsaga).
GenesandGenomes)pathway(Additionalfile1:TableS2). The comparison of functional categories as defined
When compared to other Clavibacter genomes, LMG by COG (Clusters of Orthologous Groups) showed no-
26808 appeared to not have experienced gene losses and ticeable differences in the gene content in categories of
despiteitisconsideredonlyadraft,themajorityofimport- ‘carbohydrate transport and metabolism’ and of ‘trans-
antgenesinvolvedinbasicmetabolismandgeneregulation lation, ribosomal structure and biogenesis’. All included
could be detected. Comparative analysis (based on KEGG Clavibacter strains contained a higher percentage of
pathways) showed that LMG 26808 lacks sulfate and ni- genes in these two categories than a free-living organ-
trate reduction pathways suggesting that its capability of ism Escherichia coli 081 ED1a or a tomato pathogen as
survival in soli might be similar to this observed in Cmm e.g. Pseudomonas syringae pv. tomato T1 (Additional
NCPPB 382. The core genome consists of 2316 homologs file 1: Figure S1). These observations are supporting the
found in all four Clavibacter genomes. LMG 26808 con- hypothesis that compared Clavibacter strains can most
tains 12 genomic regions exhibiting a lower GC content probably utilize different sugars as an energy source and
(Additional file 1: Table S3). Several genes coding for pro- that they possess a wide range of transport systems that
teins within these regions were found to contribute to the enable the efficient trafficking of the substrates and prod-
fitness of the bacterium (Cl_02679 coding for ABC-type ucts. The presence of a high number of genes involved in
Fe3+-siderophore transport system; Cl_03044 coding for translation, ribosomal structure and biogenesis implies
permease component, chloramphenicol acetyltransferase thatthesebacteriarespondmoreeffectivelyandrapidlyto
(EC 2.3.1.28); Cl_03094 coding for multidrug-efflux trans- nutritional resources, which can be an important advan-
porter). The genome heterogeneity and genetic diversity tageinachangingenvironment.
Figure1Syntenicdotplotsshowingthesimilarityofthegenomesincludedintheanalysis.A)non-pathogenicClavibacterLMG26808(x-axis)
andCmmNCPPB382(y-axis);B)non-pathogenicClavibacterLMG26808(x-axis)andCmnNCPPB2581(y-axis);C)non-pathogenicClavibacterLMG
26808(x-axis)andCmsATCC33113(y-axis);D)CmmNCPPB382(x-axis)andCmnNCPPB2581(y-axis)(DiagramsgeneratedinIMG-ER).
Załugaetal.BMCGenomics2014,15:392 Page6of14
http://www.biomedcentral.com/1471-2164/15/392
Table3MobilegeneticelementsfoundinthegenomeofLMG26808(BasedontheannotationresultsfromIMG-ER,
RASTandEDGAR)
CDSidentifiers COG COGannotation Pfam Position Length(bp)
Cl_00892 COG4974 Site-specificrecombinaseXerD,phage_integrase pfam00589 Contig3(50199to51350) 1152
Cl_00935 COG4974 Site-specificrecombinaseXerD,phage_integrase pfam00589 Contig5(8554to9540) 987
Cl_01562 COG4974 Site-specificrecombinaseXerD,phage_integrase pfam00589 Contig5(682570to683556) 987
Cl_01811 COG1842 PhageshockproteinA(IM30),suppresses pfam04012 Contig5(938633to939376) 744
sigma54-dependenttranscription
Cl_01968 COG3600 Uncharacterizedphage-associatedprotein pfam13274 contig7(7492to7956) 753
Cl_03043 COG4679 Phage-relatedprotein pfam05973 contig15(974to1303) 330
Cl_03056 COG4974 Site-specificrecombinaseXerD,phage_integrase pfam00589 contig15(22760to23485) 726
Cl_03252 COG4974 Site-specificrecombinaseXerD,phage_integrase pfam00589 contig46(189to866) 678
Cl_02713 COG2452 Predictedsite-specificintegrase-resolvase pfam12728 contig11(350076to350522) 447
Cl_03047 COG2801 Transposaseandinactivatedderivatives,Tnp1,IS3_IS150 pfam01527 contig15(9545to11077) 1533
Cl_03189 - TransposaseDDEdomain,Tnp1,IS1380 pfam01609 contig28(1417to2532) 1116
Cl_03190 COG4644 Transposaseandinactivatedderivatives,TnpAfamily,Tn3 pfam01526 contig28(4206to5753) 1530
Cl_03209 COG4644 Transposaseandinactivatedderivatives,TnpAfamily,Tn3 pfam01526 contig33(10984to12801) 1818
Cl_03210 COG3316 Transposaseandinactivatedderivatives,IS6 pfam13610 contig33(12848to13552) 705
Cl_03211 COG2801 Transposaseandinactivatedderivatives,IS3_IS150 pfam13276 contig33(14010to13498) 513
Cl_03212 COG2963 Transposaseandinactivatedderivatives,Tnp1,IS3 pfam01527 contig33(14489to14343) 147
Cl_03214 COG3316 Transposaseandinactivatedderivatives,IS6 pfam13610 contig34(391to116) 276
Cl_03235 - Transposase,Tnp1,IS5_IS903 pfam13737 contig39(1373to2212) 840
Cl_03261 - TransposaseDDEdomain,Tnp1,IS4 pfam01609 contig51(1to1188) 1188
Cl_03204 - Mobileelementprotein - contig33(530to366) 165
peg.1244 - Mobileelementprotein - contig28(149to742) 594
Cl_03063 - Mobileelementprotein - contig15(32765to34252) 1488
peg.807 - Resolvase-like - contig15(6452to6847) 396
peg.1245 - Tn1transposase - contig28(741to1088) 348
Cl_03045 - Gifsy-2prophageprotein - contig15(7251to7544) 294
Cl_03251 - putativebacteriophageprotein - contig45(3200to4171) 972
Cl_01918 - elementsofexternalorigin;phage-related - contig6(14793to15281) 489
functionsandprophages
Genomicislands regionsaccountsfor265kb(~7%ofthegenomesize).The
The analysis of the LMG 26808 genome showed that at equivalent of PAI of Cmm NCBI 382 (130 kb) containing
least12regionswithlowerGCcontentsdistributedamong two regions chp and tomA with important genes respon-
different contigs could be distinguished (Additional file 1: sible for effective plant colonization, was not found in
TableS3).Partsofgenomicislands3and4foundinLMG LMG 26808, nor in other Clavibacter genomes. However,
26808 partly overlap with the chp region of pathogenicity a number of orthologswerefound in all three Clavibacter
island (PAI) described previously in Cmm NCPPB 382 genomes (Additional file 1: Table S4). The higher number
(Additional file 1: Table S3). Genomic islands with lower of orthologs of genes encoded on chp and tomA regions
GC% are thought to be integrative elements that exhibit (as detected by OrthoMCL) was found in Cms ATCC
different codon usage relative to the rest of the genome, 33113(32),followedbyLMG26808(17)andCmnNCPPB
encodefortransposases, integrases and aretypicallyfound 2581(10). Onlysix orthologs of PAI (chpregion) found in
attRNAloci.Theiracquisitionismostlyaresultofactions LMG 26808 were located on the low GC region 3 and 4
of phages, transposons or horizontal gene transfer [29]. (Additionalfile1:TableS3).
SomeofthegenespresentintheseregionsinLMG26808 Genomic regions with lower GC content can contain
were detected previously in the genome of Cmm NCPPB diverse genes exhibiting functions in many metabolic
382 but the majority represents regions that were not processes. The longest region found in LMG 26808
found in Clavibacter subspecies. The total size of these (more than 100 kb) included genes coding for antibiotic
Załugaetal.BMCGenomics2014,15:392 Page7of14
http://www.biomedcentral.com/1471-2164/15/392
resistance (beta-lactamase class A, Cl_03208, Cl_03230), neededtoelucidatetheexactfunctionsofthesegenes.The
transposases (Cl_03209, Cl_03212) and many hypothet- smaller plasmid pCM1 was not detected during the plas-
ical proteins(Cl_03223,Cl_03183)someofwhichshowed mid extraction, nor were its sequences found in the gen-
the highest similarity on the protein level to genes found ome sequence of LMG 26808. Despite that two DNA
on Klebsiella pneumoniae plasmids. Genomic region 3 fragments of LMG 26808 showed to be highly similar to
containssomegenes that showed similarities tothe genes two plasmid-encoded genes, namely pCM1_0018 and
foundinpCM2plasmidofCmmNCPPB382.Themajor- pCM1_0020, the reciprocal BLAST search revealed that
ity of them are hypothetical proteins and two of them these sequences from the non-pathogenic Clavibacter
codeforacetyltransferasesAdditionalfile1:TableS3.Pre- genome are more similar to the chromosomally encoded
vious studies indicated that some pathogenic Clavibacter CMM_1065 and CMM_2443, respectively. Interestingly,
strains lacking pCM1 and pCM2 showed a positive signal the latter gene encodes CelB, which is a putative secreted
inhybridizationexperimentswithspecificplasmidregions cellulase that contains a cellulose-binding domain (endo-
of Cmm NCPPB 382 implying that some of the genes 1,4-beta-glucanase). Chromosomally encoded celB misses
found originally in Cmm plasmids may be actually one of three protein coding domains present in the celA
chromosomally-encodedinotherCmmstrains[30]. gene. The missing endoglucanase C-terminal domain is
Genomic regions 7, 9 and 10 with lower GC content similar to the α-expansin protein family that occurs in
contained some genes encoding transposases and recom- plants and is essential for development of wilting and for
binases, which might imply their possible exchange/ degradationofcrystallinecellulose[8,32].Thelackordis-
mobilization ability. In region 7 one phage-related gene ruption of any of these domains of celA inevitably led to
(Cl_03043), showing homology to prophage protein gp49, the disability to induce disease symptoms in a tomato
was detected. Its presence may represent the remains of plant[32].
prophagegenes. Surprisingly, the genome analysis showed the presence
of sequences found in Klebsiella pneumoniae plasmid
Plasmidcontent pOXA-48 (61881 bp). The presence of sequences from a
LMG 26808 did not contain any of two known virulence plasmid of Gram-negative bacteria in a Gram-positive
plasmids found in pathogenic Cmm NCPPB 382. How- Clavibacter strain is rather unusual and has not been re-
ever, the plasmid extraction demonstrated the presence ported previously. Although the genome sequence of
of one plasmid, which size was slightly smaller than that LMG 26808 is only a draft and therefore incomplete, we
of plasmid pCM2 (70 kb). Initially, we assumed that it could not detect any sequences that could be attributed
might be a pCM2 that lost some genes because in the toaKlebsiellapneumoniaeKp11978chromosome.
previous study we could not detect the presence of the Theexchange ofgeneticmaterial between various pro-
pat-1 gene, which is encoded on the Cmm plasmid [4]. karyotes is well known and has been extensively studied
Even though we did not detect the complete pCM2, over the last few decades [33-37]. Although it was dem-
some of the genes originally encoded on this plasmid onstratedforbacteriathatthegeneexchangeisobserved
were found in LMG 26808 (Additional file 1: Table S5). more frequently between closely related genera with a
Except for two genes involved in the putative conjugal similar GC content and exhibiting high sequence simi-
transfer (pCM2_0013 and pCM_0019, coding for TraA larities there are examples of recent gene transfers be-
and TraG, respectively), all the remaining genes showed tween distantly related bacteria (e.g. Actinobacteria and
to code for hypothetical or putative secreted proteins. gammaproteobacteria) [38].
All of them were detected on contig 6 but the order in Conjugational transfer is considered the most efficient
which they were found in LMG 26808 did not match way of LGT [39,40] that contribute the most to the
the order demonstrated in pCM2. Moreover, there are spread of antibiotic resistance among different bacteria
more genes present on contig 6, some of which showed [41]. This type of LGT is widely encountered among
to be homologous to proteins from the Cmm chromo- various bacterial species and even between bacteria and
some (Cl_01961-Cmm_02708, Cl_01957-Cmm_01374). Archaea [42] on the one hand and between bacteria and
These observations may suggest that some of these plas- higher organisms such as Saccharomyces cerevisiae [43],
mid genes were incorporated in the genome of LMG or plants [44] on the other hand. Conjugational plasmid
26808. exchange was also observed within the genus Clavibac-
The observation that some genes from pCM2 that ter in which the endophytic CMM100 strain (cured of
were expressed during tomato infection by Cmm [31] pCM1 and pCM2 plasmids) was able to acquire these
had orthologs found in LMG 26808, might suggest that plasmids from other pathogenic Cmm strains and re-
although their function is unknown, they may be essential store pathogenicity [45]. Furthermore, transformation
fornon-pathogenicClavibacteraswellaspathogenicCmm experiments carried out with Clavibacter xyli subsp.
strains(Additionalfile1:TableS5).Furtherinvestigationis cynodontis (currently reclassified to the genus Leifsonia)
Załugaetal.BMCGenomics2014,15:392 Page8of14
http://www.biomedcentral.com/1471-2164/15/392
demonstrated the possibility to acquire an IncP plasmid factors in recent reports [30,31,52,53]. The genome of
fromEnterobacteriaceaebythisGram-positiveActinobac- LMG26808didnotcontainthemostprominentvirulence
teria, which provided another evidence of conjugational factors pat-1 and celA. Their absence may be directly
transferbetweendiversetaxa[46]. linkedwiththeabsenceofthepCM1andpCM2plasmids
Klebsiella pneumoniae strains were found in many im- in the non-pathogenic Clavibacter. However, southern
portant crops such aspotato, maize, soybean, cotton and hybridization experiments with plasmid fragments con-
tomato[47,48].Manyofthesestrainscarryplasmidsthat taining virulence factors showed that in some plasmid-
contain antibiotic resistance genes and possess the conju- freepathogenicCmmstrainsthesevirulencedeterminants
gationtransfersystemswhichenablethegenemobilization hadhomologuesonthechromosome[30].
and exchange among and outside Enterobacteriaceae and A proteomic study of tomato-Cmm interaction identi-
other bacterial genera [49]. Some genes encoded on the fiedseveralbacterialproteinswithaputativeroleinsignal
Klebsiellapneumoniaeplasmidsexhibithighsimilaritiesto perception, transduction and response to impulses. They
regions found previously in Escherichia coli and Yersinia belongtotwo-componentsystemproteins,transcriptional
genomes,implyingthatthereisanactivegeneticexchange regulators and other DNA binding proteins. They are be-
amongstrainsofthesegenera[50]. lieved to play a role in sensing the tomato plant environ-
Although an acquisition by LMG 26808 of a relatively ment and initiating pathways, possibly leading to disease
large plasmid originating most probably from a member development [31]. All putative genes encoding proteins
of Enterobacteriaceae/Klebsiella relative by LMG 26808 thatare probablyinvolved in signalexchange betweento-
(Additional file 1: Figure S2) was unexpected and unpre- matoandbacteriumcouldbeidentifiedinthegenomese-
cedented, a similar occurrence was previously reported quenceofLMG26808(Additionalfile1:TableS6).
by Baltrus and coworkers. They detected a recent acqui- As a non-pathogenic bacterium, LMG 26808 was hy-
sition of a megaplasmid by two cucumber isolates of pothesizedtocontainlessgenomicinformationforhydro-
Pseudomonaslachrymans [51]. It was suggested that this lytic enzymes that are known to be expressed during
acquisition resulted from an important ecological shift tomato infection with Cmm [31]. As expected, the most
across closely related Pseudomonas members and that important group containing genes coding for secreted
the plasmid-encoded genes may be advantageous for the proteases from Pat-1 family was largely absent in LMG
recipientbacteria. 26808 (Additional file 1: Table S6). Additional pat-1
AsKlebsiellapneumoniaeandClavibacterstrainsthrivein homologues encoded on the pCM2 plasmid (plasmid
thesameenvironmental niche (associated withtomato)and homologs of pat-1, phpA and phpB) were also absent.
because of examples of possible genetic material exchange From seven genes encoding putative serine proteases
between distantly related bacteria we can hypothesize that chpA-chpG (chromosomal homologs of pat-1) [54] only
theacquisitionofplasmid sequencesencodingantibioticre- sequences similar to chpF and chpG were detected (chpF
sistancegenesmightprovideacompetitiveadvantageforthe and chpG are orthologs with nucleic acid sequence simi-
non-pathogenicClavibacterstrainLMG26808. larity of 69.1% and amino acids sequence similarity of
68%). Both these sequences, however, matched to the
Non-pathogeniclifestyle same region and a reciprocal best BLAST hit confirmed
Non-pathogenic Clavibacter strains from tomato seeds the presence of only chpF. Interestingly, the lack of chpG
tested in the previous study [4] as well as other strains may be a possible explanation for the disability of LMG
tested by Jacques and coworkers [3] did not introduce 26808 to produce a HR in nonhost plants since the chpG
any disease symptoms when tested on tomato plants. mutant in Cmm was unable to cause an HR in Mirabilis
Possible explanations for the non-pathogenic nature of jalapa [8]. Moreover, the low colonization efficiency of
these strains are i) the lack of two plasmids present in LMG 26808 could be attributed to the lack of the chpC
pathogenic Cmm and carrying virulence factors, ii) the gene. A chpC mutation in the pathogenic Cmm NCPPB
absenceofthepathogenicityisland andiii)a significantly 382resultedinadrasticreductionincolonizationabilities
lower number of genes coding for extracellular hydrolytic in tomato plants [8,55]. Pseudogenes chpA, chpB and
enzymesincludingseveralimportantserineproteases,glyco- chpD were not found in LMG 26808. The family of chp
sylhydrolasesandtheplantcellwall-hydrolyzingenzymes. genes is important for plant-pathogen interaction in
In pathogenic Cmm, main virulence factors cel-A and Cmm, but probably also in Cms where four orthologs
pat-1, encoded on pCM1 and pCM2, respectively, are re- were found. Cmn genome had no orthologs of these
quiredtoinducediseasesymptoms(wiltingandcanker)in genes.
tomatoplants[52,32].Moreover,genescodingforthepro- The majority of members of secreted serine proteases
duction of extracellular enzymes, such as endoglucanase, of the Ppa family (PpaA-PpaJ) that are encoded in sev-
polygalacturonase,xylanase,serineproteasesandotherse- eral different loci on the chromosome and on pCM1
cretedproteinshavebeenimplicatedaspossiblevirulence could not be found in LMG 26808. Orthologs of ppaI
Załugaetal.BMCGenomics2014,15:392 Page9of14
http://www.biomedcentral.com/1471-2164/15/392
and ppaF were found in LMG 26808. Cms ATCC 33113 LMG 26808 suggesting that their function might not be
contained orthologs of ppaB1, ppaB2, ppaF, ppaI, ppaA restricted to disease development alone. The glycosyl hy-
in the chromosome and ppaC on pCS1 plasmid. On the drolasesarenotconsideredas bonafide virulencefactors,
contrary Cmn NCPPB 2581 had only one ortholog of but as reflecting the adaptation to the differing compos-
ppaF. Since ppaA and ppaC genes are important for ition of nutrients in planta allowing the survival inside of
plant colonization [8] and they were absent in LMG theplant.
26808, it can be another evidence why non-pathogenic Very important functions involved in transport and
Clavibacter strains are poorly colonizing tomato plants. metabolism are linked to the presence of ABC and other
Indeed, secreted serine proteases studied in pathogenic transporters that ensure the uptake of amino acids,
Cmm are thought to presumably facilitate the inter- metals, sugars, oligopeptides, etc. Some of these trans-
action between Cmm and its host plant and are believed porters that were expressed during tomato infection by
toplayafunctioninpathogenicitybyapossibleutilization Cmm may utilize plant molecules to support its metab-
of plant proteins [31]. Their lack might imply that inter- olism. ThegenesfoundinthegenomeofLMG26808that
action between LMG 26808 and tomato is actually very codefortransportersthatareknowntobeexpresseddur-
limited. TomA gene of Cmm NCPPB 382 (CMM_0090), ing plant infectionby Cmm are listed in Additional file 1:
encodingendo-1,4-betagalactosidaseinvolvedindetoxifi- Table S7. Interestingly, only five orthologs of fifty seven
cation of the alfa-tomatine, had orthologs in three other transporters could not be found in LMG 26808. Further-
Clavibacter genomes. However, the similarity based on more, the genome of non-pathogenic Clavibacter con-
the amino acid sequence was rather low (coverage (%)/ tainedadditionaltransportersthatwerenotpresentinthe
identity (%): 47/24 in Cms, 47/23 in Cmn and 47/22 in pathogenic Cmm genome (Additional file 1: Table S2).
LMG26808). Some of them are supposed to play a role in the active
Genes coding for subtilases sbtA, sbtB and sbtC are drug transport and cell protection from toxic metabolites
known to be secreted during the plant infection [31]. (C_03094 and Cl_03219). Another very important ex-
Orthologs of these three subtilases genes were found in ampleofadditional ABC transporters in the genome of
all four Clavibacter genomes. Sbt proteins of Cmm are LMG 26808 (not found in other three Clavibacter ge-
highly similar to different tomato subtilases, some of nomes) aretransportersinvolvedinironcomplextransport
which have been associated with wound formation and (ABC-typeFe3+siderophoretransportsystemCl_02679and
pathogen responses [56] and may play a role in the dis- ABC-type cobalamin/Fe3+−siderophores Cl_ 02677)
ease development. Because they are present in the non- (Additional file 1: Table S8). An alternative iron uptake
pathogenic Clavibacter strain their function probably system found in LMG 26808 might be advantageous in
cannot be solely associated with the disease develop- an iron deficient environment. This data suggests that
ment. Cellulases and pectinases are the most important LMG 26808 is probably able to utilize a broad variety
enzymes degrading plant cell walls. In many bacteria of compounds to maximize its survival changes. Many
genes encoding these enzymes were found to be viru- environmental bacteria were shown to contain a high
lence determinants [57]. In the genome of LMG 26808 number of transporter genes in support of an environ-
genes for pectate lyases, pelA1 and pelA2 and cellulase mental lifestyle [60].
celA were not found. However, another cellulase celB, Observations described above correlate well with the
the polygalacturonase pgaA (whose substrate is pectin), initial assumptions that suggested that non-pathogenic
xysA (whose substrate is β-1,4-xylan) and an arylesterase Clavibacter strains must have lost or never contained
(which hydrolyzes ester bonds between hemicelluloses prominent virulence determinants responsible for dis-
and lignin) [58] were present in LMG 26808 (Additional ease induction in tomato plants. These hypotheses were
file 1: Table S6). These findings support the thesis that partially underpinned by similar findings in another draft
the non-pathogenic Clavibacter strain is probably less genomeofnon-pathogenicClavibacterLMG26811,which
efficient in digesting pectins and cellulose into simpler lacks the majority of virulence factors including two main
by-products than the pathogenic Cmm that is equipped determinants.Italsocontainedlessplantcelldegradingen-
with manyvariousplant celldegradingenzymes. zymes than pathogenic Cmm NCPPB 382 (data not
Enzymes from a large group of glycosyl hydrolases shown). Furthermore, the comparative genome analysis
(GH) which hydrolyze the glycosidic bond between two of LMG 26808 and Cmm NCPPB 382 revealed that
carbohydrates or between a carbohydrate and a noncar- some putative virulence factors, determined based on
bohydrate molecule[59]werealsoexpressedduringplant expression levels obtained from the proteomic study of
infection of Cmm. Therefore, many of them are assumed tomato-Cmm interaction [31], were also present in
to be potential virulence factors that can hydrolyze sub- LMG 26808, which may indicate their redundant func-
strates of plant origin [31]. Ourresults demonstrated that tions and suggest that they are not critical for Cmm
genes for the majority of these enzymes are present in virulence.
Załugaetal.BMCGenomics2014,15:392 Page10of14
http://www.biomedcentral.com/1471-2164/15/392
Survivalintheenvironmentalniche and YoeB T-A genes were found in many bacterial ge-
Non-pathogenic Clavibacter strain LMG 26808 was iso- nomes and sometimes more than one copy per genome
lated from tomato seeds yet knowledge on its environ- [63]. It was demonstrated that T-A systems are present
mental niche is largely lacking. Ecological niche(s) and onlyinenvironmentalandfree-livingorganismsandwere
transmission routes have not yet been found. Prelimin- not detected in intracellular bacteria [63]. The BLASTp
ary results with colonization experiments showed poor analysis of YefM and YoeB genes from LMG 26808 re-
colonization of vascular tissues of tomato and seemingly vealedhighsimilaritiestoproteinsfromRhodococcuspyri-
lower survival potential of LMG 26808 in comparison to dinivoransAK37andMicrobacteriumtestaceumStLB037,
Cmm [4]. The HR was not induced in Mirabilis jalapa respectively. The T-A system found in LMG 26808 was
(J. Van Vaerenbergh, data not published), indicating that not present in the pathogenic Cmm,but YefM(peg.1235)
non-pathogenic Clavibacter strains do not contain genes was present in another non-pathogenic Clavibacter strain
that would be recognized by the plant to trigger the ac- LMG 26811 (data not shown). Interestingly, Cmn con-
tive plant defense response. tained another putative toxin-antitoxin system. T-A sys-
Antibiotic resistance. In the genome of LMG 26808 temsarenotessentialfornormalcellgrowth,nevertheless
several additional genes responsible for antibiotic resist- theyarepresentinmanybacteriaandArchaea[63].Based
ance were detected (Additional file 1: Table S2). They on the frequency of T-A systems, it was suggested that
coded for beta-lactamases (Cl_03263, peg.1233, peg.1766, they play subtle roles that are advantageous for cell sur-
peg.1776), chloramphenicol acetyltransferase (Cl_03044) vivalintheirnaturalhabitats.Toxinsmayfacilitatecellular
and tetracycline efflux protein TetA (peg.1764). They adaptation of an organism to changing environments by
showed the highest similarity togenes found in Klebsiella slowingdownits cell growth,inhibitingits cell growth,or
pneumonia, Escherichia coli and Salmonella enterica causing some ofits cells to die [64]. Itis possible that the
suggesting that they could originate from these bacteria. presence of a T-A system in the LMG 26808 genome
In addition to the above genes, the genome of LMG increases the fitness of this bacterium in the occupied
26808 contains two drug efflux transporters (Cl_03219, environmental niche. Differences in the detected toxin-
Cl_03094) not found in pathogenic Cmm NCPPB 382. antitoxin systems in particular Clavibacter subspecies
Interestingly,LMG26808containedglyoxalase/bleomycin might be attributed to different ecological niches and
resistance protein (Cl_03100), which probably constitutes inhabited hosts.
the resistance to bleomycin-antibiotic produced by some ErrorproneUmuDCoperon.SOSmutagenesisresponse
Streptomycesstrains[61]. inbacteriaincludeserror-proneanderror-freeDNAdam-
Thepresenceofadditionalacetyltransferasesmightsug- age repair responses that are activated after exposure to
gest that LMG 26808 exhibits broad resistance to certain different antibiotics, chemical compounds or radiation
antibioticsassomeoftheacetyltransferases(GNATsuper- [65]. In Escherichia coli UmuDC proteins are involved in
family)catalyzetheselectiveacetylationofoneofthefour error-pronebypassofUVlesionsandUmuCproteinspos-
amino groups found on a diverse set of aminoglycosides sess DNA polymerase activity. In the SOS process many
withantibioticproperties[62].Acetylationreducestheaf- genesgetinducedandtheirproductsareinvolvedinDNA
finity of these compounds for the acceptor tRNA site on repair, replication and cell cycle control in order to repair
the 30S ribosome. As a consequence, bacteria expressing the DNA damage [66]. The genes coding for this operon
these genes are resistant to some aminoglycosides with were found in LMG 26808 and also in another non-
antibiotic properties. The ability to cope with antibiotics pathogenic Clavibacter, LMG 26811 (data not shown),
producedbyorganismswithwhichnon-pathogenicClavi- implying that their cells might have higher abilities to
bacterstrainsshare the environmental niche is a signifi- recover after exposures to UV and/or other types of
cant adaptive advantage. The growth of Clavibacter chemicals retrieved during the seed certification proce-
strains in culture is often inhibited by other faster grow- dures. Some of the sequences coding for genes of
ing organisms. Therefore, the presence of genes coding UmuC operon (e.g. peg.1211) and antibiotic resistance
for antibiotic resistance might be the reason why non- genes (e.g. Cl_03263) described above are associated
pathogenic Clavibacter strains are more frequently en- with the plasmid and/or low GC regions (Additional
countered and isolated from the semi-specific medium file 1: Table S2).
duringthetomatoseedcertification. The extracellular polysaccharide (EPS). The genomes
Toxin-antitoxin system. The presence of the toxin- of all four analyzed Clavibacter strains contained four
antitoxin (T-A) genes (YefM Cl_00198, peg.1235 and gene clusters involved in exopolysaccharides production
YoeB Cl_00197) in thegenomeofnon-pathogenicClavi- (Additional file 1: Table S9). The EPS production in
bacter is intriguing and raises questions concerning their LMG 26808 is expected to effectively occur since all
originandpotentialfunctionsinrelationtothephysiology genes involved in that process are functional (no frame-
of the bacterium (Additional file 1: Table S2). The YefM shifts, no pseudogenes). LMG 26808 contains almost a