Detection and Characterization of Miniature Inverted-RepeatTransposable Elements in “Candidatus Liberibacter asiaticus”

Xuefeng Wang,a,b Jin Tan,a,c Ziqin Bai,a,c* Huanan Su,a,c Xiaoling Deng,d Zhongan Li,a Changyong Zhou,a Jianchi Chenb

National Citrus Engineering Research Center, Citrus Research Institute, Southwest University, Chongqing, People’s Republic of Chinaa; San Joaquin Valley AgriculturalSciences Center, Agricultural Research Services, U.S. Department of Agriculture, Parlier, California, USAb; College of Plant Protection, Southwest University, Chongqing,People’s Republic of Chinac; Citrus Huanglongbing Research Center, South China Agricultural University, Guangzhou, People’s Republic of Chinad

Miniature inverted-repeat transposable elements (MITEs) are nonautonomous transposons (devoid of the transposasegene tps) that affect gene functions through insertion/deletion events. No transposon has yet been reported to occur in“Candidatus Liberibacter asiaticus,” an alphaproteobacterium associated with citrus Huanglongbing (HLB, yellow shootdisease). In this study, two MITEs, MCLas-A and MCLas-B, in “Ca. Liberibacter asiaticus” were detected, and the genomewas characterized using 326 isolates collected in China and Florida. MCLas-A had three variants, ranging from 237 to 325bp, and was inserted into a TTTAGG site of a prophage region. MCLas-A had a pair of 54-bp terminal inverted repeats (TIRs),which contained three tandem repeats of TGGTAACCAC. Both “filled” (with MITE) and “empty” (without MITE) states weredetected, suggesting the MITE mobility. The empty sites of all bacterial isolates had TIR tandem repeat remnants (TRR). Fre-quencies of TRR types varied according to geographical origins. MCLas-B had four variants, ranging from 238 to 250 bp, and wasinserted into a TA site of another “Ca. Liberibacter” prophage. The MITE, MCLas-B, had a pair of 23-bp TIRs containing no tan-dem repeats. No evidence of MCLas-B mobility was found. An identical open reading frame was found upstream of MCLas-A(229 bp) and MCLas-B (232 bp) and was predicted to be a putative tps, suggesting an in cis tps-MITE configuration. MCLas-Aand MCLas-B were predominantly copresent in Florida isolates, whereas MCLas-A alone or MCLas-B alone was found in Chi-nese isolates.

Miniature inverted repeat transposable elements (MITEs) area type of nonautonomous transposons that involve inser-

tion/deletion of DNA in the genomes of both prokaryotes andeukaryotes and influence gene functions (1, 2). MITEs are gener-ally small (�400 bp) and contain a noncoding short central region(CR) flanked by a pair of terminal inverted repeats (TIRs). MITEsequences can form highly stable secondary structures at the tran-scriptional level. Direct repeats (DRs) are typically found outsidethe TIRs on the host bacterial genomic DNA (1, 3). MITEs werefirst described in Neisseria sp. (4) and are now reported in manyother bacteria (5–14).

“Candidatus Liberibacter asiaticus” is a nonculturable, phlo-em-restricted alphaproteobacterium associated with citrus huan-glongbing (HLB; yellow shoot disease, also known as greeningdisease). HLB is one of the most destructive diseases in citrusproduction worldwide (15). In China, HLB was observed over 100years ago (16), and the association with “Ca. Liberibacter asiati-cus” was confirmed in 1996 (17, 18). The bacterium was found inSão Paulo, Brazil, in 2004 (19), and in Florida a year later (20).Following the reports of its occurrence in several southern states,“Ca. Liberibacter asiaticus” was detected in California in 2012(21). Due to the lack of in vitro culture, much of the “Ca. Liberib-acter asiaticus” biology, including the status of transposons, re-mains to be studied.

Analysis of the whole-genome sequence concluded that notransposon was identified in a Florida isolate of “Ca. Liberib-acter asiaticus” (22). No transposon was annotated in the com-plete sequences of three phages/prophages of “Ca. Liberibacterasiaticus,” SC1, SC2, and FP2 (23, 24). In sequence analyses,two characteristics are highly indicative of transposons: pres-ence of TIRs and identification of a transposase gene (tps).Since MITEs are devoid of tps, their detection is even more

difficult, if not impossible, should a single bacterial genomesequence be studied. Analysis of a single genome sequence can-not simultaneously reveal the presence (“filled”) and absence(“empty”) states of a MITE. Detection of the two states is adirect evidence of transposon mobility.

Tettelin et al. (25) described the concept of the pangenome,with emphasis on genomic sequences from multiple isolates toreveal a “core genome” and a “dispensable genome” of a bacterialspecies. Mobile elements, such as phages and transposons, con-tribute to the “dispensable genome” and may be present in thegenomes of some isolates and absent in others. Thus, analysesof a bacterial population may provide an opportunity for trans-poson detection. In the past few years, there have been severalreports of DNA polymorphisms in “Ca. Liberibacter asiaticus”populations using single primer sets (24, 26–29). By analyzingtwo Chinese “Ca. Liberibacter asiaticus” populations, one with120 isolates and the other with 39 isolates, Liu et al. (27) ob-served a significant difference in the frequency of a prophagegene that suggested the presence of an active phage. In thisstudy, we applied the concept of population or pangenomeanalysis to search for transposons in “Ca. Liberibacter asiati-cus.” A polymorphic genomic locus was identified and ana-

Received 11 April 2013 Accepted 25 June 2013

Published ahead of print 28 June 2013

Address correspondence to Jianchi Chen, jianchi.chen@ars.usda.gov, orChangyong Zhou, zhoucy@cric.cn.

* Present address: Ziqin Bai, Guizhou Fruit Institute, Guiyang, China.

Copyright © 2013, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JB.00413-13

September 2013 Volume 195 Number 17 Journal of Bacteriology p. 3979–3986 jb.asm.org 3979

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

lyzed using a total of 326 “Ca. Liberibacter asiaticus” isolatesfrom China and Florida. As a result, two MITEs, MCLas-A andMCLas-B, were discovered and are reported here.

MATERIALS AND METHODSSources of “Ca. Liberibacter asiaticus” DNA. Because “Ca. Liberibacterasiaticus” is not culturable in vitro, pure bacterial DNA could not be ob-tained. Instead, “Ca. Liberibacter asiaticus” DNA was extracted fromHLB-affected citrus trees along with host DNA. Leaves showing charac-teristic yellowing and/or mottling HLB symptoms (15, 16) were collectedfor DNA extraction. The presence of “Ca. Liberibacter asiaticus” was in-dicated by positive PCR detection with primer set OI1/OI2c (30). A bac-terial isolate was represented by DNA from a “Ca. Liberibacter asiaticus”-infected tree. DNA extracted from leaves of non-“Ca. Liberibacterasiaticus”-infected citrus trees in both California, where only a single caseof “Ca. Liberibacter asiaticus” was reported, and Chongqing, China,where “Ca. Liberibacter asiaticus” has not been reported, was used asnegative control. “Ca. Liberibacter asiaticus” samples were collected fromnine provinces in China and from Florida in the United States (Table 1).Samples in China were collected between December 2008 and October2011 and extracted using a modified cetyltrimethylammonium bromide(CTAB) method (31). DNA preparations from Florida were kindly pro-vided by X. Sun and M. Irey. Details of HLB sample collection and DNAextractions in Florida were described previously (26).

PCR primers and procedures. DNA sequences of “Ca. Liberibacterasiaticus” strain Psy62 (CP001677), phage SC1 (HQ377372), phage SC2(HQ377373), and phage FP2 (JF773396) were downloaded from theGenBank DNA database in the National Center for Biotechnology Infor-mation (NCBI; http://www.ncbi.nlm.nih.gov/). Primers were designedusing Primer 3 software (32). As an extension of a genomic diversityresearch (29), primer set LapPF1-f/LapPF1-r (5=-GCCACTTTGGGGTAGCAGTA-3=/5=-AAAACTTTCGTCACGGCTTT-3=) were found to yieldmultiple amplicons from some “Ca. Liberibacter asiaticus” isolates (Fig.1) and therefore selected for this investigation.

A previously published procedure was followed for PCR (29). Briefly,a reaction mixture (25 �l) included 2 �l of template DNA, 0.3 �l of TaqDNA polymerase at 5 U/�l (TaKaRa Bio Inc., Shiga, Japan), 0.4 �l of eachforward and reverse primer (10 �M), and 2.5 �l of 2.5 mM deoxynucleo-side triphosphate. Thermal cycling comprised an initial denaturing of94°C for 2 min, followed by 35 cycles of amplification (94°C for 30 s, 55°Cfor 30 s, and 72°C for 30 s) and a final extension for 7 min. PCR productswere electrophoresed in a 1.5% agarose gel and visualized by ethidiumbromide staining under UV light.

DNA sequencing and analysis. PCR amplicons from “Ca. Liberibac-ter asiaticus” isolates were purified from agarose gels using a QIAquick

Gel Extraction kit (Qiagen, Valencia, CA) and cloned with pGEM T-easyvector (Promega Corp., Fitchburg, WI). Four to seven clones were ran-domly selected from each isolate for sequencing using the BigDye Termi-nator v3.1 Cycle Sequencing kit (Applied Biosystems, Inc., Foster City,CA) in a 3130 � 1 Genetic Analyzer (Perkin-Elmer, Boston, MA). Se-quences were aligned using the ClustalW program (Ver.1.74) (33) hostedby the European Bioinformatics Institute (www.ebi.ac.uk/Tools/msa/clustalw2) with the default parameters. Manual adjustment was madewhen appropriate.

Identification and characterization of MITEs. A candidate MITE wasidentified for its small size (�400 bp) with TIRs, a noncoding CR, andDRs (1). Evidence of mobility and a putative tps further confirmed theMITE status. MITEs were classified based on TIR sequence and then sub-grouped into variants based on CR similarity.

TIRs were identified by the einverted program in the EMBOSS pack-age (34). DRs and internal TIR repeats were identified manually. The CRwas checked for amino acid encoding by GeneMark web software (http://opal.biology.gatech.edu/GeneMark/) (35). Secondary structure and sta-bilities were predicted using Vienna RNA Websuite (36). MITE copynumbers were estimated based on BLASTn searches (37) using MITEsequences as queries against the whole-genome sequences of “Ca. Liberib-acter asiaticus” strain Psy62 and phages SC1, SC2, and FP2 with a cutoff Evalue of �e�5. Phylogenetic trees of related MITE variants were gener-ated using the neighbor-joining method implemented through MEGA 4.0(38), and bootstrap analysis (1,000 replicates) was performed to assess thetree reliability.

To search for possible in cis tps, sequences of the open readingframes (ORFs) both up- and downstream of the identified MITEs inthe SC1 and SC2 genomes (23) were used as queries for BLASTxsearches in the ISfinder database (http://www-is.biotoul.fr/) (39) witha cutoff E value of 10�3.

MITE mobility and remnant types. Based on a PCR result, a “Ca.Liberibacter asiaticus” isolate showed a larger amplicon that was filled(with a MITE), a smaller amplicon that was empty (without a MITE), orboth large and small amplicons (a mixture of filled and empty). The pres-ence of both filled and empty states was considered evidence of MITEmobility. In the MITE empty “Ca. Liberibacter asiaticus” sequences, var-ious TIR remnant sequences were found. The remnant sequences wereclassified into remnant types. The frequency and distribution of remnanttypes of “Ca. Liberibacter asiaticus” isolates from different geographicpopulations were compared. Statistical significance was evaluated by chi-square analyses.

Nomenclature of MITE. A synopsis of MCLas-Xi was used for MITEnomenclature, where MCLas indicated that MITE origined from “Ca.Liberibacter asiaticus,” X was an uppercase letter assigned to a MITEgroup based on TIR sequence similarity, and i was a number for a variantbased on sequence similarity in CR.

TABLE 1 Distributions of three amplicon types with primer set Lap-PF1-f/Lap-PF1-r from “Candidatus Liberibacter asiaticus” isolates fromChina and Florida

OriginTotal no.of isolates

No. of isolates with electrophoretic type(s):

B720 B630 B350B720,B350

B630,B350

B720,B630,B350 None

ChinaYunnan 63 25 1 3 23 4 3 4Guizhou 10 7 0 0 3 0 0 0Sichuan 20 12 0 2 6 0 0 0Guangdong 67 7 0 31 0 29 0 0Guangxi 39 5 0 21 1 12 0 0Fujian 37 8 0 23 0 4 0 2Jiangxi 13 3 0 10 0 0 0 0Zhejiang 18 4 0 14 0 0 0 0Total 267 71 1 104 33 49 3 6

Florida, USA 59 0 0 5 6 0 48 0

FIG 1 Electrophoretic profiles of representative “Candidatus Liberibacter asi-aticus” isolates from PCR amplification with primer set LapPF-1f/LapPF-1r.Three amplicon types, B720, B630, and B350, are designated. CK representsDNA from healthy (non-“Ca. Liberibacter asiaticus”-infected) citrus. Lane Mon the right shows molecular markers in bp.

Wang et al.

3980 jb.asm.org Journal of Bacteriology

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

RESULTSDetection and characterization of MCLas-A. RepresentativePCR results with primer set LapPF1-f/LapPF1-r for “Ca. Liberib-acter asiaticus” isolates from China and Florida are shown in Fig.1. Three DNA bands, B720 (�720 bp), B630 (�630 bp), and B350(�350 bp), were amplified. In silico analyses showed that B720 waspresent in phage SC1 (720 bp) and in a prophage region in thegenome of “Ca. Liberibacter asiaticus” strain Psy62 (719 bp). Sim-ilarly, B630 was found in phages SC2 and FP2 (635 bp in both). Incontrast, no identical sequence of the size of B350 was found in thepublished genomes of “Ca. Liberibacter asiaticus” strain Psy62and phages SC1, SC2, and FP2. PCR results showed that B350 waspresent in 70.8% (189/267) of the isolates from China and in allisolates from Florida (59/59). Primer set LapPF1-f/LapPF1-r didnot yield any PCR amplicon from DNA samples extracted fromhealthy (non-“Ca. Liberibacter asiaticus”-infected) citrus leavescollected in California or Chongqing. The distributions of thethree DNA fragments in all isolates from China and Florida arelisted in Table 1.

A Chinese isolate, YN-JS-835, harboring all three bands, B720,B630, and B350 (Fig. 2), was selected to demonstrate the relation-ships among the three amplicons. In this isolate, the exact sizes ofB720, B630, and B350 were 731, 598, and 303 bp, respectively. TheB630 sequence of isolate YN-JS-835 was identical to the B350 se-quence except for an insertion of a 295-bp fragment (325 � 30 �295, as shown in Fig. 2). The 325-bp sequence in the B630 wascharacterized by (i) a pair of 54-bp perfect TIRs, containing threetandem repeats of 10 bp (note that the single G/C was interspacedbetween the second and the third repeat units in Fig. 3 but the

sequence was still considered to be in tandem for the convenienceof discussion); (ii) a noncoding CR of 217 bp; (iii) an entire325-bp sequence, which could form a highly stable secondarystructure at the transcriptional level, i.e., RNA (�G � �160.80kcal/mol); and(iv) a pair of 6-bp DRs outside the TIR pair with a83.3% (5/6) similarity. All these are characteristic features of aMITE. Therefore, the 325-bp insert was designated MCLas-A1, avariant of MCLas-A (Fig. 3).

Amplicons of 36 representative isolates from China and Flor-ida were sequenced (Table 2). Eight isolates yielded single B630sequences ranging from 509 to 635 bp. One isolate, GD-HZ-D64,yielded two B630 sequences of 598 and 512 bp. MCLas-A wasfound in all B630 sequences and the genomes of phage SC2 andFP2 as mentioned above. Three MCLas-A variants were identified,with MCLas-A1 being dominant (10/12 or 83.3%) (Fig. 3).Among the 10 MCLas-A1s, 6 originated from China and 4 fromFlorida. The other two variants (GD-HZ-D64-C2 and YN-MG-4)had partially deleted left TIRs. A BLASTn search revealed thatall MCLas-A sequences had no significant hits other than thesingle copies in genome sequences of phages SC2 and FP2 in theGenBank sequence database.

Referenced to the complete genome sequence of phage SC2 (aswell as phage FP2, which is identical to phage SC2), an ORF,SC2_gp120, annotated as a hypothetical protein, was 229 bp up-stream of MCLas-A (Fig. 2). Evaluation of conserved domainsshowed that the hypothetical protein is a member of theHTH_Tnp_1 (PF01527) family, which can bind to DNA and havea transposase function (3, 40). The putative transposase belongedto the IS3 family based on BLASTx searching against the ISfinder

FIG 2 Schematic comparison of MITE-related loci among the genomes of “Candidatus Liberibacter asiaticus” Psy62, phages SC1, SC2, and PF2, and “Ca.Liberibacter asiaticus” isolate YN-JS-835. Primer set LapPF1-f/LapPF1-r is marked with a pair of small arrowheads. Three amplicon types, B720, B630, and B350,are indicated. Dashed lines represent MCLas-As flanked by two dashed-line boxes indicating terminal inverted repeats (TIRs). Solid lines represent MCLas-Bflanked by two solid-line boxes (TIRs). Open reading frames (ORFs) from SC1_gp120, CLIBASIA_05620, SC2_gp120, and Psy-FP2-gm125 are identical andencode a putative transposase. ORFs from SC1_gp125, CLIBASIA_05625, SC2_gp125, and Psy-FP2-gm130 encode a putative XRE family transcriptionalregulator.

MITEs in “Candidatus Liberibacter asiaticus”

September 2013 Volume 195 Number 17 jb.asm.org 3981

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

database (39). Downstream 770 bp of MCLas-A in the genome ofSC2/FP2 was an ORF annotated as a putative transcriptional reg-ulator of the XRE family, not known to be associated with trans-posases.

Detection of B630 (MCLas-A filled) was mostly associated withB350 (MCLas-A empty). The only exception was in the case ofMCLas-A3 (Fig. 3). The detection of this MITE was not associatedwith any empty form of the MITE in isolate YN-MG-4 (Table 1).On the other hand, 148 isolates (137 [104 � 33] from China and 11[5�6] from Florida) of “Ca. Liberibacter asiaticus” were empty, withMCLas-A (B350) only, i.e., without the filled MCLas-A (B630) state(Table 1).

All MCLas-A empty sites in B350 were filled with TIR tandemrepeat remnants (TRRs) (Fig. 4). A total of 132 B350 clones (96from China and 36 from Florida) selected from 26 “Ca. Liberib-acter asiaticus” isolates (20 from China and six from Florida) weresequenced and formed six TRR types (Fig. 4). GenBank accessionnumbers of 46 representative sequences are listed in Table 2.Among the 26 “Ca. Liberibacter asiaticus” isolates (Table 2), 3isolates harbored three TRR types, 11 isolates had two TRR types,and 12 isolates had a single TRR type. There was no significantdifference (P � 0.19) in TRR type distribution between Yunnanand Florida isolates. However, isolates from Guangdong were sig-nificantly different from those in Yunnan (P � 0.001) and in Flor-ida (P � 0.001) (Fig. 4).

Identification of MCLas-B. A total of 161 isolates (107 fromChina and 54 from Florida) yielded B720 with primer set Lap-PF1-f/Lap-PF1-r (Table 1). As shown in Table 2, 16 (12 fromChina and 4 from Florida) B720 sequences, ranging from 718 to733 bp, were obtained. Each sequence contained a MITE, rangingfrom 238 to 250 bp. The MITE shared no sequence similarity withMCLas-A and was, therefore, designated MCLas-B with four vari-ants (Fig. 5). MCLas-B1 and MCLas-B2 were flanked by two23-bp perfect TIRs with no internal tandem repeats. TIRs ofMCLas-B3 were 78.3% similar to those of MCLas-B1 and MCLas-B2. MCLas-B4, represented in a single isolate, YN-JS-4, shared

almost identical TIRs with MCLas-B1 and MCLas-B2 but exhib-ited significant variation in CR. MCLas-B1 was also present inphage SC1 and “Ca. Liberibacter asiaticus” strain Psy62 and wasthe dominant type (10/16, 62.5%). In the SC1 sequence, an ORF,SC1_gp120, identical to SC2_gp120 of phage SC2 was located 232bp upstream of MCLas-B1 and therefore was a putative tps.Downstream 606 bp was an ORF not known to be related to anytransposase in the genomes of both phage SC1 and “Ca. Liberib-acter asiaticus” Psy62. MCLas-B was able to form a stable second-ary structure at the RNA level (�G � �80.00 kcal/mol). No “Ca.Liberibacter asiaticus” isolates with empty MCLas-B were found,and the MITE mobility could not be evaluated. A BLASTn searchrevealed that all MCLas-B sequences were unique in the GenBanksequence database with the exception of the genomes of “Ca. Li-beribacter asiaticus” Psy62 and phage SC1, where a single ho-molog was found in each.

Coexistence status of MCLas-A and MCLas-B. In the Floridapopulation of “Ca. Liberibacter asiaticus,” 48 of the 59 (81.4%)isolates were detected with both MCLas-A (B720) and MCLas-B(B630), indicating a high level of coexistence or mix of the twoMITEs (Table 1). In the China population of “Ca. Liberibacterasiaticus,” separation of MCLas-A and MCLas-B was obvious. Atotal of 50 “Ca. Liberibacter asiaticus” isolates were MCLas-Aalone (B630 alone or B630 � B350), which accounted for 18.5%(50/267) of the total collection. Among them, 58.0% (29/50) werefrom Guangdong, whereas only 8.0% (4/50) were from Yunnan.For MCLas-B, 71 MCLas-B alone isolates were detected, whichaccounted for 26.6% (71/267) of the total collection. Amongthem, 35.2% (25/71) were from Yunnan, in contrast to the 9.9%(7/71) from Guangdong.

DISCUSSION

One characteristic of transposons, including MITEs, is the pres-ence of TIRs, which can be conveniently identified by a computerprogram such as the einverted program. However, not all se-quences with TIRs are transposons. It was the mobility of a

FIG 3 Schematic representation of MCLas-A and comparison of MCLas-A1s, MCLas-A2, and MCLas-A3. TIR, terminal inverted repeat; DR, direct repeat (graytriangle); CR, central region. Black triangle, a tandem repeat unit. In sequence alignments, a dot represents nucleotide identity to the first sequence (YN-JS-835);a dash represents a missing base; a space flanked by two slashes represents omitted nucleotides. The relatedness of MITEs is summarized by a neighbor-joiningtree on the left. Subgroups of MITEs are numbered on the right.

Wang et al.

3982 jb.asm.org Journal of Bacteriology

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

TIR-flanked sequence (both filled and empty states) that served asthe strongest evidence of a transposon, or a MITE, in this study. Akey contribution was the concept of pangenome or populationgenome, which encouraged the analyses of DNAs from multipleisolates. In this study, we selected a genomic locus with multiam-plicons from a single primer set, analyzed the variations using alarge collection of “Ca. Liberibacter asiaticus” isolates, and iden-tified a MITE, MCLas-A, something that could not be achieved bysingle-genome analyses. It should be noted that evidence ofMCLas-B mobility was not found in this study. However, the or-ganization similarity to MCLas-A, i.e., the presence of DRs, TIRs,noncoding CR, and putative tps, strongly suggested that MCLas-Bis, or was, a MITE.

A unique feature of MCLas-A is the presence of three unitsof tandem repeats in its TIRs. To our knowledge, this featurehas not been reported in any MITE. Stavrinides et al. (13) re-cently reported a MITE, E622, in Pseudomonas syringae. The168-bp TIRs of E622 contained two sets of interspaced repeats.The first set had two 12-bp units interspaced by 10 bp. Thesecond set had two 15-bp units interspaced by 18 bp. This isstill structurally very different from the TIRs of MCLas-A,where the repeats are basically in tandem (Fig. 3). Mahillon andChandler (3) concluded that TIRs could be divided into twofunctional domains: the outer terminal 2 or 3 bp, involved incleavages and strand transfer of transposition; and the innerpart, involved in transposase binding. For MCLas-A, the tan-

TABLE 2 Comparative information of 36 representative isolates of “Candidatus Liberibacter asiaticus” in China and the United States selected forsequence analyses of three amplicon types

Origin anddesignation ofisolate Yr Location

Amplicon type (accession no.)

B720 B630 B350

YunnanYN-JS-835 2008 Jianshui 731 (KC478846) 598 (KC478847) 303 (KC478848); 292 (KC478849); 303 (KC478850)YN-JS-4 2010 Jianshui 721 (KC478851) 292 (KC478852); 303 (KC478853)YN-GJ-831 2008 Gejiu 731 (KC478854) 291 (KC478855)YN-MG-7 2010 Maguan 719 (KC478856)YN-MG-4 2010 Maguan 509 (KC478857)YN-MLP-9 2010 Malipo 303 (KC478858); 313 (KC478859)

GuizhouGZ-CH-3 2009 Ceheng 731 (KC478860) 303 (KC478861); 313 (KC478862)

SichuanSC-DC-D-2 2010 Dechang 719 (KC478863)SC-NN-6 2011 Ningnan 722 (KC478864)SC-NN-9 2011 Ningnan 303 (KC478865);

GuangdongGD-SH-2 2009 Sihui 719 (KC478867)GD-QY-874 2008 Qingyuan 598 (KC478868) 303 (KC478869)GD-QY-875 2008 Qingyuan 598 (KC478870) 303 (KC478871)GD-QY-876 2008 Qingyuan 598 (KC478872) 313 (KC478873)GD-HZ-D64 2008 Huizhou 598 (KC478874); 512 (KC478875) 303 (KC478876)GD-SH-1 2009 Sihui 303 (KC478877); 292 (KC478878); 303 (KC478879)GD-GZ-A4 2010 Guangzhou 313 (KC478880)

GuangxiGX-LP-2 2009 Lipu 722 (KC478881); 718 (KC478882)GX-LZ-5 2009 Luzai 598 (KC478883) 292 (KC478884); 303 (KC478885)GX-GL-3 2011 Guiling 313 (KC478886)GX-GL-5 2011 Guiling 292 (KC478887); 303 (KC478888)GX-FC-2 2009 Fuchuan 292 (KC478889); 303 (KC478890)

FujianFJ-CT-2 2011 Changtai 731 (KC478891)FJ-XM-1 2010 Xiamen 733 (KC478892)FJ-CT-1 2011 Changtai 292 (KC478893); 303 (KC478894)FJ-YC-3 2011 Yongchun 303 (KC478895); 292 (KC478896)FJ-FZ-1 2010 Fuzhou 313 (KC478897)

JiangxiJX-GZ-4 2011 Ganzhou 731 (KC478898)

ZhejiangZJ-TZ-3 2011 Taizhou 731 (KC478899)ZJ-TZ-2 2011 Taizhou 313 (KC478900)

Florida, USAFL-132701 2009 Florida 720 (KC478901) 329 (KC478902); 340 (KC478903)FL-30286 2009 Florida 719 (KC478904) 329 (KC478905)FL-61427 2008 Florida 720 (KC478906) 635 (KC478907) 329 (KC478908); 340 (KC478909)FL-61437 2008 Florida 720 (KC478910) 635 (KC478911) 329 (KC478912); 340 (KC478913)FL-132879 2009 Florida 302 (KC478914); 312 (KC478915); 302 (KC478916)FL-132881 2009 Florida 302 (KC478917); 312 (KC478918); 302 (KC478919)

MITEs in “Candidatus Liberibacter asiaticus”

September 2013 Volume 195 Number 17 jb.asm.org 3983

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

dem repeat regions seemed to be the sites of MITE cleavage.Examination of TIR remnants in the empty sites (Fig. 4) showsthat all repeat units were almost intact, suggesting the entire 10bp were used as a cleavage site during transposition. In fact,deletion within a tandem repeat seemed to prevent the mobilityof MCLas-A3 in isolate YN-MG-4 (Fig. 3 and Table 1).

It is interesting that the distributions of TRR types are sig-nificantly different among the “Ca. Liberibacter asiaticus” pop-ulations from different geographical regions (Fig. 4). Isolatesfrom Guangdong, China, are unique, while isolates from Yun-nan, China, and Florida are similar. Based on this, it can bespeculated that the “Ca. Liberibacter asiaticus” population inYunnan is more related to the Florida population than that inGuangdong, China. The mechanism of how a specific remnanttype was formed is unknown. Current information may reflectthe imprints of MCLas-A behavior under different environ-mental conditions.

The mobility of MCLas-A strongly suggests the presence of afully functional transposase in “Ca. Liberibacter asiaticus.” Cur-rently, MITEs are considered to require a transposase acting intrans for transposition (1). With the close physical proximity, it is

tempting to assume that the transposase upstream of MCLas-A(also MCLas-B) would mediate the transposition. Further sub-stantiation of this notion is that both the putative transposase andMCLas-A belong to the IS3 family. A signature of IS3 transposaseis that the majority of their transposons terminate with 5=-TG andCA-3= (3). This matches with all the MCLas-A members in thisstudy, but not with any MCLas-B members. A consequence of thein cis tps MITE configuration is that the movement of these MITEsmight be limited to the vicinity of the tps. This could explain theobservations that only a single copy of both MCLas-A andMCLas-B were found in the bacterial/prophage genome. Trans-poson Tn10 was reported to have more efficient transpositionactivity if the transposase was provided by a tps located close by onthe same DNA molecule (41).

In contrast, no evidence of MCLas-B mobility was found inisolates from either China or Florida. This may be explained by thedifferent structure of MCLas-B, which may be associated with thelow or nonexistent transposition efficiency under the conditionswhere MCLas-A was active. In any case, further research is neededto provide direct proof of transposase activity. Related to MITEmobility, it should be noted that all MCLas-B coexisted or mixed

FIG 4 Schematic illustration of terminal inverted repeats (TIRs) of MCLas-As and their relations to tandem repeat remnants (TRRs) among the six TRR typesof “Candidatus Liberibacter asiaticus.” In MCLas, tandem repeat units are underlined, and central region is represented by in a solid rectangle. Each sequencebegins and ends with a 6-bp direct repeat. Frequencies of each TRR type in Yunnan, China (YN), Guangdong, China (GD), and Florida (FL) are tabulated on theright.

FIG 5 Schematic representations of MCLas-B and comparison of MCLas-B1, MCLas-B2, MCLas-B3, and MCLas-B4. TIR, terminal inverted repeat; DR, directrepeat (gray triangle); CR, central region. In sequence alignments, a dot represents nucleotide identity to the first sequence (SC-NN-6); a dash represents amissing base; a space flanked by two slashes represents omitted nucleotides. For demonstration of sequence variation, only nucleotide changes are listed;nucleotide insertions/deletions are not shown. The relatedness of MITEs is summarized by a neighbor-joining tree on the left. Subgroups of MITEs are numberedon the right.

Wang et al.

3984 jb.asm.org Journal of Bacteriology

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

with MCLas-A in the “Ca. Liberibacter asiaticus” population inFlorida. Since the introduction of “Ca. Liberibacter asiaticus” intoFlorida was recent, the bacterium should have originated from aregion dominated with both MCLas-A and MCLas-B. This inter-mediate location is currently unknown.

In summary, our knowledge about MITEs in bacteria is stilllimited. Studies of transposons/MITEs in “Ca. Liberibacter asiati-cus” are further limited by the unavailability of in vitro culture,making genome sequence analyses a current major tool of re-search. This study applied the pangenome concept to analyzeDNA polymorphisms of “Ca. Liberibacter asiaticus” detected byPCR, leading to the first detection and characterization of MITEor MITE type transposons in the bacterium. It is expected thatfuture studies on MITEs will enrich our knowledge of the “Ca.Liberibacter asiaticus” biology, particularly the bacterial niche-specific adaptation, mechanisms of population diversity, and bac-terial genomic evolution.

ACKNOWLEDGMENTS

This work was supported by Special Fund for Agro-scientific Research inthe Public Interest (201003067-02), Program for Changjiang Scholars andInnovative Research Team in University (PCSIRT, IRT0976), Natural Sci-ence Foundation Project of CQ CSTC (cstc2012jjA80025), ChongqingKey Laboratory of Citrus (CKLC201108), and California Citrus ResearchBoard (5300-151).

We thank X. Sun and M. Irey for providing “Ca. Liberibacter asiati-cus” DNAs and E. Civerolo for editing the manuscript.

Mention of trade names or commercial products in this publication issolely for the purpose of providing specific information and does notimply recommendation or endorsement by the U.S. Department of Agri-culture (USDA).

REFERENCES1. Delihas N. 2011. Impact of small repeat sequences on bacterial genome

evolution. Genome Biol. Evol. 3:959 –973.2. Feschotte C, Zhang X, Wessler SR. 2002. Miniature inverted-repeat

transposable elements and their relationship to established DNA trans-posons. In Craig NL, Craigie R, Gellert M, Lambowitz AM (ed), MobileDNA II. American Society for Microbiology, Washington, DC.

3. Mahillon J, Chandler M. 1998. Insertion sequences. Microbiol. Mol. Biol.Rev. 62:725–774.

4. Correia FF, Inouye S, Inouye M. 1988. A family of small repeated ele-ments with some transposon-like properties in the genome of Neisseriagonorrhoeae. J. Biol. Chem. 263:12194 –12198.

5. Bardaji L, Añorga M, Jackson RW, Martínez-Bilbao A, Yanguas-CasásN. 2011. Miniature transposable sequences are frequently mobilized in thebacterial plant pathogen Pseudomonas syringae pv. phaseolocola. PLoSOne 6:e25773. doi:10.1371/journal.pone.0025773.

6. Delihas N. 2007. Enterobacterial small mobile sequences carry open read-ing frames and are found intragenically-evolutionary implications for for-mation of new peptides. Gene Regul. Syst. Biol. 1:191–205.

7. Hulton CS, Higgins CF, Sharp PM. 1991. ERIC sequences: a novel familyof repetitive elements in the genomes of Escherichia coli, Salmonella typhi-murium and other enterobacteria. Mol. Microbiol. 5:825– 834.

8. Lin S, Hass S, Zemojtel T, Xiao P, Vingron M, Li R. 2011. Genome-widecomparison of cyanobacterial transposable elements, potential genetic di-versity indicators. Gene 473:139 –149.

9. Mazzone M, De Gregorio E, Lavitola A, Paqliarulo C, Alifano P, DiNocera PP. 2001. Whole-genome organization and functional propertiesof miniature DNA insertion sequences conserved in pathogenic Neisse-riae. Gene 278:211–222.

10. Nelson WC, Bhaya D, Heidelberg JF. 2012. Novel miniature transpos-able elements in thermophilic Synechococcus strains and their impact onan environmental population. J. Bacteriol. 194:3636 –3642.

11. Ogata H, Audic S, Barbe V, Artiquenave F, Fournier PE, Raoult D,Claverie JM. 2000. Selfish DNA in protein-coding genes of Rickettsia.Science 290:347–350.

12. Oggioni MR, Claverys JP. 1999. Repeated extragenic sequences in pro-karyotic genomes: a proposal for the origin and dynamics of the RUPelement in Streptococcus pneumoniae. Microbiology 145:2647–2653.

13. Stavrinides J, Kirzinger MWB, Beasley FC, Guttman DS. 2012. E622, aminiature, virulence-associated mobile element. J. Bacteriol. 194:509 –517.

14. Zhou F, Tran T, Xu Y. 2008. Nezha, a novel active miniature inverted-repeat transposable element in cyanobacteria. Biochem. Biophys. Res.Commun. 365:790 –794.

15. Bové JM. 2006. Huanglongbing: a destructive, newly-emerging, century-old disease of citrus. J. Plant Pathol. 88:7–37.

16. Lin KH. 1956. Observations on yellow shoot of citrus. Acta Phytopathol.Sin. 2:1–11.

17. Deng X, Tang W. 1996. The studies on detection of citrus Huanglongbingpathogen by polymerase chain reaction. J. South China Agric. Univ. 17:119 –120.

18. Tian Y, Ke S, Ke C. 1996. Detection and quantitation of citrus Huang-longbing pathogen by polymerase chain reaction. Acta Phytopathol. Sin.26:243–250.

19. Coletta-Filho HD, Targon MLPN, Takita MA, De Negri JD, Pompeu J,Machado MA, Amaral AM, Muller GW. 2004. First report of the causalagent of huanglongbing (‘Candidatus Liberibacter asiaticus’) in Brazil.Plant Dis. 88:1832.

20. Halbert SE. 2005. The discovery of huanglongbing in Florida. InProceed-ings of the 2nd International Citrus Canker and Huanglongbing Re-search Workshop, Florida Citrus Mutual, Orlando, FL.

21. Kumagai L, Levesque C, Blomquist CL, Madishetty K, Guo YY, WoodsP, Rooney-Latham S, Rascoe J, Gallindo T, Schnabel D, Polek M. 2013.First report of ‘Candidatus Liberibacter asiaticus’ associated with citrusHuanglongbing (HLB) in California. Plant Dis. 97:283.

22. Duan Y, Zhou L, Hall DG, Li W, Doddapaneni H, Lin H, Liu L, VahlingCM, Gabriel DW, Williams KP, Dickerman A, Sun Y, Gottwald T. 2009.Complete genome sequence of citrus huanglongbing bacterium, ‘Candi-datus Liberibacter asiaticus’ obtained through metagenomics. Mol. PlantMicrobe Interact. 22:1011–1020.

23. Zhang S, Flores-Cruz Z, Zhou L, Kang BH, Fleites L, Gooch MD, WulffNA, Davis MJ, Duan Y, Gabriel DW. 2011. ‘Ca. Liberibacter asiaticus’carries an excision plasmid prophage and a chromosomally integratedprophage that becomes lytic in plant infections. Mol. Plant Microbe In-teract. 24:458 – 468.

24. Zhou L, Powell CA, Hoffman MT, Fan G, Liu B, Lin H, Duan Y. 2011.Diversity and plasticity of the intracellular plant pathogen and insect sym-biont “Candidatus Liberibacter asiaticus” as revealed by hypervariableprophage genes with intragenic tandem repeats. Appl. Environ. Micro-biol. 77:6663– 6673.

25. Tettelin H, Masignani V, Cieslewicz MJ, Donati C, Medini D, WardNL, Angiuoli SV, Crabtree J, Jones AL, Durkin AS, Deboy RT,Davidsen TM, Mora M, Scarselli M, Margarit y Ros I, Peterson JD,Hauser CR, Sundaram JP, Nelson WC, Madupu R, Brinkac LM,Dodson RJ, Rosovitz MJ, Sullivan SA, Daugherty SC, Haft DH,Selengut J, Gwinn ML, Zhou L, Zafar N, Khouri H, Radune D,Dimitrov G, Watkins K, O’Connor KJ, Smith S, Utterback TR,White O, Rubens CE, Grandi G, Madoff LC, Kasper DL, Telford JL,Wessels MR, Rappuoli R, Fraser CM. 2005. Genome analysis ofmultiple pathogenic isolates of Streptococcus agalactiae: implicationsfor the microbial “pan-genome”.Proc. Natl. Acad. Sci. U. S. A. 102:13950 –13955. doi:10.1073/pnas.0506758102.

26. Chen J, Deng X, Sun X, Jones D, Irey M, Civerolo E. 2010. Guangdongand Florida populations of ‘Candidatus Liberibacter asiaticus’ distin-guished by a genomic locus with short tandem repeats. Phytopathology100:567–572.

27. Liu R, Zhang P, Pu X, Xing X, Chen J, Deng X. 2011. Analysis of aprophage gene frequency revealed population variation of ‘CandidatusLiberibacter asiaticus’ from two citrus-growing provinces in China. PlantDis. 95:431– 435.

28. Tomimura K, Miyata S, Furuya N, Kubota K, Okuda M, Subandiyah S,Hung TH, Su HJ, Iwanami T. 2009. Evaluation of genetic diversityamong ‘Candidatus Liberibacter asiaticus’ isolates collected in SoutheastAsia. Phytopathology 99:1062–1069.

29. Wang X, Zhou C, Deng X, Su H, Chen J. 2012. Molecular characteriza-tion of a mosaic locus in the genome of ‘Candidatus Liberibacter asiaticus.’BMC Microbiol. 12:18. doi:10.1186/1471-2180-12-18.

MITEs in “Candidatus Liberibacter asiaticus”

September 2013 Volume 195 Number 17 jb.asm.org 3985

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.

30. Jagoueix S, Bové JM, Garnier M. 1994. The phloem-limited bacterium ofgreening disease of citrus is a member of the alpha subdivision of theProteobacteria. Int. J. Syst. Bacteriol. 44:379 –386.

31. Murray MG, Thompson WF. 1980. Rapid isolation of high molecularweight plant DNA. Nucleic Acids Res. 8:4321– 4325.

32. Rozen S, Skaletsky HJ. 2000. Primer 3 on the WWW for general users andfor biological programmers. Methods Mol. Biol. 132:365–386.

33. Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W: improvingthe sensitivity of progressive multiple sequence alignment through se-quencing weighting, position-specific gap penalties and weight matrixchoice. Nucleic Acids Res. 22:4673– 4680.

34. Rice P, Longden I, Bleasby A. 2000. EMBOSS: the European MolecularBiology Open Software Suite. Trends Genet. 16:276 –277.

35. Besemer J, Borodovsky M. 2005. GeneMark: web software for gene find-ing in prokaryotes, eukaryotes and viruses. Nucleic Acids Res. 33:W451–W454. doi:10.1093/nar/gki487.

36. Gruber AR, Lorenz R, Bernhart SH, Neuböck R, Hofacker IL. 2008. TheVienna RNA Websuite. Nucleic Acids Res. 36:W70 –W74. doi:10.1093/nar/gkn188.

37. Altschul Gish SF, W, Miller W, Lipman DJ. 1990. Basic local alignmentsearch tool. J. Mol. Biol. 215:403– 410.

38. Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: molecular evo-lutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol.24:1596 –1599.

39. Siguier P, Perochon J, Lestrade L, Mahillon J, Chandler M. 2006.ISfinder: the reference centre for bacterial insertion sequences. NucleicAcids Res. 34:D32–D36.

40. Rousseau P, Gueguen E, Duval-Valentin Chandler M. 2004. The helix-turn-helix motif of bacterial insertion sequence IS911 transposase is re-quired for DNA binding. Nucleic Acids Res. 32:1335–1344.

41. Morisato D, Way JC, Kim HJ, Kleckner N. 1983. Tn10 transposase actspreferentially on nearby transposon ends in vivo. Cell 32:799 – 807.

Wang et al.

3986 jb.asm.org Journal of Bacteriology

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

b on

22

Dec

embe

r 20

21 b

y 18

7.87

.9.2

6.