A TALE of Transposition: Tn3-Like Transposons Play a Major Role inthe Spread of Pathogenicity Determinants of Xanthomonas citri andOther Xanthomonads

Rafael Marini Ferreira,a Amanda Carolina P. de Oliveira,a Leandro M. Moreira,b,c José BelasqueJr.,d Edith Gourbeyre,e

Patricia Siguier,e Maria Inês T. Ferro,a Jesus A. Ferro,a Michael Chandler,e Alessandro M. Varania

Departamento de Tecnologia, Faculdade de Ciências Agrárias e Veterinárias de Jaboticabal, UNESP, Universidade Estadual Paulista, Jaboticabal, São Paulo, Brazila;Departamento de Ciências Biológicas (DECBI), Instituto de Ciências Exatas e Biológicas (ICEB), Campus Morro do Cruzeiro, Universidade Federal de Ouro Preto, Ouro Preto,Minas Gerais, Brazilb; Núcleo de Pesquisas em Ciências Biológicas (NUPEB), Universidade Federal de Ouro Preto, Ouro Preto, Minas Gerais, Brazilc; Departamento deFitopatologia, Escola Superior de Agricultura Luiz de Queiroz, Universidade de São Paulo, Piracicaba, São Paulo, Brazild; Laboratoire de Microbiologie et GénétiqueMoléculaires, CNRS 118, Toulouse, Francee

ABSTRACT Members of the genus Xanthomonas are among the most important phytopathogens. A key feature of Xanthomonaspathogenesis is the translocation of type III secretion system (T3SS) effector proteins (T3SEs) into the plant target cells via aT3SS. Several T3SEs and a murein lytic transglycosylase gene (mlt, required for citrus canker symptoms) are found associatedwith three transposition-related genes in Xanthomonas citri plasmid pXAC64. These are flanked by short inverted repeats (IRs).The region was identified as a transposon, TnXax1, with typical Tn3 family features, including a transposase and two recombina-tion genes. Two 14-bp palindromic sequences within a 193-bp potential resolution site occur between the recombination genes.Additional derivatives carrying different T3SEs and other passenger genes occur in different Xanthomonas species. The T3SEsinclude transcription activator-like effectors (TALEs). Certain TALEs are flanked by the same IRs as found in TnXax1 to formmobile insertion cassettes (MICs), suggesting that they may be transmitted horizontally. A significant number of MICs carryingother passenger genes (including a number of TALE genes) were also identified, flanked by the same TnXax1 IRs and delimitedby 5-bp target site duplications. We conclude that a large fraction of T3SEs, including individual TALEs and potential pathoge-nicity determinants, have spread by transposition and that TnXax1, which exhibits all of the essential characteristics of a func-tional transposon, may be involved in driving MIC transposition. We also propose that TALE genes may diversify by fork slip-page during the replicative Tn3 family transposition. These mechanisms may play a crucial role in the emergence ofXanthomonas pathogenicity.

IMPORTANCE Xanthomonas genomes carry many insertion sequences (IS) and transposons, which play an important role intheir evolution and architecture. This study reveals a key relationship between transposons and pathogenicity determinants inXanthomonas. We propose that several transposition events mediated by a Tn3-like element carrying different sets of passengergenes, such as different type III secretion system effectors (including transcription activation-like effectors [TALEs]), were deter-minant in the evolution and emergence of Xanthomonas pathogenicity. TALE genes are DNA-binding effectors that modulateplant transcription. We also present a model for generating TALE gene diversity based on fork slippage associated with the repli-cative transposition mechanism of Tn3-like transposons. This may provide a mechanism for niche adaptation, specialization,host-switching, and other lifestyle changes. These results will also certainly lead to novel insights into the evolution and emer-gence of the various diseases caused by different Xanthomonas species and pathovars.

Received 16 December 2014 Accepted 28 December 2014 Published 17 February 2015

Citation Marini Ferreira R, de Oliveira ACP, Moreira LM, Belasque J, Jr, Gourbeyre E, Siguier P, Ferro MIT, Ferro JA, Chandler M, Varani AM. 2015. A TALE of transposition: Tn3-liketransposons play a major role in the spread of pathogenicity determinants of Xanthomonas citri and other xanthomonads. mBio 6(1):e02505-14. doi:10.1128/mBio.02505-14.

Editor Pascale F. Cossart, Institut Pasteur

Copyright © 2015 Marini Ferreira et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-ShareAlike 3.0Unported license, which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

Address correspondence to Michael Chandler, mike@ibcg.biotoul.fr, or Alessandro M. Varani, amvarani@fcav.unesp.br.

This article is a direct contribution from a Fellow of the American Academy of Microbiology.

The genus Xanthomonas represents an important group of bac-terial plant pathogens with global economic agricultural im-

pacts. Xanthomonas species are clearly distinctive in their pathol-ogy and plant host ranges (1). Among the more economicallyimportant are Xanthomonas citri, causing citrus canker, Xan-thomonas oryzae pathovars, causing bacterial blight and leaf streakin rice (Oryza sativa) (2, 3), Xanthomonas campestris pv. vesicato-

ria and Xanthomonas arboricola pv. pruni, responsible for bacte-rial spot on a wide range of commercial plants, including pepper,tomato, and Prunus species (1, 4), and Xanthomonas fuscanssubsp. fuscans and Xanthomonas axonopodis pv. manihotis, re-sponsible for common bean blight and bacterial cassava blight,respectively (1, 5).

Citrus canker is an important disease affecting many citrus

RESEARCH ARTICLE crossmark

January/February 2015 Volume 6 Issue 1 e02505-14 ® mbio.asm.org 1

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

species. It is endemic in many parts of the world and has a severeeconomic impact on the citrus-producing industry (2, 6). Asiatictype A citrus canker caused by X. citri subsp. citri pathotype Ainfection is the most widespread and destructive citrus canker type(2). Types B and C result from infection by different strains of X.fuscans subsp. aurantifolii, and these are restricted to South Amer-ica (2, 7). Citrus canker can be easily transmitted by wind-blownrain. Infection causes hyperplasia and necrosis, leading to cell lysis(8). As is the case for other diseases caused by the Xanthomonasgenus, citrus canker involves a complex network of interactionsimplicating bacterial type II (T2SS) and type III secretion systems(T3SS) and biofilm formation (8). As observed in other pathosys-tems (9, 10), T3SS proteins are an important part of the Xan-thomonas pathogenesis repertoire (11, 12), creating a bridge be-tween the bacterium and the plant host cell and permitting theinjection of bacterial T3SS effector proteins (T3SEs) into the plantcell cytoplasm (6). Over 60 type III effector proteins (http://www.xanthomonas.org/t3e.html), including transcriptionactivator-like effectors (TALEs; see below) (13, 14), have beenidentified.

Xanthomonas genome sequences have proved a rich source ofinformation concerning the evolution, lifestyle, and pathogenicityof members of the genus. The genome of X. citri subsp. citri patho-type A was deciphered over a decade ago (6), and the genomes offour other citrus canker-related species (X. fuscans subsp. auran-tifolii strains ICPB 11122 and ICPB 10535, X. citri subsp. citristrain Aw 12879, and X. axonopodis pv. citrumelo strain F1) (2, 7)were partially or completely sequenced recently. Those of severaladditional species and pathovars are currently available or in pro-cess (http://www.xanthomonas.org).

An intriguing feature of the xanthomonads is their extraordi-nary genome plasticity and highly mosaic architecture (8), largelydue to genomic rearrangements involving genomic islands andinsertion sequences (IS) (15, 16). Indeed, IS are important land-marks of some Xanthomonas genomes. X. oryzae pv. oryzae and X.oryzae pv. oryzicola are rich in IS transposases, which can repre-sent up to 10% of their genomes, whereas in X. citri subsp. citristrain 306 and X. campestris pv. vesicatoria, IS transposases repre-sent less than 2% of the genome (17).

It has been noted that several Xanthomonas T3SEs are flankedby short (�100 bp) inverted repeat (IR) sequences (18) that formgenetic structures typical of mobile insertion cassettes (MIC) (19).These are composed of passenger genes flanked by IS-related in-verted repeats. MICs do not encode their own transposases butcan be activated by a cognate transposase from a related IS (19). Ithas been suggested that these effector modules may have beentransmitted by transposition and lateral gene transfer (18), al-though no specific IS family repeats have been reported to bepredominantly associated with T3SEs (3).

The T3SEs found in the MIC structures include TALE genes.The products of these are transcription factors which modulateplant host gene expression to the benefit of the pathogen by inter-acting with various host cell factors and interfering with cellgrowth and the defense response (20, 21). Members of this class ofDNA binding protein are found in a number of xanthomonadsand are a key pathogenic feature (8, 13, 18, 22). Host DNA se-quence specificity is determined by conserved near-identical tan-dem repeats, usually 34 amino acids long, that include a repeat-variable diresidue (RVD) at positions 12 and 13 (23). Each RVDrecognizes a given nucleotide, and the string of tandem repeats

therefore recognize a specific DNA binding sequence (20, 23, 24).TALEs are becoming powerful DNA-targeting tools in biotech-nology (23, 25).

One of the largest and most diverse transposon groups is theTn3 family. Members range in size and structure from 3,000 bp tomore than 50,000 bp. All include a transposase (TnpA) of the DDEmotif (RNase H fold) class (26, 27), and most carry a site-specificrecombinase system, as well as passenger genes for functions suchas resistance to antibiotics and heavy metals like mercury (28, 29)or copper (30), which have been sprayed for many years on vege-table and fruit crops to limit the spread of plant-pathogenic bac-teria and fungi (31). They transpose by replicative replicon fusion,generating a cointegrate with a copy of the transposon in directorientation at each junction between the fused donor and targetreplicons (26). Transposition terminates by recombination (res-olution) between the two directly repeated Tn copies at a partic-ular site, res or rst, in a reaction catalyzed by a transposon-encodedsite-specific recombinase. In many Tn3 family members, this is aserine recombinase, the resolvase (TnpR) (26). In others, it may bea single tyrosine recombinase related to phage integrases (TnpI)(32) or a pair of genes encoding TnpT and TnpS (33). TnpS isrelated to the bacteriophage P1 Cre tyrosine recombinase, whereasTnpT shows no similarity to any known protein. Shorter deriva-tives have also been identified that encode only the transposaseand are devoid of the resolvase (e.g., ISVsa19, ISShfr9, ISBusp1,and IS1071D) (34). These therefore resemble simple bacterial ISs.

In the studies reported here, high-quality IS annotation cou-pled with comparative genomics revealed an important group ofxanthomonad Tn3 family transposons and derivatives with asso-ciated T3SEs in general and TALE genes in particular. This in-cludes a complete Tn3 family transposon in the X. citri subsp. citripathotype A genome and a large number of derivatives in plas-mids and chromosomes of other Xanthomonas species.

The X. citri subsp. citri pathotype A Tn3-like transposons carryT3SE passenger genes, together with a highly conserved mureinlytic transglycosylase gene (mlt). The analysis also underlined theimportance of related transposon-like structures in the transmis-sion of other pathogenicity factors between different xanthomon-ads. Not only were the four previously known X. citri subsp. citripathotype A plasmid (pXAC33 and pXAC64) TALE genes locatedin a MIC structure flanked by the same Tn3-like repeats as theTnXax1 transposon, but many other TALE and pathogenicitygenes from other xanthomonads were observed to be associatedwith similar structures. These results provide strong evidence that,at least in the case of X. citri subsp. citri pathotype A and otherclosely related species, key pathogenicity and virulence genes arespread by a Tn3-like transposition mechanism and use (or, in thecase of X. oryzae, have used) conjugative plasmids as intercellularvectors. We propose that the transposition process can inherentlygenerate TALE gene diversity and is a key agent in modulatingbacterial virulence and host specificity in this group of plantpathogens.

RESULTSTn3 family elements in the pXAC64 plasmid and the distribu-tion of their specific IRs among the xanthomonads. PlasmidpXAC64 and closely related plasmids are widely distributed inxanthomonads (4, 7), including X. oryzae strains isolated fromNorth America, although they appear to be absent in the se-quenced X. oryzae and X. oryzae pv. oryzicola genomes (3, 35).

Marini Ferreira et al.

2 ® mbio.asm.org January/February 2015 Volume 6 Issue 1 e02505-14

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

This distribution presumably occurred by conjugal transfer, sincethe plasmids carry a transfer operon (a type IV secretion system[T4SS]) with a relaxase gene related to trwC of plasmid R388 andtraI of the classical conjugative F plasmid (36, 37).

Noël et al. (38) identified this region in the X. citri subsp. citripathotype A plasmid pXAC64 (6) and noted that the xopE2 effec-tor and mlt genes were associated with three transposon-relatedgenes and that the entire structure was flanked by IR sequences.However, the nature of these transposon-related genes was notaddressed.

We have revisited this question during our ongoing studies ofX. citri subsp. citri pathotype A pathogenesis. We found that notonly do the flanking IR sequences correspond to the highly con-served terminal IR of members of the large Tn3 transposon familybut the associated transposon-related genes are also typical of Tn3family members. This potentially transposable structure, whichwe have called TnXax1, has a length of 8,944 bp (Fig. 1A).

The TnXax1 IR sequences are shown in Fig. 1B. They are 92 bplong with 72 bp of identity. Although this is rather long for Tn3

IRs, several other family members are also known to carry long IRs(ISfinder database). The TnXax1 IRs begin with GAGGG insteadof the more common GGGGG. These full-length TnXax1-like IRsare generally restricted to the Gammaproteobacteria, particularlyto Xanthomonas species. There are at least 264 different occur-rences of these in the Xanthomonadales (see Table S1 in the sup-plemental material), often forming part of TnXax1 derivativesand putative MIC elements, although there are several examples ofsolo IRs. Intriguingly, 6 full-length TnXax1-like IRs are also foundin a unique sequenced betaproteobacterial genome, that of thecucurbit pathogen Acidovorax avenae subsp. citrulli (Table S1),flanking a XopJ2 effector (see below). Tn3 family IRs withGAGGG tips occur in several other Tn3-like transposons (Fig. 1B;Table S2). However, the IRs are all significantly shorter than thoseof TnXax1.

TnXax1 transposition functions. The putative transpositionfunctions of TnXax1 (Fig. 1A) are closely related to those ofTnPa43. The longest open reading frame (ORF), XACb0008, re-sembles a Tn3 transposase (62% identity at the protein level to

FIG 1 (A) Genetic organization of the canonical TnXax1 located in X. citri subsp. citri strain 306 plasmid pXAC64. Genes are indicated by colored boxes, withthe direction of transcription shown by the arrowheads. Transposition-related genes are shown in purple, and passenger genes in red and blue. The presumedresolution site is located between the tnpT and tnpS genes, as observed in the transposable element Tn4651. This includes two palindromes: IR1 and IR2 areprobably part of the core site at which recombination occurs, recognized by the TnpS recombinase, whereas IRa and IRb are potential binding sites for TnpT. Theterminal inverted repeats (IRL and IRR) are shown as black triangles. The convention used for orientation of the transposon is that the transposase (TnpA) istranscribed from left to right. TnpA is closely related to other Tn3-like elements from Shewanella (ISSod9; 47% identity), Acinetobacter (ISAcsp1; 46%),Arthrobacter (ISArsp6; 43%), Azospirillum (ISAzs17; 41%), Salmonella (ISSwi1; 42%), and Nostoc (ISNpu13; 43%) species and to the mercury resistancetransposon Tn5044 from X. campestris (84%). (B) IRs identified from other Tn3-like transposable elements. All include the GAGGG tips sharing sequencesimilarities to TnXax1 IRs. These are TnPa43 from Pseudomonas aeruginosa, TnStma1 from Stenotrophomonas maltophilia strain D457, TnTin1 from Thiomonasintermedia strain K12, and TnXca1 from X. campestris pv. vesicatoria plasmid pXCV183.

Transposition and Pathogenicity in Xanthomonas

January/February 2015 Volume 6 Issue 1 e02505-14 ® mbio.asm.org 3

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

that of TnPa43). The second, XACb0009, shows high similarity tothe TnPa43 tyrosine recombinase, TnpS (49% identity), and thethird, XACb0010, is related to the Tn4651 auxiliary cointegrateresolution protein T (TnpT) (43% identity) (33). TnpS and TnpTare oriented divergently in both Tn4651 and TnXax1.

For Tn4651, the intergenic region between the divergent tnpSand tnpT genes is 188 bp and includes two short palindromeswithin a 136-bp functional resolution recombination site (33), theTn3 family rst site (26). One pair are putative binding sites forTnpT, whereas the other pair are part of the core site at whichrecombination occurs (33).

The TnXax1 tnpT-tnpS intergenic region is 193 bp and alsoincludes two pairs of palindromic sequences; one pair, IR1 andIR2, are 14 bp with a potential core recombination site of 5 bp(Fig. 1A), and the other pair are two short IRs, IRa and IRb, whichcould represent a putative TnpT binding site (33).

Other TnXax1-related derivatives. A TnXax1-related struc-ture was previously identified in plasmid pXap41, which is presentin a variety of X. arboricola pv. pruni species (4). We have identi-fied a number of additional TnXax1-related structures in a varietyof xanthomonads (Fig. 2), both in plasmids and chromosomes.One striking feature of these is that, while the right ends are quitesimilar (Fig. 2B), the left ends exhibit a large degree of variability(Fig. 2A). Clearly the left end can act as a platform for accommo-dating various passenger genes, as well as different IS and otherTn3 family transposons. Additional IR copies are also observed insome cases, permitting the transposition of alternative structuresusing alternative combinations of ends. Many contain mutationswhich would render them unable to transpose (e.g., an in-framestop codon in tnpT and the absence of a right IR [IRR] in X. camp-estris pv. vesicatoria strain 85-10 chromosomal copy B) (Fig. 2A).In addition, it seems probable that intertransposon recombina-tion has also played a role in generating these structures. For ex-ample, the X. citri subsp. citri strain 29-1 chromosomal TnXax1derivative (called XccA-29-1 herein) includes xopE2 and xopAIeffector genes and two left IRs (IRLs). It resembles a chimera de-rived by recombination between the TnXax1 from the pXAC64plasmid and the X. citri subsp. citri pathotype A chromosomalTnXax1 copy (Fig. 2A).

Passenger genes: effector proteins and Mlt. These transposon-like structures were initially identified in regions containing anumber of T3SE pathogenicity genes. The annotated regionsshown in Fig. 2 show that these occur as passenger genes withinmany of the TnXax1 derivative structures. These are principallythe effector genes xopE2, xopE3, xopAI, xopC, and mlt, which havereceived much attention over the past few years. They are centralfor pathogenicity (38–40).

The mlt gene is not considered to be a T3SE family member (4),but except for a single case from X. campestris pv. vesicatoria 85-10chromosomal copy B, it is consistently found in these structures(Fig. 2B). Laia and coauthors previously suggested that Mlt may beimportant for T3SS apparatus assembly and activity (41). Thisconservation suggests that mlt is maintained by a significant selec-tive pressure. It is noteworthy that mlt homologues are also pres-ent in a variety of xanthomonads (see Tables S3 and S4 in thesupplemental material) and in other genera, such as Pseudomonasand Ralstonia. Mlt shares 63% sequence identity with HopAJ1from Pseudomonas syringae pv. tomato DC3000 (4, 41). AlthoughHopAJ1 is not a T3SS substrate, it probably enables the T3SS topenetrate the peptidoglycan layer in the bacterial periplasm and

deliver virulence proteins into host cells (42). In X. citri subsp. citripathotype A, it appears to be expressed specifically during in vivomultiplication, and its deletion reduces the severity of citrus can-ker (41, 43).

TnXax1 from plasmid pXAC64 and from the X. citri subsp. citripathotype A chromosome carry two different T3SE genes. Thatfrom pXAC64 includes xopE2, a putative transglutaminase thatacts on the plant cell plasma membrane, suggesting a role in viru-lence and in the suppression of the hypersensitive response of thehost (40). The X. citri subsp. citri pathotype A chromosome copyincludes xopAI, homologous to the P. syringae pv. tomato viru-lence factors hopU1, hopO1-1, and hopO1-2. These probably havea posttranslational modification function (44) (Fig. 2A). Interest-ingly, the two effectors share about 200 bp at their 5= ends. TheC-terminal end of XopAI shows some similarity to the arginineADP-ribosyltransferase family (PF01129). The chromosomal X.citri subsp. citri pathotype A TnXax1 derivative also includes anadditional 274 bp of noncoding DNA between xopAI and the IRL,whereas the derivative X. citri subsp. citri strain Aw 12879 has204 bp of noncoding DNA between xopE2 and IRL. Other differ-ences between the TnXax1 structures are much more complex,suggesting that a high level of recombination may be involved intheir formation (Fig. 2A).

Both X. axonopodis pv. citrumelo strain F1 TnXax1 relatives (Aand B) carry a disrupted xopC gene whose product is secretednormally and translocates into the plant cell (Fig. 2A) (38). ThexopC copies found in the X. axonopodis pv. citrumelo F1 ATnXax1 derivative are disrupted by IS404 and IS1389 insertions(Fig. 2A) that are not present in X. axonopodis pv. citrumelo F1 B.

As shown in Fig. 2 and in Table S2 in the supplemental mate-rial, many other passenger genes carried by these transposon-likestructures are also related to virulence.

TALE gene distribution among the Xanthomonadales: astory of transposition. Although several of the �264 different IRoccurrences (see Table S1 in the supplemental material) werefound to form part of TnXax1 derivatives, many occurred as in-verted repeats flanking passenger genes without associated trans-position genes and form putative MIC elements.

Analysis of the passenger genes neighboring the 92 bp IRshowed that many were TALE gene family members. TALE geneshave attracted attention by their simple and predictable specificDNA sequence recognition readout, which uses a number of re-peating peptide blocks, some of which include a dipeptide withspecific recognition properties for a given base (20, 24). In theiroriginal xanthomonad hosts, TALE proteins are transmitted tothe host plant cell nucleus, where they activate or inactivate im-portant genes, thus contributing directly to pathogenicity. Indeed,a single amino acid substitution in a TALE sequence can affectcitrus canker development (22). However, the detailed mecha-nism of transcriptional regulation by TALEs is still not fully un-derstood (20).

Previous studies had revealed that a number of these in differ-ent Xanthomonas strains are flanked by 62 bp IRs, and it was sug-gested that they may have been transmitted as mobile cassettes(18). Two were identified on each of the two X. citri subsp. citristrain 306 plasmids, pXAC33 (carrying pthA1 and pthA2) andpXAC64 (carrying pthA3 and pthA4) (18). The four genes areslightly different, with the principal variations located in the re-peating peptide blocks. The analysis performed here showed thatthey are flanked by the same 92 bp (rather than 62 bp) IR sequence

Marini Ferreira et al.

4 ® mbio.asm.org January/February 2015 Volume 6 Issue 1 e02505-14

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

FIG 2 Genetic organization of the left (A) and right (B) ends of the TnXax1-related structures found in other Xanthomonas species. Abbreviations for theTnXax1-related structures: Xap pXap41, X. arboricola pv. pruni, plasmid pXap41; XccA, X. citri subsp. citri pathotype A, chromosomal copy; TnXax1, X. citrisubsp. citri pathotype A, plasmid copy and canonical element; Xac F1 A and Xac F1 B, X. axonopodis pv. citrumelo strain F1, 2 chromosomal copies; Xcv A andXcv B, X. campestris pv. vesicatoria strain 85-10, chromosomal copies A and B; XccA(w), X. citri subsp. citri strain Aw 12879, chromosomal copy; XccA-29-1, X.citri subsp. citri strain 29-1, chromosomal copy; XccA-29-1 pXac64, X. citri subsp. citri strain 29-1, plasmid copy; XauB ctg 621_0, X. fuscans subsp. aurantifoliistrain ICPB 11122; XauC ctg 1147_0, X. fuscans subsp. aurantifolii strain ICPB 10535.

Transposition and Pathogenicity in Xanthomonas

January/February 2015 Volume 6 Issue 1 e02505-14 ® mbio.asm.org 5

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

as the Xanthomonas-specific TnXax1 IR. This suggests that theTn3-like element may provide transposition functions for theseMIC elements.

Boch and Bonas described 113 known TALEs in different Xan-thomonas species (18) (see also Bogdanove et al. [14]). Here, weanalyzed these genes together with newly identified TALE genesdeposited in the public databases, focusing principally on thosefrom complete and fully sequenced genomes. A maximum-likelihood tree constructed from the full-length DNA sequences of122 different TALEs is shown in Fig. 3. The topology shows well-defined groups, including two major groups (Fig. 3B and C) and

three minor groups (Fig. 3D, F, and G) composed of TALEs fromvarious X. oryzae strains (Fig. 3, purple, green, blue, and white), agroup composed of TALEs from X. citri, X. axonopodis, X. camp-estris pv. vesicatoria, and X. campestris (Fig. 3A, gray), and a grouprepresenting most of the X. oryzae pv. oryzicola BLS256 TALEs(pink) (Fig. 3E). A similar topology was obtained using the set offull-length proteins (data not shown). The general topology isconsistent with that obtained in other studies, which used nucle-otide sequences without the RVD repeats (14), concatenated N-and C-terminal portions of the proteins (45), or C-terminal re-gions alone (46).

FIG 3 Molecular phylogenetic analysis by maximum-likelihood method of the TALEs genes found in completely sequenced Xanthomonas genomes. Each TALEgene is shown with its respective genomic coordinates (for TALEs extracted from complete genomes) or GenBank accession number inside the brackets. Symbols:red diamonds, TALE genes associated with MIC structures; yellow triangles, TALE genes associated with a solo IR flanking one of their extremities; blue circles,genomes which carry TnXax1 or related structures; asterisks, MICs with the direct TACTC(G) target repeat; blue arrow, the TALE gene from X. fuscans subsp.fuscans strain 4834-R plasmids pla and plc. TALEs from different X. oryzae pathovars are highlighted as follows: blue, X. oryzae pv. oryzae strain PXO99A; green,X. oryzae pv. oryzae strain KACC 10331; purple, X. oryzae pv. oryzae strain MAFF311018; pink, X. oryzae pv. oryzicola strain BLS256. TALEs from X. citri,X. axonopodis, X. campestris pv. vesicatoria, and X. campestris are highlighted in gray.

Marini Ferreira et al.

6 ® mbio.asm.org January/February 2015 Volume 6 Issue 1 e02505-14

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

Association of TALE genes with TnXax1 IR. A substantialnumber of TALE genes from this collection (55 or 45%) (Fig. 3)proved to be associated with the TnXax1 IRs and to form MICstructures. These include both MICs with full-length IRs andMICs with IRs ranging from 33 bp to 97 bp but always maintain-ing the GAGGG tip (see Table S3 in the supplemental material).MIC-associated TALEs are found in a majority of the Asian X.oryzae and the X. citri/axonopodis/campestris/campestris pv. vesi-catoria group, suggesting that transposition by a Tn3 mechanismmay be an important driver of TALE dispersion and diversifica-tion in these groups. However, they are much less frequent in theX. oryzae pv. oryzicola strains.

In addition to the MIC structures, 20 solo IRs were identifiednext to TALEs from the X. oryzae pv. oryzae and X. oryzae pv.oryzicola groups, and a potential solo IR was identified in the X.citri/axonopodis/campestris/campestris pv. vesicatoria group (Fig. 3;see Table S1 in the supplemental material). Solo IRs may reflectprevious transposition or recombination events involving the ad-jacent TALE gene. The chromosomal TALE-containing MICs areparticularly notable in the Asian X. oryzae group (Fig. 4), where,assuming that there have been no major errors in genome assem-bly, there are small clusters of tandemly repeated MIC units. Fig-ure 4 and Fig. S1 show the distribution of all MIC structures andtheir association with the TALE genes and solo IRs. Table S3 liststhese elements in more detail. In all three X. oryzae strains, thetandemly repeated MIC units are distributed in a complex clusterorganization that is distinct for each strain (Fig. 4).

Moreover, the DNA located between the MICs of each clustershows a higher-than-average percentage of genes encoding trans-posases from other transposable element (TE) families, integrases,and phage-related functions and could represent a plasticity re-gion in which insertions and rearrangements do not adverselyaffect cell viability. Since more than one MIC structure and IR areassociated with each locus, transposition of alternative structuresusing different combinations of neighboring ends may be possi-ble. In addition, these X. oryzae genomes also carry other isolatedTALE genes and solo IRs which may also act as substrates forrecombination and transposition of alternative structures.

For some MICs, it was possible to identify 5 bp target duplica-tions that are commonly generated by Tn3 family elements, indi-cating that these insertions represent true transposition events(Fig. 3; see Table S3 in the supplemental material). These DRs arediverse in sequence (Table S3), but a particular sequence,TACTC(G), is shared by a number of MIC elements of all three X.oryzae strains, although the associated TALEs are different andgrouped in different X. oryzae subgroups (Fig. 3B, C, and G, aster-isks). This may indicate either that these insertions exhibit a pre-ferred target sequence or that the TALEs have evolved from asingle MIC which has been displaced and amplified and has di-verged as part of a larger structure (Table S3).

Relationship between TALEs from different xanthomonads.Not all TALE genes are included in well-defined MICs. The X.oryzae pv. oryzicola strain BSL256 carries many such copies(Fig. 3, pink). The X. oryzae pv. oryzicola tal5a without IRs islocated on the phylogenetic tree within a small group of X. oryzaeMIC-associated TALEs (Fig. 3D, blue, purple, and green) in abranch separate from the main X. oryzae groups (Fig. 3B and C)and more closely related to X. oryzae pv. oryzicola strain BSL256TALEs (Fig. 3E). Another subset of X. oryzae pv. oryzicola BSL256TALEs without IRs, tal2c, tal2b, and tal2h, forms part of the sec-ond X. oryzae TALE clade (Fig. 3G), which also includes MIC-associated TALEs (Fig. 3G, blue, purple and green). It is possiblethat X. oryzae pv. oryzicola BSL256 first acquired these TALEs bytransposition and has subsequently lost the IRs.

Although the majority of X. oryzae pv. oryzicola BSL256 TALEsare not associated with flanking IRs (Fig. 3C), three appear to havea single IR copy. This is also consistent with the notion of IR decay.

There are a number of unexpected incongruences in the phy-logenetic tree. In particular, X. fuscans subsp. fuscans strain4834-R carries both X. oryzae and X. citri/axonopodis/campestris/campestris pv. vesicatoria-related TALEs. These are plasmid lo-cated (5). The TALE from one of the three resident plasmids,X. fuscans subsp. fuscans plasmid A, is grouped with those of the X.oryzae strains (Fig. 3B, blue arrow), whereas that of X. fuscanssubsp. fuscans plasmid C is grouped with those of the X. citri/axonopodis/campestris/campestris pv. vesicatoria strains (Fig. 3A,

FIG 4 Linear genomic representation of the distribution of the IRs and tandemly repeated MIC clusters in X. oryzae genomes. Abbreviations: XooP, X. oryzaepv. oryzae strain PXO99A: XooM, X. oryzae pv. oryzae strain MAFF311018; XooK, X. oryzae pv. oryzae strain KACC 10331; XocB, X. oryzae pv. oryzicola strainBLS256.

Transposition and Pathogenicity in Xanthomonas

January/February 2015 Volume 6 Issue 1 e02505-14 ® mbio.asm.org 7

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

blue arrow). The presence of both types of TALE in differentclades is therefore most simply explained by plasmid-mediatedhorizontal transfer.

MICs that include other pathogenicity functions. In additionto TALE genes, MIC structures contain other T3SEs. The uniqueMIC of X. oryzae pv. oryzicola BLS256 includes several completeand partial TEs of different families and the effector gene xopU(see Table S3 in the supplemental material). This effector is exclu-sive to X. oryzae strains and is found in different MIC structures.Other potential effector proteins and pathogenicity functions areshown in Fig. S1 and Table S3. While MICs carrying TALE genestend to be rather simple, those carrying other potential effectorand pathogenicity factors appear more complex and can includemore than two IRs. They vary in length from 3 to 19 kb. Closelyrelated elements are present in different related strains. This is thecase for the two 16.2 kb chromosomal Mic A from X. citri subsp.citri ATCC 33913 and X. citri subsp. citri 8004 (which are invertedwith respect to each other in the genome) and the 18.2 kb X. oryzaeMIC J (strain MAFF311018) and MIC G (strain KACC 10331). Itshould also be noted that MIC B (X. oryzae strain PXO99A) isnearly identical to the MAFF311018 and KACC 10331 copies butcontains additional DNA, including an extra IR and two differentT3SEs (xopU and xopX).

In general, all these MICs carry known T3SS effector genes,such as xopJ2, xopJ5, xopT, xopU and xopX, as well as mlt and genesfor pectin lyase, protein tyrosine phosphatase, a resistance-nodulation-cell division (RND) efflux pump, and members of atwo-component system. Type IA DNA topoisomerase, potentialtyrosine recombinase, toxin/antitoxin, and acriflavin resistancegenes are also represented, as well as many unrelated hypotheticalprotein genes (see Table S3 in the supplemental material). More-over, MIC H from X. oryzae KACC 10331, carrying a polymerase/histidinol phosphatase-like protein, generated 5 bp target site du-plication, clearly indicating that it had inserted into the genome bytransposition.

Transposase sources for MIC mobility. Clearly, the presenceof flanking target DR observed for a number of MIC structures inthese Xanthomonas strains indicates that they have undergonetrue transposition events (e.g., those marked with asterisks inFig. 3B, C, and G and in Table S3 in the supplemental material).This requires an external source of the cognate transposase. ManyTALEs of the X. citri/axonopodis/campestris/campestris pv. vesica-toria group are located on plasmids. In these cases, transpositionfunctions could be supplied either from the plasmid TnXax1 de-rivatives or, when present, from a chromosomal TnXax1 deriva-tive. On the other hand, Asian X. oryzae and X. oryzae pv. oryzicolagroup strains, whose TALEs are uniquely chromosomal, do notcarry plasmid or chromosomal TnXax1 derivatives or indeed iso-lated transposition functions. Thus, further MIC transposition isunlikely. To explain the appearance of TALEs in Asian X. oryzaeand X. oryzae pv. oryzicola group strains, it seems probable thatthey were acquired by transmission from an ancestral plasmidwhich has subsequently been lost. The loss of this type of plasmidhas been observed in some North American X. oryzae strains(which are more closely related to X. oryzae pv. oryzicola [35]).Most contain a plasmid similar to that of X. campestris pv. vesica-toria pXCV38, with a Tn3-like element (ISXc4) whose transposaseis 40% identical to TnpA of TnXax1, and others appear to havelost the plasmid during laboratory culture (35).

Moreover, a second related plasmid, pXCV183 from X. camp-

estris pv. vesicatoria, carries a potential transposon, TnXca1 (seeTable S2 in the supplemental material), which shares the same IRtips and carries TnpA, TnpS, and TnpT with 59, 98, and 91%identity, respectively, to those from TnXax1 (Table S4). Althoughthe Asian X. oryzae group strains do not carry TnXax1 relatives, X.oryzae PXO carries structures similar to TnXca1, which carriesonly 29 instead of 92 bp TnXax1 IRs. The X. oryzae PXO TnXca1derivative carries tnpA and tnpS genes with 59 and 80% identity,respectively, to those in TnXax1 (Table S4). An additional andpartial copy without IRs is also found in X. citri subsp. citri strainB100 (Table S4). Since the TnXca1 IRs and transposition-relatedgenes share considerable similarity to those of TnXax1, TnXca1may be able to provide a functional transposase for the MICsfound in X. oryzae PXO. It is noteworthy that a partial TnXca1-related structure is also found in the genome of Pseudoxanthomo-nas spadix strain BD-a59, thus indicating that this class of trans-poson is not exclusive to the Xanthomonas genus (Table S4).

DISCUSSION

We have identified a novel class of Tn3 family-derived transpos-able elements in Xanthomonas species which could potentiallymediate combinatorial spread and diversification of plant viru-lence factors. The group includes TnXax1, an apparently intacttransposon from Xanthomonas citri, and we have also identified anumber of related structures in a variety of xanthomonads. Theseexhibit several different genetic organizations and carry differenttypes of passenger genes, mostly T3SEs. Although the right sides ofthese structures fall into relatively homogenous groups, the leftsides are highly variable: each derivative appears to have a uniqueleft side, indicating the occurrence of different recombination andduplication events. The IRs of these structures are commonly as-sociated with passenger genes, mostly TALE genes, forming MICelements. Target site duplications of 5 bp, commonly generated byelements from the Tn3 family, were also identified in some of theTnXax1-related structures and MICs, strongly supporting the ideathat these are products of true transposition events.

Studies with Bacillus cereus strains showed that their MICs(which carry genes encoding products such as endopeptidase, an-tibiotic resistance proteins, or regulatory factors) can be mobi-lized in trans by a cognate transposase (47). Different MIC struc-tures carrying TALE genes are often found in Xanthomonasstrains, and their role in the emergence of pathogenicity may beconsiderable.

Relationship of MIC clusters to the emergence of Xanthomo-nas pathogenicity. The transposition of TALE-carrying MICs inXanthomonas strains presents an intriguing puzzle, since they ap-pear to be grouped in clusters. This may indicate that the MICs ineach cluster were amplified and diverged from a single insertionevent. Moreover, in several citrus pathogens and related species,the MIC-associated TALE genes are often found in plasmids,whereas in the X. oryzae and X. oryzae pv. oryzicola species, theseoccur in the chromosome. In addition, the presence of plasmidsmay be essential to the development of disease caused by X. fuscanssubsp. aurantifolii strains. For instance, the plasmid pXcB from X.fuscans subsp. aurantifolii can be fully transferred in planta and iscentral to the development of citrus canker (37). Since some X.oryzae group members from North America carry plasmids simi-lar to pXCV38, it is possible that a plasmid-borne TnXax1 relativeand TALE-carrying MICs transiently invaded members of the X.oryzae group and subsequently spread by transposition driven by

Marini Ferreira et al.

8 ® mbio.asm.org January/February 2015 Volume 6 Issue 1 e02505-14

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

TnXax1 or by a cognate transposase from a closely related element(such as ISXc4 or TnXca1). This mechanism may have been re-sponsible for delivering the pthA4- and pthAw-carrying MICs toX. citri subsp. citri pathotype A and X. citri subsp. citri strain Aw

12879, respectively. These effectors are considered central for thedevelopment of the citrus canker (43). The X. oryzae pv. oryzicolastrain BLS256 does not have TALE-carrying MICs, and the major-ity of its TALEs group in a monophyletic clade. Moreover, thelines of evidence presented and the tree topology, shown in Fig. 3,strongly suggest that, after acquisition by transposition, the X.oryzae pv. oryzicola BSL256 TALEs evolved principally by dupli-

cation rather than by transposition. Although gene duplication isconsidered one of the major mechanisms through which new ge-netic material is generated, transposition can promote bacterialevolution in quantum leaps. Since X. oryzae pv. oryzicola BSL256is less widespread and causes less severe disease than members ofthe X. oryzae group (3), the absence of MIC-carrying TALEs ob-served in this strain may have affected its pathogenicity.

Generating T3SE and TALE diversity by transposition. In ad-dition to insertion and genome expansion, the Tn3 family repli-cative transposition mechanism could itself generate sequencevariation in several ways. Reassortment of passenger genes leading

FIG 5 A model for generating variability in the number of TALE gene repeats. The proposed steps of Tn3 family replicative transposition are shown. Top tobottom: double-strand transposon showing the terminal inverted repeats (IRs, red boxes) and three repeat sequences (green scale boxes); the transposon in situin the transposon donor molecule (flanking black bars), indicating the polarity of the two DNA strands and the polarity of cleavage (vertical blue arrows); transferof the 3= OH strand at each end of the transposon (thin arrows) at staggered positions (green square) into the target molecule (blue bars); formation of thebranched Shapiro intermediate with both donor (black bars) and target (blue bars) DNA attached; replication of the intermediate structure showing leadingstrand synthesis (bottom branch of the fork) and lagging strand synthesis (top branch of the fork); replication fork slippage on the leading strand template (left)resulting in removal of repeat sequences or slippage on the nascent strand (right) resulting in expansion of repeat sequences, which are each resolved in a secondround of replication.

Transposition and Pathogenicity in Xanthomonas

January/February 2015 Volume 6 Issue 1 e02505-14 ® mbio.asm.org 9

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

to changes in the T3SE repertoire could occur by “resolution”(site-specific recombination) between alternative rst sites fromdifferent elements, as shown for Tn4651-related transposons (33).Replicative Tn3 transposition would itself provide an exquisitemechanism for generating TALE diversity. TALE genes include anumber of conserved nearly identical tandem repeat sequencesthat encode the 34-amino-acid repeats, including the RVD site.Changes in the repeat numbers, RVDs, and sequences in the TALEgenes might occur by replication slippage between direct repeatsand errors during the replicative stage of transposition (Fig. 5) orby unequal crossing-over, as has been proposed for amplificationof other repeated DNA segments (48, 49). The example presentedin Fig. 5 shows that replication slippage on the template strand(bottom left) would give rise to a reduction in the number ofrepeat sequences, while slippage on the nascent strand (bottomright) would result in repeat expansion following a second roundof replication. Similar activities might occur on the lagging strandand lagging strand template. This mechanism may remodel thebacterial T3SE arsenal, producing novel effectors capable of bind-ing to new DNA motifs to regulate new sets of host genes. Alter-natively, such rearrangements may be generated similarly duringreplication of the host replicons or during conjugative transfer ofthe plasmid vectors. Experiments are under way to test these hy-potheses.

Tn3 family elements from closely related genera. The in-volvement of Tn3 family elements with virulence and pathogenic-ity is not an exclusive feature of the Xanthomonas genus. Addi-tional Tn3 family transposons identified in otherGammaproteobacteria, such as Pseudomonas aeruginosa (TnPa43)(50, 51) and Stenotrophomonas maltophilia (TnStma1) (52, 53),share TnXax1 IR sequences (GAGGG tips). These are also in-volved in assisting the bacterial host-related lifestyle, since theyalso carry genes directly or indirectly involved in pathogenicityand virulence or tolerance of high levels of various toxic metals.Xanthomonas, Pseudomonas, and Stenotrophomonas species aregammaproteobacteria and share a close common ancestor. Lateralgene transfer mediated by these Tn3-like transposable elementsmay have had an important role in speciation, shaping the lifestyles of their bacterial hosts. The Xanthomonas and Pseudomonasgenera are composed of several species capable of infecting differ-ent plant hosts. Even though more than 200 plant species are hostsfor Xanthomonadales and Pseudomonadales species (54), mostXanthomonas and Pseudomonas species exhibit a strict host range,resulting in specific combinations of plant-pathogen interactionsthat result in pathogen infection (or disease). The mobilization ofTnXax1 and its relatives and the potential recombination andshuffling of the T3SEs and TALEs in MIC structures are probablymajor sources of pathogenicity variation in these plant pathogens.However, we would like to point out that, in this study, we haveonly addressed transposons of a subclass of the Tn3 family, thosewith IRs whose sequences end in GAGGG. It is possible that someof the many other transposable elements present in some of thesexanthomonad genomes are also involved in transmitting patho-genic determinants. Work is under way to determine this.

MATERIALS AND METHODSBioinformatics analysis. Insertion sequences were annotated using theISfinder database (55) and ISsaga tool (56). Global comparisons usingsequenced Xanthomonadales genomes deposited in the public GenBankrepository were made using BLAST (57) with standard parameters, except

that the low-complexity-region filter was deactivated and a word size of 7was used.

The molecular phylogenetic analysis of the TALE genes was performedusing the maximum-likelihood method based on the general time-reversible model (58). The full-length nucleotide sequences of the TALESgenes were aligned and edited to maintain the order of the conservedtandem repeats, with no gap inclusion in order to respect the targetednucleotide sequence in the plant and, therefore, the TALE specificity. Thebootstrap consensus tree was inferred with 500 replicates (59). Branchescorresponding to partitions reproduced in less than 50% of bootstrapreplicates were collapsed. Initial trees for the heuristic search were ob-tained automatically by applying the Neighbor Joining and BioNJ algo-rithms to a matrix of pairwise distances estimated using the maximumcomposite likelihood (MCL) approach and then selecting the topologywith superior log likelihood value. A discrete gamma distribution wasused to model evolutionary rate differences among sites (3 categories[�G, parameter � 1.0586]). The analysis involved 122 TALE sequences,including a total of 8,174 positions in the final data set. Evolutionaryanalyses were conducted in MEGA6 (60).

SUPPLEMENTAL MATERIALSupplemental material for this article may be found at http://mbio.asm.org/lookup/suppl/doi:10.1128/mBio.02505-14/-/DCSupplemental.

Figure S1, PDF file, 0.8 MB.Table S1, PDF file, 0.1 MB.Table S2, PDF file, 0.1 MB.Table S3, PDF file, 0.1 MB.Table S4, PDF file, 0.1 MB.

ACKNOWLEDGMENTS

R.M.F. was supported by a Ph.D. fellowship from CAPES (Coordenaçãode Aperfeiçoamento do Pessoal de Ensino Superior), A.C.P.D.O. was re-cipient of an Undergraduate fellowship from UNESP—Universidade Es-tadual Paulista, and M.I.T.F. and J.A.F. are recipients of ProductivityGrants from CNPq (Conselho Nacional de Desenvolvimento Científico eTecnológico).

We thank Dave Lane (LMGM/CNRS) and all the members of theLaboratory of Molecular Biology of the Department of Technology(UNESP-Universidade Estadual Paulista).

We would also like to thank Mathieu Arlat, Bernard Hallet and Mar-shal Stark for comments which greatly improved the manuscript.

REFERENCES1. Mansfield J, Genin S, Magori S, Citovsky V, Sriariyanum M, Ronald P,

Dow, M, Ronald P, Dow M, Verdier V, Beer SV, Machado MA, Toth I,Salmond G, Foster GD. 2012. Top 10 plant pathogenic bacteria in mo-lecular plant pathology. Mol Plant Pathol 13:614 – 629. http://dx.doi.org/10.1111/j.1364-3703.2012.00804.x.

2. Jalan N, Yan Q, Kogenaru S, Guo Y, Jones JB, Graham JH, Wang N.2014. Genomics of Xanthomonas citri and related species, p 151614 –176.In Gross DC, Lichens-Park A, Kole C (ed), Genomics of plant-associatedbacteria. Springer, Berlin, Germany.

3. Triplett L, Koebnik R, Verdier V, Leach JE. 2014. The genomics ofXanthomonas oryzae, p 127614 –150. In Gross DC, Lichens-Park A, KoleC (ed), Genomics of plant-associated bacteria. Springer, Berlin, Germany.

4. Pothier JF, Vorhölter F-J, Blom J, Goesmann A, Pühler A, Smits THM,Duffy B. 2011. The ubiquitous plasmid pXap41 in the invasive phyto-pathogen Xanthomonas arboricola pv. pruni: complete sequence andcomparative genomic analysis. FEMS Microbiol Lett 323:52– 60. http://dx.doi.org/10.1111/j.1574-6968.2011.02352.x.

5. Darrasse A, Carrère S, Barbe V, Boureau T, Arrieta-Ortiz ML, BonneauS, Briand M, Brin C, Cociancich S, Durand K, Fouteau S, Gagnevin L,Guérin F, Guy E, Indiana A, Koebnik R, Lauber E, Munoz A, Noël LD,Pieretti I, Poussier S, Pruvost O, Robène-Soustrade I, Rott P, Royer M,Serres-Giardi L, Szurek B, van Sluys M-A, Verdier V, Vernière V, ArlatM, Manceau C, Jacques MA 2013. Genome sequence of Xanthomonasfuscans subsp. fuscans strain 4834-R reveals that flagellar motility is not a

Marini Ferreira et al.

10 ® mbio.asm.org January/February 2015 Volume 6 Issue 1 e02505-14

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

general feature of xanthomonads. BMC Genomics 14:761. http://dx.doi.org/10.1186/1471-2164-14-761.

6. Da Silva ACR, Ferro JA, Reinach FC, Farah CS, Furlan LR, Quaggio RB,Monteiro-Vitorello CB, Van Sluys MA, Almeida NF, Alves LMC, DoAmaral AM, Bertolini MC, Camargo LEA, Camarotte G, Cannavan F,Cardozo J, Chambergo F, Ciapina LP, Cicarelli RMB, Coutinho LL,Cursino-Santos JR, El-Dorry H, Faria JB, Ferreira AJS, Ferreira RCC,Ferro MIT, Formighieri EF, Franco MC, Greggio CC, Gruber A, Kat-suyama AM, Kishi LT, Leite RP, Lemos EGM, Lemos MVF, Locali EC,Machado MA, Madeira AMBN, Martinez-Rossi NM, Martins EC,Meidanis J, Menck CFM, Miyaki CY, Moon DH, Moreira LM, NovoMTM, Okura VK, Oliveira MC, Oliveira VR, Pereira HA, et al. 2002.Comparison of the genomes of two Xanthomonas pathogens with differ-ing host specificities. Nature 417:459 – 463. http://dx.doi.org/10.1038/417459a.

7. Moreira LM, Almeida NF, Jr, Potnis N, Digiampietri LA, Adi SS,Bortolossi JC, da Silva AC, da Silva AM, de Moraes FE, de Oliveira JC,de Souza RF, Facincani AP, Ferraz AL, Ferro MI, Furlan LR, GimenezDF, Jones JB, Kitajima EW, Laia ML, Leite RP, Jr, Nishiyama MY,Rodrigues Neto J, Nociti LA, Norman DJ, Ostroski EH, Pereira HA, Jr,Staskawicz BJ, Tezza RI, Ferro JA, Vinatzer BA, Setubal JC. 2010. Novelinsights into the genomic basis of citrus canker based on the genomesequences of two strains of Xanthomonas fuscans subsp. aurantifolii.BMC Genomics 11:238. http://dx.doi.org/10.1186/1471-2164-11-238.

8. Ryan RP, Vorhölter F-J, Potnis N, Jones JB, Van Sluys M-A, BogdanoveAJ, DOW JM. 2011. Pathogenomics of Xanthomonas: understandingbacterium-plant interactions. Nat Rev Microbiol 9:344 –355. http://dx.doi.org/10.1038/nrmicro2558.

9. Galán JE, Collmer A. 1999. Type III secretion machines: bacterial devicesfor protein delivery into host cells. Science 284:1322–1328. http://dx.doi.org/10.1126/science.284.5418.1322.

10. Cornelis GR, Van Gijsegem F. 2000. Assembly and function of type IIIsecretory systems. Annu Rev Microbiol 54:735–774. http://dx.doi.org/10.1146/annurev.micro.54.1.735.

11. Rossier O, Wengelnik K, Hahn K, Bonas U. 1999. The XanthomonasHrp type III system secretes proteins from plant and mammalian bacterialpathogens. Proc Natl Acad Sci U S A 96:9368 –9373. http://dx.doi.org/10.1073/pnas.96.16.9368.

12. Büttner D, He SY. 2009. Type III protein secretion in plant pathogenicBacteria. Plant Physiol 150:1656 –1664. http://dx.doi.org/10.1104/pp.109.139089.

13. Kay S, Bonas U. 2009. How Xanthomonas type III effectors manipulatethe host plant. Curr Opin Microbiol 12:37– 43. http://dx.doi.org/10.1016/j.mib.2008.12.006.

14. Bogdanove AJ, Schornack S, Lahaye T. 2010. TAL effectors: finding plantgenes for disease and defense. Curr Opin Plant Biol 13:394 – 401. http://dx.doi.org/10.1016/j.pbi.2010.04.010.

15. Lima WC, Paquola ACM, Varani AM, Van Sluys M-A, Menck CFM.2008. Laterally transferred genomic islands in Xanthomonadales relatedto pathogenicity and primary metabolism. FEMS Microbiol Lett 281:87–97. http://dx.doi.org/10.1111/j.1574-6968.2008.01083.x.

16. Monteiro-Vitorello CB, de Oliveira MC, Zerillo MM, Varani AM,Civerolo E, Van Sluys MA. 2005. Xylella and Xanthomonas Mobil’omics.Omics 9:146 –159. http://dx.doi.org/10.1089/omi.2005.9.146.

17. Varani AM, Monteiro-Vitorello CB, Nakaya HI, Van Sluys M-A. 2013.The role of prophage in plant pathogenic bacteria. Annu Rev Phytopathol51:429 – 451. http://dx.doi.org/10.1146/annurev-phyto-081211-173010.

18. Boch J, Bonas U. 2010. Xanthomonas AvrBs3 family-type III effectors:discovery and function. Annu Rev Phytopathol 48:419 – 436. http://dx.doi.org/10.1146/annurev-phyto-080508-081936.

19. Siguier P, Varani A, Perochon J, Chandler M. 2012. Exploring bacterialinsertion sequences with ISfinder: objectives, uses, and future develop-ments. Methods Mol Biol 859:91–103. http://dx.doi.org/10.1007/978-1-61779-603-6_5.

20. Grau J, Wolf A, Reschke M, Bonas U, Posch S, Boch J. 2013. Compu-tational predictions provide insights into the biology of TAL effector tar-get sites. PLoS Comput Biol 9:e1002962. http://dx.doi.org/10.1371/journal.pcbi.1002962.

21. Moscou MJ, Bogdanove AJ. 2009. A simple cipher governs DNA recog-nition by TAL effectors. Science 326:1501. http://dx.doi.org/10.1126/science.1178817.

22. Shiotani H, Fujikawa T, Ishihara H, Tsuyumu S, Ozaki K. 2007. A pthAhomolog from Xanthomonas axonopodis pv. citri responsible for host-

specific suppression of virulence. J Bacteriol 189:3271–3279. http://dx.doi.org/10.1128/JB.01790-06.

23. Streubel J, Blücher C, Landgraf A, Boch J. 2012. TAL effector RVDspecificities and efficiencies. Nat Biotechnol 30:593–595. http://dx.doi.org/10.1038/nbt.2304.

24. Boch J, Scholze H, Schornack S, Landgraf A, Hahn S, Kay S, Lahaye T,Nickstadt A, Bonas U. 2009. Breaking the code of DNA binding specific-ity of TAL-type III effectors. Science 326:1509 –1512. http://dx.doi.org/10.1126/science.1178811.

25. Moore R, Chandrahas A, Bleris L. 2014. Transcription activator-likeeffectors: a toolkit for synthetic biology. ACS Synth Biol 3:708 –716. http://dx.doi.org/10.1021/sb400137b.

26. Grindley NDF. 2002. The movement of Tn3-like elements: transpositionand cointegrate resolution, p 272708 –302. In Craig N, Craigie R, GellertM, Lambowitz A (ed), Mobile DNA II. ASM Press, Washington, DC.

27. Ohta S, Tsuchida K, Choi S, Sekine Y, Shiga Y, Ohtsubo E. 2002.Presence of a characteristic D-D-E motif in IS1 transposase. J Bacteriol184:6146 – 6154. http://dx.doi.org/10.1128/JB.184.22.6146-6154.2002.

28. Partridge SR, Hall RM. 2004. Complex multiple antibiotic and mercuryresistance region derived from the r-det of NR1 (R100). AntimicrobAgents Chemother 48:4250 – 4255. http://dx.doi.org/10.1128/AAC.48.11.4250-4255.2004.

29. Kholodii G, Yurieva O, Mindlin S, Gorlenko Z, Rybochkin V, NikiforovV. 2000. Tn5044, a novel Tn3 family transposon coding for temperature-sensitive mercury resistance. Res Microbiol 151:291–302. http://dx.doi.org/10.1016/S0923-2508(00)00149-2.

30. Chiou CS, Jones AL. 1993. Nucleotide sequence analysis of a transposon(Tn5393) carrying streptomycin resistance genes in Erwinia amylovoraand other gram-negative bacteria. J Bacteriol 175:732–740.

31. Voloudakis AE, Reignier TM, Cooksey DA. 2005. Regulation of resis-tance to copper in Xanthomonas axonopodis pv. vesicatoria. Appl Envi-ron Microbiol 71:782–789. http://dx.doi.org/10.1128/AEM.71.2.782-789.2005.

32. Vanhooff V, Galloy C, Agaisse H, Lereclus D, Révet B, Hallet B. 2006.Self-control in DNA site-specific recombination mediated by the tyrosinerecombinase TnpI. Mol Microbiol 60:617– 629. http://dx.doi.org/10.1111/j.1365-2958.2006.05127.x.

33. Yano H, Genka H, Ohtsubo Y, Nagata Y, Top EM, Tsuda M. 2013.Cointegrate-resolution of toluene-catabolic transposon Tn4651: determi-nation of crossover site and the segment required for full resolution activ-ity. Plasmid 69:24 –35. http://dx.doi.org/10.1016/j.plasmid.2012.07.004.

34. Di Gioia D, Peel M, Fava F, Wyndham RC. 1998. Structures of homol-ogous composite transposons carrying cbaABC genes from Europe andNorth America. Appl Environ Microbiol 64:1940 –1946.

35. Triplett LR, Hamilton JP, Buell CR, Tisserat NA, Verdier V, Zink F,Leach JE. 2011. Genomic analysis of Xanthomonas oryzae isolates fromrice grown in the United States reveals substantial divergence from knownX. oryzae pathovars. Appl Environ Microbiol 77:3930 –3937. http://dx.doi.org/10.1128/AEM.00028-11.

36. Garcillán-Barcia MP, Francia MV, de la Cruz F. 2009. The diversity ofconjugative relaxases and its application in plasmid classification. FEMSMicrobiol Rev 33:657– 687. http://dx.doi.org/10.1111/j.1574-6976.2009.00168.x.

37. El Yacoubi B, Brunings AM, Yuan Q, Shankar S, Gabriel DW. 2007. Inplanta horizontal transfer of a major pathogenicity effector gene. ApplEnviron Microbiol 73:1612–1621. http://dx.doi.org/10.1128/AEM.00261-06.

38. Noël L, Thieme F, Gäbler J, Büttner D, Bonas U. 2003. XopC and XopJ,two novel type III effector proteins from Xanthomonas campestris pv.vesicatoria. J Bacteriol 185:7092–7102. http://dx.doi.org/10.1128/JB.185.24.7092-7102.2003.

39. Jalan N, Kumar D, Andrade MO, Yu F, Jones JB, Graham JH, White FF,Setubal JC, Wang N. 2013. Comparative genomic and transcriptomeanalyses of pathotypes of Xanthomonas citri subsp. citri provide insightsinto mechanisms of bacterial virulence and host range. BMC Genomics14:551. http://dx.doi.org/10.1186/1471-2164-14-551.

40. Thieme F, Szczesny R, Urban A, Kirchner O, Hause G, Bonas U. 2007.New type III effectors from Xanthomonas campestris pv. vesicatoria trig-ger plant reactions dependent on a conserved N-myristoylation motif.Mol Plant Microbe Interact 20:1250 –1261. http://dx.doi.org/10.1094/MPMI-20-10-1250.

41. Laia ML, Moreira LM, Dezajacomo J, Brigati JB, Ferreira CB, Ferro MI,Silva AC, Ferro JA, Oliveira JC. 2009. New genes of Xanthomonas citri

Transposition and Pathogenicity in Xanthomonas

January/February 2015 Volume 6 Issue 1 e02505-14 ® mbio.asm.org 11

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

subsp. citri involved in pathogenesis and adaptation revealed by atransposon-based mutant library. BMC Microbiol 9:12. http://dx.doi.org/10.1186/1471-2180-9-12.

42. Oh H-S, Kvitko BH, Morello JE, Collmer A. 2007. Pseudomonas syrin-gae lytic transglycosylases coregulated with the type III secretion systemcontribute to the translocation of effector proteins into plant cells. J Bac-teriol 189:8277– 8289. http://dx.doi.org/10.1128/JB.00998-07.

43. Hu Y, Zhang J, Jia H, Sosso D, Li T, Frommer WB, Yang B, White FF,Wang N, Jones JB. 2014. Lateral organ boundaries 1 is a disease suscep-tibility gene for citrus bacterial canker disease. Proc Natl Acad Sci U. S. A.111:201313271. http://dx.doi.org/10.1073/pnas.1313271111.

44. Nicaise V, Joe A, Jeong BR, Korneli C, Boutrot F, Westedt I, Staiger D,Alfano JR, Zipfel C. 2013. Pseudomonas HopU1 modulates plant im-mune receptor levels by blocking the interaction of their mRNAs withGRP7. EMBO J 32:701–712. http://dx.doi.org/10.1038/emboj.2013.15.

45. Yu Y, Streubel J, Balzergue S, Champion A, Boch J, Koebnik R, Feng J,Verdier V, Szurek B. 2011. Colonization of rice leaf blades by an Africanstrain of Xanthomonas oryzae pv. oryzae depends on a new TAL effectorthat induces the rice nodulin-3 Os11N3 gene. Mol Plant Microbe Interact24:1102–1113. http://dx.doi.org/10.1094/MPMI-11-10-0254.

46. Pereira AL, Carazzolle MF, Abe VY, de Oliveira ML, Domingues MN,Silva JC, Cernadas RA, Benedetti CE. 2014. Identification of putativeTAL effector targets of the citrus canker pathogens shows functional con-vergence underlying disease development and defense response. BMCGenomics 15:157. http://dx.doi.org/10.1186/1471-2164-15-157.

47. Chen Y, Braathen P, Léonard C, Mahillon J. 1999. MIC231, a naturallyoccurring mobile insertion cassette from Bacillus cereus. Mol Microbiol32:657– 668. http://dx.doi.org/10.1046/j.1365-2958.1999.01388.x.

48. Meyer J, Iida S. 1979. Amplification of chloramphenicol resistance trans-posons carried by phage P1Cm in Escherichia coli. Mol Gen Genet 176:209 –219.

49. Chandler M, de la Tour EB, Willems D, Caro L. 1979. Some propertiesof the chloramphenicol resistance transposon Tn9. Mol Gen Genet 176:221–231.

50. Zaborin A, Gerdes S, Holbrook C, Liu DC, Zaborina OY, Alverdy JC.2012. Pseudomonas aeruginosa overrides the virulence inducing effect of

opioids when it senses an abundance of phosphate. PLoS One 7:e34883.http://dx.doi.org/10.1371/journal.pone.0034883.

51. Bains M, Fernández L, Hancock REW. 2012. Phosphate starvation pro-motes swarming motility and cytotoxicity of Pseudomonas aeruginosa.Appl Environ Microbiol 78:6762– 6768. http://dx.doi.org/10.1128/AEM.01015-12.

52. Pages D, Rose J, Conrod S, Cuine S, Carrier P, Heulin T, Achouak W.2008. Heavy metal tolerance in Stenotrophomonas maltophilia. PLoS One3:e1539. http://dx.doi.org/10.1371/journal.pone.0001539.

53. Ryan RP, Monchy S, Cardinale M, Taghavi S, Crossman L, Avison MB,Berg G, van der Lelie D, DOW JM. 2009. The versatility and adaptationof bacteria from the genus Stenotrophomonas. Nat Rev Microbiol7:514 –525. http://dx.doi.org/10.1038/nrmicro2163.

54. Kado CI. 2010. Plant bacteriology. American Phytopathological Society,St. Paul, MN.

55. Siguier P, Perochon J, Lestrade L, Mahillon J, Chandler M. 2006.ISfinder: the reference centre for bacterial insertion sequences. NucleicAcids Res 34:D32–D36. http://dx.doi.org/10.1093/nar/gkj014.

56. Varani AM, Siguier P, Gourbeyre E, Charneau V, Chandler M. 2011.ISsaga is an ensemble of web-based methods for high throughput identi-fication and semi-automatic annotation of insertion sequences in pro-karyotic genomes. Genome Biol 12:R30. http://dx.doi.org/10.1186/gb-2011-12-3-r30.

57. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W,Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation ofprotein database search programs. Nucleic Acids Res 25:3389 –3402.http://dx.doi.org/10.1093/nar/25.17.3389.

58. Kumar S, Tamura K, Nei M. 2004. MEGA3: integrated software forMolecular Evolutionary Genetics Analysis and sequence alignment. BriefBioinform 5:150 –163. http://dx.doi.org/10.1093/bib/5.2.150.

59. Felsenstein J. 1997. An alternating least squares approach to inferringphylogenies from pairwise distances. Syst Biol 46:101–111. http://dx.doi.org/10.1093/sysbio/46.1.101.

60. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6:Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30:2725–2729. http://dx.doi.org/10.1093/molbev/mst197.

Marini Ferreira et al.

12 ® mbio.asm.org January/February 2015 Volume 6 Issue 1 e02505-14

m

bio.asm.org

on February 17, 2015 - P

ublished by m

bio.asm.org

Dow

nloaded from

Table S4. Organization of TnXax1 related elements

Main chromosome or Plasmid

Mlt TnpA TnpS res site [TnpS to TnpT] TnpT

Xoo KACC 10331 Not found Not found Not found Not found Not found

Xoo MAFF 311018 Not found Not found Not found Not found Not found

Xoo PXO99A Not found A) 2,302,697 .. 2,305,654 (59%)

[IS630 partial] – [TnpS]

B) 1,625,991 .. 1,626,503 (92%) – Partial

[VirG] – [IS256 partial]

A) 2,301,719 .. 2,302,666 (80%)

[arsenical membrane pump] –

[TnpA]

A) 2,294,680 .. 2,294,784 (79%)

[ISXoo5] – [TnpT]

A) 2,293,677..2,294,663 (82%)

[ISXoo5] – [TnpT]

Xoo BLS256 A)4,233,640.. 4,234,908 (95%)

[C.H] – [IS3 ssgr IS407 partial]

Not found Not found Not found Not found

Xff 4834-R plc A)29,119 .. 30,390 (98%)

[XopE3] - [C.H]

Not found Not found Not found A)4,947,724 .. 4,948,167 (84%) - Partial

X. albilineans str. GPE PC73 A) 2,246,792 .. 2,248,069 (87%)

[lysineaminomutase] – [trna/rrnamethyltransferase]

Not found Not found Not found Not found

Xcv pXCV183 Notfound A) 105,147 .. 108,086 (59%)

[tnpS] – [repA]

B) 119,656 .. 122,448 (41%)

[C.H] – [C.H]

C) 124,208 .. 127,144 (59%)

[TnpS] – [hypothetical]

A) 104,170 .. 105,120 (97%)

[tnpT] – [tnpA]

B) 123,255 .. 124,181 (98%)

[C.H] – [TnpA]

A) 103,979 .. 104,170 (90%)

[TnpT] – [ TnpS]

A) 102,969 .. 103,979 (91%)

[Type III] – [TnpS]

B) 101,026 .. 101532 (94%)

Partial

Xcc Aw12879 pXcaw58 Not found A) 6,161 .. 9,097 (59%)

[TnpS] –[Tn3 transposase - partial]

B) 18,384 .. 21,176 (41%)

[hypothetical] – [hypothetical]

C) 54,640 .. 57,432 (41%)

[hypothetical] – [hypothetical]

A) 3,983 .. 6,134 (95%) - frameshift

[ Type III] – [TnpA]

A) 3,792 .. 3,983 (85%)

[TnpT] - [TnpS]

A) 3,218 .. 3,790 (94%)

[hypothetical] – [TnpS]

B) 2,739 .. 3,218 (95%)

Partial

Xcc B100 Not found A) 2,759,319 .. 2,762,255 (59%)

[Peptidase] – [TnpS]

B) 2,787,410 .. 2,788,048 (87%) – Partial

[IS3 ssgr IS407 partial ] – [type III]

A)2,762,282.. 2,763,226 (90%)

[TnpA] – [TnpT]

A) 2,763,259 .. 2,763,416 (79%)

[TnpS] – [TnpT]

A) 2,763,416 .. 2764465 (93%)

[TnpS] – [hypothetical ]

Pseudoxanthomonasspadix BD-a59 Not found A) 636,723 .. 633,745 (85%)

[alcohol dehydrogenases] – [TnpS]

A) 636,754 .. 638,019 (71%)

[ TnpA] – [transcription factor ]

B) 651,884 .. 653,152 (71%)

[H] – [transcription factor]

A) 666,719 ..666,827 (76%)

[ATPase domain] – [TnpT]

A) 666,842 .. 667,873 (86%)

[endonuclease] – [sec-C metal-binding protein]

Abbreviations: H= hypothetical protein; C.H= conserved hypothetical protein

The gene products inside the brackets indicate the genome context of each TnXax1 genes

Table S2. Transposons related to TnXax1

A) TnPa43 :19,903 bp – Source: ISfinder Locus Tag Element ID Coordinates Length Orientation Predicted Product

n/d ISPa43_00001 43..1,824 1,782 - Methyltransferase domain proteinn/d ISPa43_00002 2,224..2,451 228 - hypothetical proteinn/d ISPa43_00003 2,769..3,629 861 + ectoine hydroxylasen/d ISPa43_00004 3,654..4,550 897 + Phosphate-import protein PhnD precursorn/d ISPa43_00005 4,674..5,522 849 + Phosphate-import permease protein PhnEn/d ISPa43_00006 5,556..6,368 813 + Phosphate-import ATP-binding protein PhnCn/d ISPa43_00007 6,380..6,925 546 + phosphonate ABC transporter, permease protein PhnEn/d ISPa43_00008 7,011..7,412 402 + Phosphate-import permease protein PhnEn/d ISPa43_00009 7,439..7,918 480 + phosphonate C-P lyase system protein PhnGn/d ISPa43_00010 7,911..8,495 585 + Alpha-D-ribose 1-methylphosphonate 5-triphosphate synthase subunit PhnHn/d ISPa43_00011 8,495..9,616 1,122 + Alpha-D-ribose 1-methylphosphonate 5-triphosphate synthase subunit PhnIn/d ISPa43_00012 9,613..10,518 906 + Alpha-D-ribose 1-methylphosphonate 5-phosphate C-P lyasen/d ISPa43_00013 10,744..11,316 573 + Putative phosphonates utilization ATP-binding protein PhnKn/d ISPa43_00014 11,330..12,055 726 + Alpha-D-ribose 1-methylphosphonate 5-triphosphate synthase subunit PhnLn/d ISPa43_00015 12,052..13,206 1,155 + Alpha-D-ribose 1-methylphosphonate 5-triphosphate diphosphatasen/d ISPa43_00016 13,220..13,789 570 + Ribose 1,5-bisphosphate phosphokinase PhnNn/d ISPa43_00017 13,780..14,547 768 + Phosphoribosyl 1,2-cyclic phosphodiesterasen/d ISPa43_00018 14,544..15,353 810 + phosphonate utilization associated putative membrane proteinn/d ISPa43_00019 15,471..16,805 1,335 + site-specific tyrosine recombinase XerD (TnpI)n/d ISPa43_00020 16,862..19,849 2,988 + Transposase, TnpA family (DDE domain)

B) TnStma1: 28,714 bp from Stenotrophomonas maltophilia D457 - NC_017671 Locus Tag Element ID Coordinates Length Orientation Predicted ProductSMD_2129 ISStma18_00001 84..1,142 1,059 - chromosome segregation protein SMCSMD_2130 ISStma18_00002 1,465..1,857 393 + hypothetical proteinSMD_2131 ISStma18_00003 1,929..3,233 1,305 + type I secretion outer membrane protein, TolC familySMD_2132 ISStma18_00004 3,230..4,726 1,497 + Cation efflux system protein CusB precursorSMD_2133 ISStma18_00005 4,723..7,884 3,162 + Cation efflux system protein CusASMD_2134 ISStma18_00006 7,930..8,286 357 + hypothetical proteinSMD_2135 ISStma18_00007 8,379..8,780 402 + Cation efflux system protein CusF precursorSMD_2136 ISStma18_00008 9,152..9,661 510 - Lipoprotein signal peptidaseSMD_2137 ISStma18_00009 9,654..10,901 1,248 + Sodium/calcium exchanger proteinSMD_2138 ISStma18_00010 10,924..11,244 321 - Nitrogen regulatory protein P-IISMD_2139 ISStma18_00011 11,244..14,369 3,126 - Cation efflux system protein CzcASMD_2140 ISStma18_00012 14,407..15,597 1,191 - Nickel and cobalt resistance protein CnrBSMD_2141 ISStma18_00013 15,594..16,871 1,278 - Cation efflux system protein CzcCn/d ISStma18_00014 17,170..17,337 168 + hypothetical proteinSMD_2142 ISStma18_00015 17,575..19,971 2,397 - putative cadmium-transporting ATPaseSMD_2143 ISStma18_00016 19,968..21,938 1,971 - Ferrous iron uptake proteinSMD_2144 ISStma18_00017 22,127..22,567 441 + Copper export regulatorSMD_2145 ISStma18_00018 22,605..24,251 1,647 + Glutamine-dependent NAD(+) synthetaseSMD_2146 ISStma18_00019 24,226..25,527 1,302 + site-specific tyrosine recombinase XerD (TnpI)SMD_2147 ISStma18_00020 25,566..28,550 2,985 + Transposase, TnpA family (DDE domain)

C) TnXca1: 11,523 bp from X. campestris pv. vesicatoria str. 85-10 plasmid pXCV183 - NC_007507Locus Tag Element ID Coordinates Length Orientation Predicted ProductXCVd0093 ISXca_00001 352..1,389 1,038 + hypothetical proteinXCVd0094 ISXca_00002 1,377..2,444 1,068 - chromosome segregation protein SMCXCVd0095 ISXca_00003 2,635..2,838 204 + hypothetical proteinXCVd0096 ISXca_00004 2,890..3,603 714 + Site-specific recombinase XerDXCVd0097 ISXca_00005 3,612..6,575 2,964 + Transposase, TnpA familyXCVd0098 ISXca_00006 6,906..8,132 1,227 - Plasmid encoded RepA proteinXCVd0099 ISXca_00007 8,412..9,599 1,188 - UDP-N-acetylglucosamine kinaseXCVd0100 ISXca_00008 9,606..9,869 264 - hypothetical proteinXCVd0101 ISXca_00009 9,894..10,295 402 - tRNA(fMet)-specific endonuclease VapCXCVd0102 ISXca_00010 10,292..10,552 261 - Virulence-associated proteinXCVd0103 ISXca_00011 10,742..11,308 567 + DNA-invertase hin

D) TnTin1: 6,123 bp from Thiomonas intermedia K12 - CP002021 - Only Transposition related genesLocus Tag Element ID Coordinates Length Orientation Predicted ProductTint_1402 ISTin1_00001 49..2,268 2,219 - Transposase, TnpA family Tint_1403 ISTin1_00002 2,265..3,083 818 - recombinaseTint_1404 ISTin1_00003 3,076..4,596 1,520 - IntegraseTint_1405 ISTin1_00004 4,981..6,081 1,100 + chromosome segregation protein SMC

Table S1. Genomic positions of the TnXax1 IRs

Genome or Plasmid identified IR start IR stop Conserved nucleotidesXanthomonas axonopodis Xac29-1 pXAC64 (64,118 bp) 3989 4080 92/92

12932 12842 74/9213749 13839 78/9117700 17609 75/9425197 25108 80/9052639 52729 74/9253945 54035 78/9158406 58315 75/94

Xanthomonas axonopodis Xac29-1 (5,153,455 bp) 3802552 3802643 92/923811473 3811382 72/943814001 3813911 70/914151804 4151713 76/92

Xanthomonas axonopodis pv. citrumelo F1 (4,967,469 bp) 2452240 2452149 92/922558748 2558839 92/922438697 2438771 66/76

Xanthomonas campestris pv. vesicatoria 85-10 (5,178,466 bp) 662686 662630 51/582480425 2480345 66/852603770 2603802 30/332610728 2610693 32/362614800 2614710 75/922764198 2764272 66/762776534 2776443 92/924369321 4369401 67/83

Xanthomonas axonopodis pv. citri str. 306 pXAC64 (64,920 bp) 3350 3441 92/9212294 12204 74/9214311 14401 78/9117959 17868 75/94

25458 25369 80/9053668 53758 74/9254974 55064 78/9158823 58732 75/94

Acidovorax citrulli AAC00-1 (5,352,772 bp) 2975223 2975303 66/852978700 2978785 69/862978963 2978872 90/923224781 3224872 90/923225044 3224959 69/863228521 3228441 66/85

Xanthomonas fuscans subsp. fuscans str. 4834-R, pla (45,224 bp) 93 1 75/94

8706 8797 77/928796 8771 24/26

14156 14247 76/9214246 14221 24/2617444 17353 76/9217354 17379 24/2640015 40105 83/91

Xanthomonas citri pv. aurantifolii strain C340 PthC (4,876bp) 455 545 82/914325 4278 44/494519 4453 58/68

Xanthomonas citri plasmid pXcB (37,106 bp) 31463 31553 82/9135333 35286 44/4935527 35461 58/68

Xanthomonas fuscans subsp. fuscans str. 4834-R, plc (41,950 bp) 19199 19108 75/92

21844 21934 75/9122338 22247 73/92

26372 26462 75/9130746 30837 73/9231908 31820 76/9038186 38276 81/9141950 41904 42/47

Xanthomonas axonopodis pv. manihotis strain CFBP1851 pthBXam (10,736 bp) 3263 3353 81/91

Xanthomonas axonopodis Xac29-1 pXAC33 (31,801 bp) 9094 9004 78/9117429 17518 80/9024927 25018 75/9428584 28494 78/91

Xanthomonas axonopodis pv. citri str. 306 pXAC33 (33,700 bp) 5469 5558 80/9012968 13059 75/9416706 16616 78/9130836 30746 78/91

Xanthomonas citri subsp. citri Aw12879 pXcaw58 (58,317 bp) 12492 12582 78/9122394 22485 74/9426247 26156 78/9227553 27463 74/92

Xanthomonas citri pv. citri strain X0053 PthAW (4,559 bp) 460 550 78/914312 4221 74/94

Xanthomonas smithii subsp. citri pB3.7 (4,500 bp) 133 223 78/914378 4287 75/94

Xanthomonas smithii subsp. citri pthA-KC21 (4,070 bp) 109 199 78/913959 3867 75/95

Xanthomonas smithii subsp. citri pB3.1 (3,985 bp) 119 209 78/91

3767 3676 75/94

Xanthomonas citri hssB3.0 gene for avirulence protein (3,718 bp) 101 191 78/91

Xanthomonas axonopodis pv. glycines strain AG1 plasmid pAG1 (15,143 bp) 3912 3821 75/94

4663 4573 74/9214711 14802 78/9214897 14987 78/91

Xanthomonas citri strain 3213 PthA (4,275 bp) 349 439 78/914198 4107 75/94

Xanthomonas campestris pv. armoraciae Hax (3,507 bp) 1 91 78/91

Xanthomonas campestris pv. armoraciae Hax3 (3,218 bp) 1 91 78/913218 3150 56/71

Xanthomonas citri apl3 (4,639 bp) 101 191 78/914562 4471 75/94

Xanthomonas citri apl2 (3,823 bp) 101 191 78/91

Xanthomonas citri apl1 (4,027 bp) 101 191 78/913950 3859 75/94

Xanthomonas vesicatoria avrBs4 (4,000 bp) 72 162 78/913912 3821 75/94

Xanthomonas vesicatoria plasmid pXV11 avrBs3 (4,366 bp) 374 464 78/914226 4135 75/94

Xanthomonas campestris pv. campestris str. ATCC 33913 1902124 1902190 57/68

(5,076,188 bp)2466511 2466591 67/822482712 2482622 72/912482622 2482650 26/292491914 2492005 76/922492195 2492272 60/843434576 3434666 74/914438478 4438566 75/91

Xanthomonas campestris pv. campestris str. B100 (5,079,002 bp) 1453050 1452960 74/91

2409940 2409863 60/842410221 2410130 76/922791168 2791078 73/912791078 2791106 26/29

Xanthomonas campestris pv. campestris str. 8004 (5,148,708 bp) 1476237 1476147 74/91

2419112 2419035 60/842419393 2419302 76/922496533 2496623 72/912496623 2496595 26/292512734 2512654 67/823137704 3137638 57/684496061 4496149 75/91

Xanthomonas fuscans subsp. fuscans str. 4834-R (4,981,995 bp) 1884500 1884596 74/97

1886283 1886240 40/452750592 2750676 69/852751922 2751831 74/924939801 4939891 76/924946417 4946327 75/914946455 4946423 31/35

4953056 4953122 59/68

Xanthomonas citri subsp. citri Aw12879 (5,321,499 bp) 1262821 1262911 71/913952635 3952544 72/944729175 4729084 76/92

Xanthomonas axonopodis pv. citri str. 306 (5,175,554 bp) 3807325 3807235 70/914143510 4143419 76/92

Xanthomonas campestris PthN (3,994 bp) 38 128 76/92135 225 76/92

3579 3488 75/923489 3517 26/29

Xanthomonas oryzae pv. oryzae strain PXO99 type III effector AvrXa23 (6,452 bp) 100 9 72/92

754 844 72/915541 5450 73/926195 6285 74/91

Xanthomonas oryzae pv. oryzae PXO99 (5,240,075 bp) 321285 321375 73/92324403 324314 72/91559020 559111 73/92842065 842003 51/63

1625939 1625971 31/331633509 1633420 65/911645151 1645242 72/921649608 1649518 73/911650262 1650353 72/921781506 1781425 65/821838885 1838816 57/712359265 2359175 73/912359919 2360010 73/922363552 2363461 73/92

2387282 2387191 73/922387936 2388026 72/912392130 2392039 72/922527364 2527426 51/632681275 2681206 58/722682578 2682668 74/912686432 2686341 72/922687086 2687176 71/912739451 2739513 51/632893362 2893293 58/722894665 2894755 74/912898519 2898428 72/922899173 2899263 71/914104892 4104801 72/924105546 4105636 72/914110333 4110242 73/924110987 4111077 74/914114733 4114642 72/924115387 4115477 72/914118731 4118640 73/924119385 4119475 72/91

Xanthomonas oryzae pv. oryzae strain C8 Avr/pthC8b (4,525 bp) 3617 3526 72/92

4271 4361 74/91

Xanthomonas campestris pv. malvacearum Avrb6 (3,856 bp) 314 404 76/923763 3672 75/94

Xanthomonas oryzae pv. oryzae MAFF 311018 (4,940,217 bp) 1232038 1231948 74/911232692 1232783 73/921236438 1236348 74/911237092 1237183 72/921241876 1241786 72/91

1242530 1242621 72/922204911 2204821 74/912205565 2205656 73/922209623 2209533 74/912210920 2210989 59/722218495 2218433 51/632352097 2352188 72/922356288 2356197 73/922356941 2357032 73/922360576 2360486 73/912386917 2386826 73/922387571 2387661 72/913028589 3028658 57/713085100 3085181 65/823218732 3218641 73/923219386 3219476 74/913223441 3223350 73/923224095 3224185 74/913228882 3228791 73/923229536 3229626 73/913234212 3234121 71/923244797 3244886 65/913252366 3252334 31/334300513 4300575 51/634529444 4529353 73/924763458 4763547 72/914765690 4765600 72/92

Xanthomonas oryzae pv. oryzae AvrXa27 (22,214 bp) 3347 3256 72/924001 4091 72/918788 8697 73/929442 9532 74/91

13188 13097 72/9213842 13932 72/91

17186 17095 73/9217840 17930 72/91

Xanthomonas oryzae pv. oryzae KACC 10331 (4,941,439 bp) 1265943 1265853 72/911266597 1266688 73/922225929 2225839 74/912226583 2226674 71/922230641 2230551 74/912231938 2232007 59/722239521 2239482 36/402377246 2377337 72/922381013 2380923 73/912401087 2400996 73/922401741 2401831 72/912406648 2406604 40/452407301 2407392 73/923035786 3035855 57/713092324 3092405 65/823223775 3223684 72/923224429 3224519 72/913229927 3229836 72/923230581 3230671 73/913234546 3234455 71/923252722 3252690 31/334307290 4307352 51/634534573 4534482 73/924767482 4767569 72/914770598 4770508 72/92

Xanthomonas campestris pv. vesicatoria plasmid pXCV183 (182,572 bp) 101536 101571 31/36

112968 113058 73/91113058 113030 26/29

Xanthomonas arboricola pv. pruni str. CFBP 5530 plasmid pXap41 (41,102 bp) 1514 1598 70/86

11847 11812 31/3615966 16031 57/6925065 24975 74/92

Xanthomonas axonopodis pv. glycines strain 8ra plasmid pXAG81 (26,721 bp) 3912 3821 75/94

4663 4573 74/9222797 22887 74/92

Xanthomonas campestris pv. campestris beta-galactosidase Gal35I (3,423 bp) 2801 2892 74/92

Xanthomonas oryzae pv. oryzicola BLS256 (4,831,739 bp) 653841 653779 53/631558706 1558796 73/921686668 1686616 47/531697100 1697009 74/932548617 2548706 66/912556024 2555991 31/34

Xanthomonas oryzae pv. oryzicola Avr/Pth13 (3,170 bp) 2852 2761 74/93

Xanthomonas campestris pv. vesicatoria outer protein B (23,471 bp) 8848 8904 51/58

Xanthomonas campestris pv. vesicatoria avirulence protein AvrBsT (1,931 bp) 1395 1484 71/92

Xanthomonas oryzae avirulence protein AvrXa10 (3,720 bp) 4 92 70/91

Xanthomonas albilineans GPE PC73 (3,768,695 bp) 2246665 2246755 68/911840960 1840907 42/54

Xanthomonas campestris pv. vesicatoria outer protein J (1,441 61 141 66/85

bp)

Xanthomonas albilineans str. GPE PC73, plasmid (24,837 bp) 7897 7932 33/36

* conserved nucleotides: x/y where y is the full length of the IR compared to the canonical 92 bp of TnXax1 and x is the number of conserved nucleotides

XccA pXAC64

XccA(w) pXcaw58

Xff pla

Xff plc

Xc pXcB

XccA-29-1 pXAC64

XccA-29-1 pXAC33

XccA pXAC33

Xag AG1 pAG1

Xap pXap41

Xcv pXCV183

Xag 8ra pXAG81

Xa GPE PC73

Other genes (no TALEs) TnXca1

TnXax1 relative structure TALE gene (unit)

TnXax1 IRR

TnXax1 IRL

TnXca1    

XccA-29-1

Xac F1

Xcv 85-10

A. citrulli

Xcc ATCC33913

Xcc B100

Xcc 8004

XccA(w)

XccA

Xa GPE PC73

A   B  

A  

A  

TnXax1 IRL TnXax1 IRR

TnXax1 relative structure

Other genes (no TALEs) TALE gene (unit)

MICs

Xff A   B   C