Genomic sequence of ’Candidatus Liberibacter solanacearum’ haplotype C and its comparison with haplotype A and B genomes

RESEARCH ARTICLE Genomic sequence of ’Candidatus Liberibacter solanacearum’ haplotype C and its comparison with haplotype A and B genomes Jinhui Wang 1 *, Minna Haapalainen 1 , Thomas Schott 2 , Sarah M. Thompson 3,4 , Grant R. Smith 3,4,5 , Anne I. Nissinen 6 , Minna Pirhonen 1 1Department of Agricultural Sciences, FI-00014 University of Helsinki, Helsinki, Finland, 2Herne Genomik, Neustadt, Germany, 3The New Zealand Institute for Plant & Food Research Limited, Lincoln, New Zealand, 4Plant Biosecurity Cooperative Research Centre, Canberra, ACT, Australia, 5Better Border Biosecurity, Lincoln, New Zealand, 6Management and Production of Renewable Resources, Natural Resources Institute Finland (Luke), Jokioinen, Finland *[email protected]i Abstract Haplotypes A and B of ‘Candidatus Liberibacter solanacearum’ (CLso) are associated with diseases of solanaceous plants, especially Zebra chip disease of potato, and haplotypes C, D and E are associated with symptoms on apiaceous plants. To date, one complete genome of haplotype B and two high quality draft genomes of haplotype A have been obtained for these unculturable bacteria using metagenomics from the psyllid vector Bactericera cockerelli. Here, we present the first genomic sequences obtained for the carrot-associated CLso. These two genomic sequences of haplotype C, FIN114 (1.24 Mbp) and FIN111 (1.20 Mbp), were obtained from carrot psyllids (Trioza apicalis) harboring CLso. Genomic comparisons between the haplotypes A, B and C revealed that the genome organization differs between these haplotypes, due to large inversions and other recombinations. Comparison of proteincoding genes indicated that the core genome of CLso consists of 885 ortholog groups, with the pan-genome consisting of 1327 ortholog groups. Twenty-seven ortholog groups are unique to CLso haplotype C, whilst 11 ortholog groups shared by the haplotypes A and B, are not found in the haplotype C. Some of these ortholog groups that are not part of the core genome may encode functions related to interactions with the different host plant and psyllid species. Introduction ‘Candidatus Liberibacter solanacearum’ (CLso) was first described in connection with diseases of solanaceous crops, including potato, tomato and capsicum in New Zealand and North America [1–7]. Later, the same bacterial species was found in Europe, associated with diseases in the Apiaceae family plants carrot and celery [8–12]. Phylogenetic analysis using the combination of the 16S rRNA, 16S-23S rRNA intergenic spacer region (ISR) and 50S ribosomal protein gene sequences, revealed that the CLso bacteria found in different geographic regions were diverse and PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 1 / 21 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Wang J, Haapalainen M, Schott T, Thompson SM, Smith GR, Nissinen AI, et al. (2017) Genomic sequence of ’Candidatus Liberibacter solanacearum’ haplotype C and its comparison with haplotype A and B genomes. PLoS ONE 12(2): e0171531. doi:10.1371/journal. pone.0171531 Editor: Chih-Horng Kuo, Academia Sinica, TAIWAN Received: October 7, 2016 Accepted: January 23, 2017 Published: February 3, 2017 Copyright: ©2017 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. All the genomic and sub-genomic DNA sequence files are available in the NCBI GenBank database (accession numbers LWEB00000000, LVWB00000000, KX431889, KX431890 and KX431891). Funding: Funding for this project was provided by the Ministry of Agriculture and Forestry of Finland (project numbers 1651/311/2011, 2052/312/2011 and 1842/312/2013), Marjatta & Eino Kolli could be assigned to five separate clades: haplotype A, B, C, D or E [12,13]. CLso haplotypes A and B are associated with Zebra chip (ZC) disease of potatoes and psyllid yellows of tomato and capsicum [4], and these two haplotypes are transmitted by the tomato/potato psyllid Bactericera cockerelli S ˇulc (Hemiptera: Triozidae) in a circulative-persistent mode [2,7,14]. CLso haplotype C is associated with carrot yellowing disease in Northern Europe, where it is transmitted by the carrot psyllid, Trioza apicalis Fo¨rster [11,15,16]. In addition to findings in Finland, CLso haplotype C has been detected in Sweden, Norway and Germany [17–19]. CLso haplotypes D and E were first described in carrot and celery in Spain, and the psyllid Bactericera trigonica Hodkinson is suspected to act as a vector for these haplotypes in Spain [9,12,20]. Haplotypes D and E have also been found in carrot in France and Morocco [21–23]. In 2011, the complete genome sequence of CLso haplotype B (ZC1) was obtained via metagenomics, using DNA that had been isolated from CLso bacteria collected by immuno-capture from a pooled sample of field-captured tomato/potato psyllids and then amplified by whole genome amplification [24]. CLso ZC1 is the first and only completely assembled CLso genome to date. In 2015, two high quality draft genomes of CLso haplotype A were published. The first assembly, NZ1, was obtained using metagenomics from DNA isolated from one tomato/potato psyllid individual from a greenhouse-reared colony in New Zealand. The DNA for Illumina sequencing was amplified using whole genome amplification. The other CLso haplotype A assembly, HenneA, was obtained from DNA isolated from two tomato/potato psyllid individuals from a greenhouse-reared colony in the United States. The DNA from two psyllid individuals with high CLso titres were combined into one DNA sample for sequencing [25]. Analysis of these complete or high quality draft genomes provided insights into the biology of CLso, and revealed genetic variation between haplotypes A and B. In addition, two draft genome sequences of CLso, R1 and RSTM, were obtained from a tomato plant and a tomato/potato psyllid respectively, in California [26,27]. These two draft genomes are still highly fragmented, R1 consists of 99 contigs and RSTM consists of 26 contigs, which limits their use in genome structure analysis. However, these draft genomes provide information of the genetic variation within CLso. As genome sequences from the other haplotypes (C, D and E) were missing, a more robust analysis could not be undertaken. Of the five haplotypes of CLso, haplotype C has the most distinct vector, the carrot psyllid Trioza apicalis, which occurs in the temperate and subarctic climate areas in Northern Europe, whereas the identified or suggested vectors of the other haplotypes of CLso belong to genus Bactericera and occur in areas with temperate or tropical climates. To determine if CLso haplotype C is genetically different from haplotypes A and B, we sequenced and assembled the genome of haplotype C and undertook genome comparisons. Two draft genome sequences of CLso haplotype C were obtained from DNA isolated from two carrot psyllid individuals from south-west Finland. One of these haplotype C draft genome sequences, FIN114, was compared with the complete genome sequence of haplotype B (ZC1) and one high quality draft genome sequence of haplotype A (NZ1). Genomic comparisons of these three haplotypes of the same bacterial species revealed the size of both the core and pan genomes, and identified potential haplotype-specific genes that may be involved in the different host plant or psyllid interactions. Materials and methods Psyllid and plant samples All carrot psyllids were captured from the same population in a carrot field in Forssa, southwest Finland in the summer of 2012, with the permission of the land owner. Thereafter, the psyllids were reared on carrot plants (cv. Fontana) in a greenhouse in Jokioinen, Finland. In Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 2 / 21 Foundation, China Scholarship Council, the New Zealand Institute for Plant & Food Research Limited, Better Border Biosecurity and the Australian Government via the Plant Biosecurity Cooperative Research Centre. The authors declare that one co-author (TS) provides bioinformatics services as a commercial freelancer under the name Herne Genomik, Neustadt, Germany. The funders provided support in the form of salaries and grants for authors [AN, MH, JW, ST, GS, TS] but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section. Competing Interests: We have the following interests: Thomas Schott is employed by a commercial company, Herne Genomik, Neustadt, Germany. This does not alter our adherence to PLOS ONE policies on sharing data and materials. the transmission experiment, each psyllid individual was released on one carrot seedling enclosed in an insect cage [16]. After three days’ exposure, psyllids were removed from the carrot plants, the DNA was extracted from the psyllids using DNeasy Blood and Tissue kit (Qiagen) according to the manufacturer’s protocol, and the DNA was then eluted in 30 µl of nuclease-free water. Each DNA sample was further diluted 1/100, and 5 µl per reaction was used for quantitative PCR [16]. DNA samples from two carrot psyllid females 111 and 114 that contained a very high titre of CLso, Ct values 19.25 and 18.93 at sample dilution 10 −2 respectively, were used as the material for sequencing (S1 Table). DNA was also extracted from the carrot plants nine weeks after the exposure to psyllids, using the CTAB method as previously described [16]. DNA from the CLsoinfected carrots A2F2 and A5F2 exposed to feeding by psyllids 111 and 114, respectively, was used as a PCR template for CLso sequence validation and gap closure. DNA from a healthy control carrot A7C1 grown in an insect proof cage in a greenhouse was used as a CLso-negative PCR template. Genome sequencing Whole genome amplification was conducted on the DNA of both the carrot psyllid samples 114 and 111 using a RepliG kit (Qiagen). Library construction and Illumina HiSeq2000 sequencing was conducted by Macrogen Inc. (South Korea). A 432 bp paired-end library was sequenced for sample 114 and a 670 bp paired-end library and a 3 kb mate-pair library were sequenced for sample 111. Part of the DNA sample 114 was also sequenced on two PacBio RS SMRT cells at Expression Analysis Ltd (USA) (S1 Table). Genome assembly and gap closure Paired Illumina reads were adaptor-clipped using Mira v4.0.2 [28] and quality clipped using Sickle v1.33 (github.com/najoshi/sickle). Remaining intact read pairs were assembled using idba_ud v1.1.1 [29]. Contigs shorter than 200 nt were discarded. Remaining contigs with a blastn-hit against any published Liberibacter genome among the top 10 best hits, as well as any Liberibacter and related phage sequences available from RefSeq (as at May 2015) were used to extract potential Liberibacter read pairs using Mirabait from the Mira package with default settings. Resulting read pairs were assembled using SPAdes v3.6.1 [30] in MDA mode with k = 27,45,65. Contigs longer than 1kb were edited using Gap5 from the Staden package [31]. The CLso assembly of psyllid 111 DNA was used for predicting the contig joins of the other CLso assembly 114, and the contigs were ordered using Mauve v2.4.0 [32]. The draft genome sequence FIN114 was intensively re-sequenced to obtain high quality sequence that could be used in comparative genomic analyses. Primers binding to the contig ends were designed using Primer3 [33], and these primer pairs (S2 Table) were used to bridge potential gaps between the adjacent contigs of assembly FIN114 by either conventional or long-range PCR using DNA template from the psyllid sample 114. All conventional PCR and long-range PCR amplifications were performed using Phusion High-Fidelity DNA Polymerase (Thermo Scientific) according to the PCR protocol provided by the manufacturer. The PacBio reads were also used to predict joins between the pre-assembled contigs of assembly 114, and those predictions were confirmed by long-range PCR. In addition, to confirm the locations and sequences of the rRNA operons encoding for the 16S, 23S and 5S rRNAs and to determine SNPs or indels in these operons, each of three 5.6 kb gene regions was cloned in pCR-Blunt vector and re-sequenced. Primers (S2 Table) were selected to target the flanking regions beyond each rRNA operon end. All the amplified PCR products were gel-purified and extracted using QIAquick Gel Extraction Kit (Qiagen). The purified PCR products were ligated into pCR-Blunt plasmid vector included in the Zero Blunt PCR Cloning Kit (Thermo Scientific). Ten transformed E.coli Top10 (Thermo Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 3 / 21 Scientific) colonies were picked for each rRNA copy. The re-constructed plasmids of each clone were isolated using QIAprep Spin Miniprep Kit (Qiagen) and analyzed by restriction enzyme digestion using FastDigest EcoRI (Thermo Scientific). The complete sequences of the inserted DNA fragments were obtained by primer walking and Sanger sequencing at Macrogen Europe (The Netherlands). To confirm the presence and locations of putative prophage regions, the joining regions between prophage and bacterial chromosomal sequences were amplified with designed primers (S2 Table) by long-range PCR similarly as described above, and the PCR products were sequenced through primer walking at Macrogen Europe. The quality clipped reads of dataset 111 were mapped against the draft genome sequence of FIN114 using BWA v0.7.12 [34]. The consensus sequence for regions with reads contiguously mapping were extracted. Regions where paired end reads indicated that contigs were joined but differences in the genomes prevented mapping were fixed manually. Genome annotation Two draft genome sequences of CLso haplotype C, FIN114 and FIN111, were annotated using the NCBI Prokaryotic Genome Annotation Pipeline (NCBI_PGAP) and deposited in the GenBank under accession numbers LWEB00000000 and LVWB01000000, respectively. Phylogenetic analysis To construct a phylogenic tree, four species of ‘Candidatus Liberibacter’ and the species Liberibacter crescens were compared with each other and to twelve closely related Alphaproteobacteria, Agrobacterium tumefaciens,Sinorhizobium meliloti,Brucella melitensis,Brucella abortus, Bartonella quintana,Bartonella bacilliformi,Bartonella henselae,Bartonella vinsonii,Mesorhizobium loti,Mesorhizobium opportunistum, and two Phyllobacterium species. Rhodospirillum rubrum from the order Rhodospirillales of the class Alphaproteobacteria was used as the outgroup (S3 Table). Protein coding genes were clustered into ortholog groups using OrthoMCL v2.0.9 [35]. In total, the annotations of 96 single copy ortholog groups from all the bacterial genomes included in the analysis were manually checked. Any ortholog group related to an unknown function or possible horizontal gene transfer was removed from the set before analysis. Finally, 88 ortholog groups were aligned using Muscle v3.8.31 [36], and then the multiple alignments were trimmed and concatenated into a supermatrix. The best amino acid substitute model for this supermatrix was determined using ProtTest v3.4.1 [37]. Maximum-likelihood tree was constructed using RAxML v8.2.0 [38] and applying ‘PROTGAMMAIWAGF’ setting. Genome comparisons As intensive sequence validations and gap closure had been conducted on assembly FIN114, this draft genome sequence was used for genome comparisons. The predicted protein sequences of FIN114 were analyzed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway maps to reconstruct the metabolic pathways [39] of CLso haplotype C. The result was compared to the pathways predicted by KEGG for the curated complete ‘Liberibacter’ genomes, including ‘Ca. Liberibacter solanacearum’ (ZC1), ‘Ca. Liberibacter asiaticus’ (psy62, Gxpsy, Ishi-1), ‘Ca. Liberibacter americanus’ (Sao Paulo), ‘Ca. Liberibacter africanus’ (PTSAPSY) and Liberibacter crescens (BT-1). Closer comparisons of the mevalonate pathway between ‘Liberibacter’ and Marinobacterium species was performed through the EcoCyc database [40]. To obtain an overall view of the genome synteny between the three CLso haplotypes, the sequence FIN114 was aligned against the haplotype B ZC1 genome (accession GCA_000183665.1) and the haplotype A NZ1 genome (accession GCA_000968085.1) using progressive Mauve algorithm from Mauve v2.4.0 [32]. Because the genome assemblies of FIN114 and NZ1 are in contigs, the Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 4 / 21 complete genome sequence ZC1 was set as the reference. Long-range PCR was used to confirm the re-arrangement of several genomic regions in FIN114 in relation to ZC1. The local conserved blocks (LCBs) and the guide tree were determined by the progressive Mauve algorithm and visualized using the R package genoPlotR v0.8.4. The annotated protein sequences of assemblies NZ1, ZC1 and FIN114, representing CLso haplotypes A, B and C, respectively, were clustered into ortholog groups using OrthoMCL. Short open reading frames with less than 50 amino acid codons were filtered away. In the subsequent all-blast-all step using Blastp, the required minimum coverage over both query and subject sequence was set at 70%. The ortholog clustering result of OrthoMCL analysis was converted into an orthologs vs. haplotype binary (1/0: present/not present) matrix, and visualized as a Venn diagram using the R package eVenn v2.3.2. Those genes identified as singletons (i.e. genes not assigned to any ortholog group) in each genomic assembly were further filtered by Blastn against the other two CLso genome sequences, and any Blastn-hit with e-value lower than 1e -11 and coverage more than 50% over query was discarded as being a potential homolog. These remaining singletons were then screened by Blastn against NCBI nr database using the same threshold to identify those singletons that show similarity to sequences of other species, and the remaining singletons were considered as FIN114 specific. The same Blastn filtering was applied to the ortholog groups that were present in NZ1 and ZC1, but not in FIN114. Since the integrity of assembly FIN111 was not confirmed by PCR, it was not used in Mauve alignments, but the putative haplotype specific genes or ortholog groups identified by comparisons between FIN114, NZ1 and ZC1 were also compared by reciprocal Blastp (identity above 40% and 70% coverage on both query and subject) against the FIN111 protein dataset. For these haplotype specific genes or ortholog groups, we used Blast2GO [41] to annotate gene functions via a workflow of BLAST, Gene Ontology (GO) mapping, and InterProScan. SignalP v4.1 [42] and SecretomeP v2.0 [43] were used to identify a signal peptide or a non-classical protein secretion signal respectively, as previously described [44]. For the putative protein AYJ09_01490 and its homologs, amino acid sequence alignment was performed using Clustal X [45] and secondary structure prediction was performed using Jpred4 [46]. The complete prophage regions of FIN114, NZ1 [25] and ‘Ca. Liberibacter africanus’ PTSAPSY [47] were predicted using PHASTER [48]. The complete prophage sequences that have been characterized in previous studies, P1 and P2 from ZC1 [24], SC1 and SC2 from ‘Ca. Liberibacter asiaticus’ UF506 [49], FP2 from ‘Ca. Liberibacter asiaticus’ psy62 [50], SP2 from ‘Ca. Liberibacter americanus’ Sao Paulo [51], LC1 and LC2 from Liberibacter crescens BT-1 [52] were extracted from the genome sequences. Because the prophage genomes show mosaic architectures between the different strains, a comparison method that requires long sequence alignments between the genomes could not be used. Instead, pairwise calculation of tetra-nucleotide frequencies (TETRA) between the ‘Liberibacter’ prophage sequences was employed to estimate their relationships, as this analysis is independent of longer sequence alignments. The correlation coefficients of the TETRA of all those selected complete prophage sequences were calculated using Python package pyani (github.com/widdowquinn/pyani) and the result was visualized using R package ggplot2 v2.1.0. Results Genome features The assemblies FIN114 and FIN111 have 5 and 15 non-redundant contigs respectively, each with 300 times average coverage. The draft genome FIN114 has a GC content of 35.2% and a length of 1.24 Mbp, encoding 1067 predicted proteins. The draft genome FIN111 has a GC content of 34.9% and a length of 1.20 Mbp, encoding 1040 predicted proteins. The average Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 5 / 21 nucleotide identity (ANI) between these two assemblies is 99.87% which shows they are highly similar. The draft genome FIN114 contains one complete prophage region, designated as phage A, in contig 2, and one partial prophage, designated as phage B, between contigs 4 and 5 (Fig 1). The prophage A is 38.3 kb long and has a GC content of 41.0%. Fig 1. Circular representation of the draft genome sequence of ‘Candidatus Liberibacter solanacearum’ haplotype C FIN114. The circles represent, from outer to inner, protein coding genes on the forward strand and the reverse strand, prophage regions, tRNA, rRNA, %G +C content and GC-skew. The loci indicated with red color represent the putative haplotype C specific genes. doi:10.1371/journal.pone.0171531.g001 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 6 / 21 CLso rRNA operons Like the previously sequenced ‘Candidatus Liberibacter’ genomes, the CLso haplotype C draft genome FIN114 also contains three rRNA operons. Each of the three rRNA operons, named 16SA, 16SB and 16SC, and their variable flanking sequences were amplified by long-range PCR, and contigs with sizes 6975 bp, 6298 bp and 6825 bp assembled via primer walking and sequencing. After complete sequence alignment and removing the variable flanking sequences, the size of the rRNA operon was determined to be 5670 bp, with almost identical sequence between the three copies. They all include the genes 16S rRNA, tRNA-Ile, tRNA-Ala, 23S rRNA, 5S rRNA and tRNA-Met in the same order. Only one polymorphic site was found within the 23S rRNA sequence where 16SA differs from 16SB and 16SC at position 3848: C/T. The sequences of the three rRNA operons of haplotype C were deposited at GenBank under accession numbers KX431889, KX431890 and KX431891. Phylogenetic tree A phylogenic tree for the genus ‘Candidatus Liberibacter’ and related species was constructed using the supermatrix approach and based on 88 single-copy ortholog groups (Fig 2). The tree obtained is robust, with strong bootstrap support. The analysis clearly shows that the Fig 2. Phylogenetic tree constructed of 88 protein-coding genes from 33 bacterial genome datasets, belonging to genera ‘Candidatus Liberibacter’, Liberibacter,Agrobacterium,Sinorhizobium,Brucella,Bartonella,Mesorhizobium,Phyllobacterium and Rhodospirillum.The 88 ortholog groups were aligned using Muscle v3.8.31, and these multiple alignments were trimmed and concatenated into a supermatrix. The best amino acid substitute model was determined using ProtTest v3.4.1. Maximum-likelihood tree was constructed using RAxML v8.2.0 with ‘PROTGAMMAIWAGF’ setting. The numbers shown next to the branches indicate the percentage of bootstrap support values (1000 replicates). The branch lengths indicate the evolutionary distance as the number of base substitutions per site. doi:10.1371/journal.pone.0171531.g002 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 7 / 21 ‘Liberibacter’ species are divided into two sub-clades. All the CLso clades and the huanglongbing-associated species ‘Candidatus Liberibacter africanus’ (CLaf), ‘Candidatus Liberibacter asiaticus’ (CLas) and ‘Candidatus Liberibacter americanus’ (CLam) clustered into the plant pathogen sub-clade of ‘Candidatus Liberibacter’. Liberibacter crescens, which is non-pathogenic and culturable, was the only member of the other sub-clade. As expected, CLso haplotype C sequences FIN114 and FIN111 and all the other CLso haplotypes together form the CLso clade, in which the three haplotypes further differentiate into three distinct haplotype clades with strong bootstrap support. Prophage sequence The TETRA correlation coefficient values of 11 complete prophage sequences (Fig 3) show that CLso FIN114 prophage A sequence is highly similar to the prophage sequences from NZ1 (P1) and ZC1 (P1 and P2) with correlation coefficient values between 0.88 to 0.90, and is less correlated to prophage sequences SC1, SC2 and PF2 from CLas with values between 0.77 to 0.83. The prophage sequence of CLaf PTSAPSY is correlated to all CLso prophage sequences with correlation coefficient values from 0.86 to 0.88, and less correlated to CLas prophages. This was unexpected, since the core genome of CLaf is more closely related to CLas than to CLso (Fig 2). All the prophages found in the ‘Ca. Liberibacter’ species and L.crescens belong to order Caudovirales family Podoviridae. Differences in genome organization between CLso haplotypes A, B and C The average nucleotide identity (ANI) is 97.70% between ZC1 and FIN114, and 97.91% between NZ1 and FIN114, which indicates that these lineages belong to the same species. The ANI result agrees with the multi-locus phylogeny tree (Fig 2) showing that haplotype C clade is more closely related to the clade of haplotype A, and agrees with the guide tree of Mauve alignment (Fig 4), which also suggests that NZ1 and FIN114 are more closely related to each Fig 3. Tetranucleotide frequency correlation coefficients (TETRA) of eleven prophage sequences from ‘Candidatus Liberibacter’ species and Liberibacter crescens.CLso, ‘Candidatus Liberibacter solanacearum’; CLaf, ‘Candidatus Liberibacter africanus’; CLas, ‘Candidatus Liberibacter asiaticus’; CLam, ‘Candidatus Liberibacter americanus’; and Lcr, Liberibacter crescens. doi:10.1371/journal.pone.0171531.g003 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 8 / 21 other than to ZC1. In general, the Mauve alignment shows synteny between ZC1 and NZ1, except one major rearrangement in the contig 2 of NZ1, which was split into contig 2a and 2b for the Mauve alignment in the previous report [25], and one inverted local conserved block (LCB) in contig 1 of NZ1. In contrast, the Mauve alignment analysis between ZC1 and FIN114 reveals multiple large genome rearrangements (Fig 4). Most of these large rearrangements are located within contig 2, which is the largest contig (834 kbp) of the FIN114 genome sequence. There are multiple large genomic regions within FIN114 contig 2 that are in the reverse orientation, and in different relative positions, to that in ZC1. One large inverted region, starting approximately at nucleotide position 364000, is next to prophage A, and is thus likely to be the result of a phage-mediated rearrangement. For another large inverted region, starting at position 690600, the homologous region in ZC1 has stretches of repetitive sequence on both sides. These differences in the genomic organization in FIN114 in comparison with ZC1 were all confirmed by long-range PCR and Sanger sequencing. The largest inversion, located approximately between positions 770000 and 888000 in FIN114, is between two identical rRNA operons, 16SB and 16SC, and thus the inversion is probably a result of a recombination event between these two rRNA operons. The variable flanking sequences of these two rRNA operons were confirmed by sequencing of the cloned DNA fragments. The genomic region homologous to this inverted region is found in contig 5 of NZ1, and between nucleotides 980000 and 1095000 of the ZC1 genome. Haplotype C core genome There are no significant differences in the core genome gene content between the haplotype B ZC1 genome and the haplotype C assembly FIN114. Thus, CLso haplotype C is likely to have a Fig 4. Multiple genome alignment of the ‘Candidatus Liberibacter solanacearum’ haplotypes A, B and C. Comparison was made between the complete genome sequence of CLso haplotype B ZC1 (top), and the draft genome sequences of haplotype A NZ1 (middle) and haplotype C FIN114 (bottom). Lines connect the homologous local conserved blocks (LCBs) between the genomes, with grey color showing connection between LCBs that are in the same orientation and black color showing connection between LCBs that are in the opposite orientation. The tree shown on the left represents the guide tree of the progressive Mauve alignment. doi:10.1371/journal.pone.0171531.g004 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 9 / 21 be considered as candidate effectors. The proteins encoded by ortholog group 987 are homologous to a phage-related hypothetical protein of Bartonella. As Bartonella species are insecttransmitted animal pathogens, these proteins might contribute to the bacteria-insect interaction. The proteins encoded by ortholog group 1000 and containing a DUF1640 domain are likely to have a phage origin. A DUF1640 superfamily protein of bacteriophage AKFV33, a putative biocontrol agent for Shiga toxin-producing Escherichia coli O157:H7, encodes a tail fiber protein which may play a role in bacterial surface recognition and adhesion [82]. In this study, two genomic sequences of CLso haplotype C were assembled and analyzed, and discussed in relation to the obligate parasitic lifestyle of CLso. The comparative genome analysis including three different haplotypes of CLso may help to identify potential haplotypespecific effectors. Since haplotype C is transmitted by a different psyllid vector and has a different plant host range than the haplotypes A and B, the differences found in both the gene content and genome organization may explain some of the differences in the interactions with plants and psyllids. Besides the mining of novel gene candidates, the new haplotype C genome sequences are also useful for applied research and diagnostics. Supporting information S1 Table. Carrot psyllid (Trioza apicalis) samples used for sequencing. (DOCX) S2 Table. Primers used for ‘CandidatusLiberibacter solanacearum’ haplotype C genome gap closure and for sub-cloning of the rRNA operons and the prophage regions. (DOCX) S3 Table. RefSeq protein datasets used in the phylogenetic analysis. (DOCX) S1 Fig. Amino acid sequence homology of the hypothetical protein AYJ09_01490 to the MIF4G domain of eIF(iso)4G. Alignment of sequences corresponding to AYJ09_01490 were performed using Clustal X and the secondary structure was predicted using Jpred 4. S.lyc, S. tub, S.pen, O.gla represent sequences from Solanum lycopersicum,Solanum tuberosum,Solanum pennellii,Oryza glaberrima, respectively, and ’X1/X2’ represent different isoforms of eIF4G. The amino acid residues highlighted with colors have high conservation (above 30%) of similarity regarding the hydrophobicity character. The secondary structure predictions, helix (red) or coil (black), are displayed on the bottom row. (TIF) Acknowledgments We thank Senja Tuominen at Luke for technical assistance and Anneli Virta at Luke for sequencing, and the CSC-IT Center for Science Ltd Finland for providing computing services. Author contributions Data curation: JW TS. Formal analysis: JW MH TS SMT. Funding acquisition: JW MH GRS AIN MP. Investigation: JW MH TS SMT. Methodology: JW TS SMT. Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 16 / 21 Project administration: MP. Resources: MH AIN. Supervision: GRS MP. Validation: JW MH TS SMT. Visualization: JW MH. Writing – original draft: JW MH. Writing – review & editing: JW MH SMT GRS AIN MP. References 1. Abad JA, Bandla M, French-Monar RD, Liefting LW, and Clover GRG. First report of the detection of ’Candidatus Liberibacter’ species in zebra chip disease-infected potato plants in the United States. Plant Dis. 2009; 93: 108–109. 2. Hansen AK, Trumble JT, Stouthamer R, and Paine TD. A new huanglongbing species, "Candidatus liberibacter psyllaurous," found to infect tomato and potato, is vectored by the psyllid Bactericera cockerelli (Sulc). Appl Environ Microbiol. 2008; 74: 5862–5865. doi: 10.1128/AEM.01268-08 PMID: 18676707 3. Liefting LW, Perez-Egusquiza ZC, Clover GRG, and Anderson JAD. A new ’Candidatus Liberibacter’ species in Solanum tuberosum in New Zealand. Plant Dis. 2008; 92: 1474. 4. Liefting LW, Sutherland PW, Ward LI, Paice KL, Weir BS, and Clover GRG. A new ’Candidatus Liberibacter’ species associated with diseases of solanaceous crops. Plant Dis. 2009; 93: 208–214. 5. Liefting LW, Weir BS, Pennycook SR, and Clover GRG. ’Candidatus Liberibacter solanacearum’, associated with plants in the family Solanaceae. Int J Syst Evol Microbiol. 2009; 59: 2274–2276. doi: 10. 1099/ijs.0.007377-0 PMID: 19620372 6. Lin H, Doddapaneni H, Munyaneza JE, Civerolo EL, Sengoda VG, Buchman JL, and et al. Molecular characterization and phylogenetic analysis of 16S rRNA from a new ’Candidatus Liberibacter’ strain associated with zebra chip disease of potato (Solanum tuberosum L.) and the potato psyllid (Bactericera cockerelli Sulc). J Plant Pathol. 2009; 91: 215–219. 7. Secor GA, Rivera VV, Abad JA, Lee IM, Clover GRG, Liefting LW, and et al. Association of ’Candidatus Liberibacter solanacearum’ with zebra chip disease of potato established by graft and psyllid transmission, electron microscopy, and PCR. Plant Dis. 2009; 93: 574–583. 8. Alfaro-Ferna ´ndez A, Cebria ´n MC, Villaescusa FJ, de Mendoza AH, Ferrandiz JC, Sanjuan S, and et al. First report of ’Candidatus Liberibacter solanacearum’ in carrot in mainland Spain. Plant Dis. 2012; 96: 582. 9. Alfaro-Ferna ´ndez A, Siverio F, Cebria ´n MC, Villaescusa FJ, and Font MI. ’Candidatus Liberibacter solanacearum’ associated with Bactericera trigonica-affected carrots in the Canary Islands. Plant Dis. 2012; 96: 581. 10. Munyaneza JE, Fisher TW, Sengoda VG, Garczynski SF, Nissinen A, and Lemmetty A. First report of "Candidatus Liberibacter solanacearum" associated with psyllid-affected carrots in Europe. Plant Dis. 2010; 94: 639. 11. Munyaneza JE, Fisher TW, Sengoda VG, Garczynski SF, Nissinen A, and Lemmetty A. Association of "Candidatus Liberibacter solanacearum" with the psyllid, Trioza apicalis (Hemiptera: Triozidae) in Europe. J Econ Entomol. 2010; 103: 1060–1070. PMID: 20857712 12. Teresani GR, Bertolini E, Alfaro-Ferna ´ndez A, Martı ´nez C, Tanaka FAO, Kitajima EW, and et al. Association of ’Candidatus Liberibacter solanacearum’ with a vegetative disorder of celery in Spain and development of a real-time PCR method for its detection. Phytopathology. 2014; 104: 804–811. doi: 10.1094/ PHYTO-07-13-0182-R PMID: 24502203 13. Nelson WR, Fisher TW, and Munyaneza JE. Haplotypes of "Candidatus Liberibacter solanacearum" suggest long-standing separation. Eur J Plant Pathol. 2011; 130: 5–12. 14. Sengoda VG, Cooper WR, Swisher KD, Henne DC, and Munyaneza JE. Latent period and transmission of "Candidatus Liberibacter solanacearum" by the potato psyllid Bactericera cockerelli (Hemiptera: Triozidae). PLoS One. 2014; 9: e93475. doi: 10.1371/journal.pone.0093475 PMID: 24682175 15. Munyaneza JE, Lemmetty A, Nissinen AI, Sengoda VG, and Fisher TW. Molecular detection of aster yellows phytoplasma and "Candidatus Liberibacter Solanacearum" in carrots affected by the psyllid Trioza apicalis (Hemiptera: Triozidae) in Finland. J Plant Pathol. 2011; 93: 697–700. Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 17 / 21 16. Nissinen AI, Haapalainen M, Jauhiainen L, Lindman M, and Pirhonen M. Different symptoms in carrots caused by male and female carrot psyllid feeding and infection by ’Candidatus Liberibacter solanacearum’. Plant Pathol. 2014; 63: 812–820. 17. Munyaneza JE, Sengoda VG, Stegmark R, Arvidsson AK, Anderbrant O, Yuvaraj JK, and et al. First report of "Candidatus Liberibacter solanacearum" associated with psyllid-affected carrots in Sweden. Plant Dis. 2012; 96: 453. 18. Munyaneza JE, Sengoda VG, Sundheim L, and Meadow R. First report of "Candidatus Liberibacter solanacearum" associated with psyllid-affected carrots in Norway. Plant Dis. 2012; 96: 454. 19. Munyaneza JE, Swisher KD, Hommes M, Willhauck A, Buck H, and Meadow R. First report of ’Candidatus Liberibacter solanacearum’ associated with psyllid-infested carrots in Germany. Plant Dis. 2015; 99: 1269. 20. Nelson WR, Sengoda VG, Alfaro-Ferna ´ndez AO, Font MI, Crosslin JM, and Munyaneza JE. A new haplotype of "Candidatus Liberibacter solanacearum" identified in the Mediterranean region. Eur J Plant Pathol. 2013; 135: 633–639. 21. Bertolini E, Teresani GR, Loiseau M, Tanaka FAO, Barbe ´S, Martı ´nez C, and et al. Transmission of ’Candidatus Liberibacter solanacearum’ in carrot seeds. Plant Pathol. 2015; 64: 276–285. 22. Loiseau M, Gamier S, Boirin V, Merieau M, Leguay A, Renaudin I, and et al. First report of ’Candidatus Liberibacter solanacearum’ in carrot in France. Plant Dis. 2014; 98: 839. 23. Tahzima R, Maes M, Achbani EH, Swisher KD, Munyaneza JE, and De Jonghe K. First report of ’Candidatus Liberibacter solanacearum’ on carrot in Africa. Plant Dis. 2014; 98: 1426. 24. Lin H, Lou BH, Glynn JM, Doddapaneni H, Civerolo EL, Chen CW, and et al. The complete genome sequence of ’Candidatus Liberibacter solanacearum’, the bacterium associated with potato zebra chip disease. PLoS One. 2011; 6: e19135. doi: 10.1371/journal.pone.0019135 PMID: 21552483 25. Thompson SM, Johnson CP, Lu AY, Frampton RA, Sullivan KL, Fiers MWEJ, and et al. Genomes of ’Candidatus Liberibacter solanacearum’ haplotype A from New Zealand and the United States suggest significant genome plasticity in the species. Phytopathology. 2015; 105: 863–871. doi: 10.1094/ PHYTO-12-14-0363-FI PMID: 25822188 26. Zheng Z, Clark N, Keremane M, Lee R, Wallis C, Deng X, and et al. Whole-genome sequence of "Candidatus Liberibacter solanacearum" strain R1 from California. Genome Announc. 2014; 2: e01353–14. doi: 10.1128/genomeA.01353-14 PMID: 25540355 27. Wu F, Deng X, Liang G, Wallis C, Trumble JT, Prager S, and et al. De novo genome sequence of "Candidatus Liberibacter solanacearum" from a single potato psyllid in California. Genome Announc. 2015; 3: e01500–15. 28. Chevreux B, Wetter T, and Suhai S. Genome sequence assembly using trace signals and additional sequence information. German Conference on Bioinformatics. 1999; 99: 45–56. 29. Peng Y, Leung HCM, Yiu SM, and Chin FYL. IDBA-UD: a de novo assembler for single-cell and metagenomic sequencing data with highly uneven depth. Bioinformatics. 2012; 28: 1420–1428. doi: 10. 1093/bioinformatics/bts174 PMID: 22495754 30. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, and et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol. 2012; 19: 455–477. doi: 10.1089/cmb.2012.0021 PMID: 22506599 31. Bonfield JK, and Whitwham A. Gap5-editing the billion fragment sequence assembly. Bioinformatics. 2010; 26: 1699–1703. doi: 10.1093/bioinformatics/btq268 PMID: 20513662 32. Darling ACE, Mau B, Blattner FR, and Perna NT. Mauve: multiple alignment of conserved genomic sequence with rearrangements. Genome Res. 2004; 14: 1394–1403. doi: 10.1101/gr.2289704 PMID: 15231754 33. Koressaar T, and Remm M. Enhancements and modifications of primer design program Primer3. Bioinformatics. 2007; 23: 1289–1291. doi: 10.1093/bioinformatics/btm091 PMID: 17379693 34. Li H, and Durbin R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics. 2009; 25: 1754–1760. doi: 10.1093/bioinformatics/btp324 PMID: 19451168 35. Fischer S, Brunk BP, Chen F, Gao X, Harb OS, Iodice JB, and et al. Using OrthoMCL to assign proteins to OrthoMCL-DB groups or to cluster proteomes into new ortholog groups. Curr Protoc Bioinformatics. 2011: 6–12. 36. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004; 32: 1792–1797. doi: 10.1093/nar/gkh340 PMID: 15034147 37. Darriba D, Taboada GL, Doallo R, and Posada D. ProtTest 3: fast selection of best-fit models of protein evolution. Bioinformatics. 2011; 27: 1164–1165. doi: 10.1093/bioinformatics/btr088 PMID: 21335321 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 18 / 21 38. Stamatakis A, Ludwig T, and Meier H. RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees. Bioinformatics. 2005; 2: 456–463. 39. Kanehisa M, and Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000; 28: 27–30. PMID: 10592173 40. Keseler IM, Mackie A, Peralta-Gil M, Santos-Zavaleta A, Gama-Castro S, Bonavides-Martı ´nez C, and et al. EcoCyc: fusing model organism databases with systems biology. Nucleic Acids Res. 2013; 41: D605–D612. doi: 10.1093/nar/gks1027 PMID: 23143106 41. Conesa A, Go ¨tz S, Garcı ´a-Go ´mez JM, Terol J, Talo ´n M, and Robles M. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics. 2005; 21: 3674– 3676. doi: 10.1093/bioinformatics/bti610 PMID: 16081474 42. Petersen TN, Brunak S, von Heijne G, and Nielsen H. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat Methods. 2011; 8: 785–786. doi: 10.1038/nmeth.1701 PMID: 21959131 43. Bendtsen JD, Jensen LJ, Blom N, von Heijne G, and Brunak S. Feature-based prediction of non-classical and leaderless protein secretion. Protein Engineering Design and Selection. 2004; 17: 349–356. 44. Jain M, Fleites LA, and Gabriel DW. Prophage-encoded peroxidase in ’Candidatus Liberibacter asiaticus’ is a secreted effector that suppresses plant defenses. Mol Plant-Microbe Interact. 2015; 28: 1330– 1337. doi: 10.1094/MPMI-07-15-0145-R PMID: 26313412 45. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, and et al. Clustal W and clustal X version 2.0. Bioinformatics. 2007; 23: 2947–2948. doi: 10.1093/bioinformatics/btm404 PMID: 17846036 46. Drozdetskiy A, Cole C, Procter J, and Barton GJ. JPred4: a protein secondary structure prediction server. Nucleic Acids Res. 2015; 43: W389–W394. doi: 10.1093/nar/gkv332 PMID: 25883141 47. Lin H, Pietersen G, Han C, Read DA, Lou B, Gupta G, and et al. Complete genome sequence of "Candidatus Liberibacter africanus," a bacterium associated with citrus huanglongbing. Genome Announc. 2015; 3: e00733–15. doi: 10.1128/genomeA.00733-15 PMID: 26184931 48. Arndt D, Grant JR, Marcu A, Sajed T, Pon A, Liang YJ, and et al. PHASTER: a better, faster version of the PHAST phage search tool. Nucleic Acids Res. 2016; 44: W16–W21. doi: 10.1093/nar/gkw387 PMID: 27141966 49. Zhang SJ, Flores-Cruz Z, Zhou LJ, Kang BH, Fleites LA, Gooch MD, and et al. ’Ca. Liberibacter asiaticus’ carries an excision plasmid prophage and a chromosomally integrated prophage that becomes lytic in plant infections. Mol Plant-Microbe Interact. 2011; 24: 458–468. doi: 10.1094/MPMI-11-10-0256 PMID: 21190436 50. Zhou LJ, Powell CA, Hoffman MT, Li WB, Fan GC, Liu B, and et al. Diversity and plasticity of the intracellular plant pathogen and insect symbiont "Candidatus Liberibacter asiaticus" as revealed by hypervariable prophage genes with intragenic tandem repeats. Appl Environ Microbiol. 2011; 77: 6663–6673. doi: 10.1128/AEM.05111-11 PMID: 21784907 51. Wulff NA, Zhang S, Setubal JC, Almeida NF, Martins EC, Harakava R, and et al. The complete genome sequence of ’Candidatus Liberibacter americanus’, associated with citrus huanglongbing. Mol PlantMicrobe Interact. 2014; 27: 163–176. doi: 10.1094/MPMI-09-13-0292-R PMID: 24200077 52. Leonard MT, Fagen JR, Davis-Richardson AG, Davis MJ, and Triplett EW. Complete genome sequence of Liberibacter crescens BT-1. Standards in Genomic Sciences. 2012; 7: 271–283. doi: 10. 4056/sigs.3326772 PMID: 23408754 53. Fagen JR, Leonard MT, McCullough CM, Edirisinghe JN, Henry CS, Davis MJ, and et al. Comparative genomics of cultured and uncultured strains suggests genes essential for free-living growth of Liberibacter. PLoS One. 2014; 9: e84469. doi: 10.1371/journal.pone.0084469 PMID: 24416233 54. Kuykendall LD, Shao JY, and Hartung JS. Conservation of gene order and content in the circular chromosomes of ’Candidatus Liberibacter asiaticus’ and other Rhizobiales. PLoS One. 2012; 7: e34673. doi: 10.1371/journal.pone.0034673 PMID: 22496839 55. Albar L, Bangratz-Reyser M, Hebrard E, Ndjiondjop MN, Jones M, and Ghesquière A. Mutations in the eIF(iso)4G translation initiation factor confer high resistance of rice to Rice yellow mottle virus. Plant J. 2006; 47: 417–426. doi: 10.1111/j.1365-313X.2006.02792.x PMID: 16774645 56. Duan YP, Zhou LJ, Hall DG, Li WB, Doddapaneni H, Lin H, and et al. Complete genome sequence of citrus huanglongbing bacterium, ’Candidatus Liberibacter asiaticus’ obtained through metagenomics. Mol Plant-Microbe Interact. 2009; 22: 1011–1020. doi: 10.1094/MPMI-22-8-1011 PMID: 19589076 57. Willems A. The Family Phyllobacteriaceae. In: The Prokaryotes: Alphaproteobacteria and Betaproteobacteria. Berlin, Heidelberg: Springer Berlin Heidelberg; 2014. pp. 355–418. 58. Williams KP, Sobral BW, and Dickerman AW. A robust species tree for the Alphaproteobacteria. J Bacteriol. 2007; 189: 4578–4586. doi: 10.1128/JB.00269-07 PMID: 17483224 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 19 / 21 59. Jagoueix S, Bove ´JM, and Garnier M. The phloem-limited bacterium of greening disease of citrus is a member of the alpha-subdivision of the Proteobacteria. Int J Syst Bacteriol. 1994; 44: 379–386. doi: 10. 1099/00207713-44-3-379 PMID: 7520729 60. Jagoueix S, Bove ´JM, and Garnier M. PCR detection of the two ’Candidatus’ liberobacter species associated with greening disease of citrus. Molecular and Cellular Probes. 1996; 10: 43–50. doi: 10.1006/ mcpr.1996.0006 PMID: 8684375 61. Haapalainen M, Kivima ¨ki P, Latvala S, Rastas M, Hannukkala A, Jauhiainen L, and et al. Frequency and occurrence of the carrot pathogen ‘Candidatus Liberibacter solanacearum’ haplotype C in Finland. Plant Pathol. 2016; 62. Lin H, and Gudmestad NC. Aspects of pathogen genomics, diversity, epidemiology, vector dynamics, and disease management for a newly emerged disease of potato: zebra chip. Phytopathology. 2013; 103: 524–537. doi: 10.1094/PHYTO-09-12-0238-RVW PMID: 23268582 63. Hao GX, Pitino M, Ding F, Lin H, Stover E, and Duan YP. Induction of innate immune responses by flagellin from the intracellular bacterium, ’Candidatus Liberibacter solanacearum’. BMC Plant Biol. 2014; 14: 1–12. 64. Kim B. H., and Gadd G. M. Biosynthesis and microbial growth. In: Bacterial Physiology and Metabolism. Cambridge: Cambridge University Press; 2008. pp. 126–201 65. Oshima K, Kakizawa S, Nishigawa H, Jung HY, Wei W, Suzuki S, and et al. Reductive evolution suggested from the complete genome sequence of a plant-pathogenic phytoplasma. Nat Genet. 2004; 36: 27–29. doi: 10.1038/ng1277 PMID: 14661021 66. Baumann P, Bowditch RD, Baumann L, and Beaman B. Taxonomy of marine Pseudomonas species: P.stanieri sp. nov.; P.perfectomarina sp. nov., nom. rev.; P.nautica: and P.doudoroffii. Int J Syst Bacteriol. 1983; 33: 857–865. 67. Bowditch RD, Baumann L, and Baumann P. Description of Oceanospirillum kriegii sp. nov. and O.jannaschii sp. nov. and assignment of two species of Alteromonas to this genus as O.commune comb. nov. and O.vagum comb. nov. Curr Microbiol. 1984; 10: 221–229. 68. Choi YU, Kwon YK, Ye BR, Hyun JH, Heo SJ, Affan A, and et al. Draft genome sequence of Marinobacterium stanieri S30, a strain isolated from a coastal lagoon in Chuuk state in Micronesia. J Bacteriol. 2012; 194: 1260. doi: 10.1128/JB.06703-11 PMID: 22328757 69. Kim H, Choo YJ, Song J, Lee JS, Lee KC, and Cho JC. Marinobacterium litorale sp. nov. in the order Oceanospirillales. Int J Syst Evol Microbiol. 2007; 57: 1659–1662. doi: 10.1099/ijs.0.64892-0 PMID: 17625212 70. Satomi M, Kimura B, Hamada T, Harayama S, and Fujii T. Phylogenetic study of the genus Oceanospirillum based on 16S rRNA and gyrB genes: emended description of the genus Oceanospirillum, description of Pseudospirillum gen. nov., Oceanobacter gen. nov and Terasakielia gen. nov and transfer of Oceanospirillum jannaschii and Pseudomonas stanieri to Marinobacterium as Marinobacterium jannaschii comb. nov and Marinobacterium stanieri comb. nov. Int J Syst Evol Microbiol. 2002; 52: 739– 747. doi: 10.1099/00207713-52-3-739 PMID: 12054233 71. Pasternak Z, Pietrokovski S, Rotem O, Gophna U, Lurie-Weinberger MN, and Jurkevitch E. By their genes ye shall know them: genomic signatures of predatory bacteria. ISME J. 2013; 7: 756–769. doi: 10.1038/ismej.2012.149 PMID: 23190728 72. Wang Z, Kadouri DE, and Wu M. Genomic insights into an obligate epibiotic bacterial predator: Micavibrio aeruginosavorus ARL-13. BMC Genomics. 2011; 12: 453. doi: 10.1186/1471-2164-12-453 PMID: 21936919 73. Thao ML, Moran NA, Abbot P, Brennan EB, Burckhardt DH, and Baumann P. Cospeciation of psyllids and their primary prokaryotic endosymbionts. Appl Environ Microbiol. 2000; 66: 2898–2905. PMID: 10877784 74. Waldron DE, and Lindsay JA. Sau1: a novel line age-specific type I restriction-modification system that blocks horizontal gene transfer into Staphylococcus aureus and between S.aureus isolates of different lineages. J Bacteriol. 2006; 188: 5578–5585. doi: 10.1128/JB.00418-06 PMID: 16855248 75. Kobayashi I, Nobusato A, Kobayashi-Takahashi N, and Uchiyama I. Shaping the genome—restrictionmodification systems as mobile genetic elements. Curr Opin Genet Dev. 1999; 9: 649–656. PMID: 10607611 76. Zhou L, Powell CA, Li W, Irey M, and Duan Y. Prophage-mediated dynamics of ’Candidatus Liberibacter asiaticus’ populations, the destructive bacterial pathogens of citrus huanglongbing. PLoS One. 2013; 8: e82248. doi: 10.1371/journal.pone.0082248 PMID: 24349235 77. Yoshii M, Nishikiori M, Tomita K, Yoshioka N, Kozuka R, Naito S, and et al. The Arabidopsis cucumovirus multiplication 1 and 2loci encode tip translation initiation factors 4E and 4G. J Virol. 2004; 78: 6102–6111. doi: 10.1128/JVI.78.12.6102-6111.2004 PMID: 15163703 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 20 / 21 78. O’Brien PJ, and Ellenberger T. The Escherichia coli 3-methyladenine DNA glycosylase AlkA has a remarkably versatile active site. Journal of Biological Chemistry. 2004; 279: 26876–26884. doi: 10. 1074/jbc.M403860200 PMID: 15126496 79. Wyatt MD, Allan JM, Lau AY, Ellenberger TE, and Samson LD. 3-Methyladenine DNA glycosylases: structure, function, and biological importance. Bioessays. 1999; 21: 668–676. doi: 10.1002/(SICI)15211878(199908)21:8<668::AID-BIES6>3.0.CO;2-D PMID: 10440863 80. Thomas L, Yang CH, and Goldthwait DA. Two DNA glycosylases in Escherichia coli which release primarily 3-methyladenine. Biochemistry. 1982; 21: 1162–1169. PMID: 7041972 81. Yao JX, Saenkham P, Levy J, Ibanez F, Noroy C, Mendoza A, and et al. Interactions "Candidatus Liberibacter solanacearum"—Bactericera cockerelli: haplotype effect on vector fitness and gene expression analyses. Front Cell Infect Microbiol. 2016; 6: 62 doi: 10.3389/fcimb.2016.00062 PMID: 27376032 82. Niu YD, Stanford K, Kropinski AM, Ackermann HW, Johnson RP, She YM, and et al. Genomic, proteomic and physiological characterization of a T5-like bacteriophage for control of Shiga toxin-producing Escherichia coli O157: H7. PLoS One. 2012; 7: e34585. doi: 10.1371/journal.pone.0034585 PMID: 22514640 Genome of ’Candidatus Liberibacter solanacearum’ haplotype C PLOS ONE | DOI:10.1371/journal.pone.0171531 February 3, 2017 21 / 21