Genome sequence and rapid evolution of the rice pathogen Xanthomonas oryzae pv. oryzae PXO99A

Abstract

Background: Xanthomonas oryzae pv. oryzae causes bacterial blight of rice (Oryza sativa L.), a major disease that constrains production of this staple crop in many parts of the world. We report here on the complete genome sequence of strain PXO99A and its comparison to two previously sequenced strains, KACC10331 and MAFF311018, which are highly similar to one another.

Results: The PXO99A genome is a single circular chromosome of 5,240,075 bp, considerably longer than the genomes of the other strains (4,941,439 bp and 4,940,217 bp, respectively), and it contains 5083 protein-coding genes, including 87 not found in KACC10331 or MAFF311018. PXO99A contains a greater number of virulence-associated transcription activator-like effector genes and has at least ten major chromosomal rearrangements relative to KACC10331 and MAFF311018. PXO99A contains numerous copies of diverse insertion sequence elements, members of which are associated with 7 out of 10 of the major rearrangements. A rapidly-evolving CRISPR (clustered regularly interspersed short palindromic repeats) region contains evidence of dozens of phage infections unique to the PXO99A lineage. PXO99A also contains a unique, near-perfect tandem repeat of 212 kilobases close to the replication terminus.

Conclusion: Our results provide striking evidence of genome plasticity and rapid evolution within Xanthomonas oryzae pv. oryzae. The comparisons point to sources of genomic variation and candidates for strain-specific adaptations of this pathogen that help to explain the extraordinary diversity of Xanthomonas oryzae pv. oryzae genotypes and races that have been isolated from around the world.

Figure 1

Circular representation of the Xanthomonas oryzae pv. oryzae genome. Rings illustrate, from outside to inside: protein coding genes (forward strand), protein coding genes (reverse strand), TAL effectors (green) and IS elements (red), and GC-skew plot showing (G-C)/(G+C) in 10 kilobase windows. Positive values of GC-skew indicate the leading strand of replication, negative values the lagging strand.

Figure 2

A 38.8 kb region including nonfimbrial adhesin genes that is unique to PXO99^A. A: organization of the region in the PXO99^A genome. Block arrows represent genes; inverted triangles represent insertion sequence elements. The region is flanked by DSP (dual specificity protein) and DBP (DNA binding protein) encoding genes, which are also present in MAFF and KACC. B: the corresponding locus in MAFF and KACC, missing the entire block of genes. The point of insertion/deletion maps to an ISXo5 insertion sequence element between DSP and DBP.

Figure 5

Inversions and rearrangements in PXO99^A compared to MAFF. The alignment shows regions of PXO99^A that align to the same (red) or opposite (blue) strand of MAFF. Transposase genes and their orientation (+ or -) are shown at the sites of each rearrangement. Letters A-J indicate specific rearrangement events. A: the IS element ISXoo3 is composed of two distinct and independently conserved ORFs and is responsible for an inversion spanning coordinates 267869–5114959 (all coordinates refer to the PXO99^A genome). B: ISXo8 occurs in opposite orientation at each end of a 2.6 Mbp inversion spanning positions 1356757–3898472. C: ISXo1 occurs in inverted copies at the endpoints of a 1.8 Mbp inversion spanning 1558996–3391786. D: a 33270 bp inverted region spanning 4394742–4428012 is flanked by oppositely-oriented copies of ISXo8. E: Each copy of the 212-kb duplication is flanked by ISXo5, which also occurs adjacent to two other translocations in this region. The duplication appears as two parallel diagonal lines in this box. F: ISXo8 also occurs in inverted copies at the boundaries of a 47540 bp segment that is translocated from approximately 4800000 to 685272. G: ISXoo3 flanks both ends of a 47540 bp translocation from approximately 1117000 to 4339239. H: A 9,862 bp region occurs in inverted copies at 217,455 and 4,305,307. MAFF311018 contains only one copy of this region. I,J: Segments spanning 96,753 bp (I) and 17,021 bp (J) are inverted with respect to MAFF311018 but not associated with transposases.

Figure 3

Relationship of TAL effector genes in Xoo strains PXO99^A and MAFF. The individual genes, distributed among nine loci in PXO99^A and eight in MAFF, are represented by open arrows and labeled as described in the text. Pseudogenes (truncated genes or genes with early stop codons) are indicated by an apostrophe. Genes that have identical repeat regions based on number of repeats and identity at the twelfth and thirteenth codons are connected with a black dashed line. Blue dashed lines connect genes with nearly identical repeat regions (see text). Names of previously characterized genes are centered above or below the corresponding open arrow. Colored boxes indicate TAL gene clusters (not to scale), with the same color representing loci at the same relative positions in the two genomes. Locus 4 in PXO99^A and locus 3 in MAFF are uniquely positioned in their respective genomes. The solid black rectangle and arrows beneath it represent the 212 kb direct repeat in the PXO99^A genome.

Figure 4

Alignment of PXO99^A TAL effector repetitive regions as represented by the twelfth and thirteenth residues of each repeat. Notes: 1 * indicates a proposed deletion of the thirteenth codon in the repeat; 2, novel variable codons; 3, truncation; 4, six-codon deletion; 5, N-terminal frameshift; 6, five-codon deletion in repeat.

Figure 6

Alignment of CRISPR elements from the PXO99^A, KACC, and MAFF genomes. Spacers are numbered from right (S0) to left, with the oldest elements on the right. Gaps (green boxes) indicate the positions of additional spacers in the genomes not shown here. Red lines indicate spacers shared in all three genomes, heavy black lines indicate spacers shared in just two species, and thin black lines indicate spacers that are similar but not identical between two species.

Figure 7

Compositional analysis of the PXO99A genome. Analysis of genome composition in 1000 bp windows. The red plot shows a X² analysis, in which the trinucleotide composition of each window is compared to the overall composition. The green plot shows GC content for the same windows.