Comparison of the complete genome sequence of two closely related isolates of 'Candidatus Phytoplasma australiense' reveals genome plasticity

Abstract

Background: 'Candidatus Phytoplasma australiense' is associated with at least nine diseases in Australia and New Zealand. The impact of this phytoplasma is considerable, both economically and environmentally. The genome of a NZ isolate was sequenced in an effort to understand its pathogenicity and ecology. Comparison with a closely related Australian isolate enabled us to examine mechanisms of genomic rearrangement.

Results: The complete genome sequence of a strawberry lethal yellows (SLY) isolate of 'Candidatus Phytoplasma australiense' was determined. It is a circular genome of 959,779 base pairs with 1126 predicted open reading frames. Despite being 80 kbp larger than another 'Ca. Phytoplasma australiense' isolate PAa, the variation between housekeeping genes was generally less than 1% at a nucleotide level. The difference in size between the two isolates was largely due to the number and size of potential mobile units (PMUs), which contributed to some changes in gene order. Comparison of the genomes of the two isolates revealed that the highly conserved 5' UTR of a putative DNA-directed RNA polymerase seems to be associated with insertion and rearrangement events. Two types of PMUs have been identified on the basis of the order of three to four conserved genes, with both PMUs appearing to have been present in the last common ancestor of 'Ca. Phytoplasma asteris' and 'Ca. Phytoplasma australiense'. Comparison with other phytoplasma genomes showed that modification methylases were, in general, species-specific. A putative methylase (xorIIM) found in 'Ca. Phytoplasma australiense' appeared to have no analogue in any other firmicute, and we believe has been introduced by way of lateral gene transfer. A putative retrostransposon (ltrA) analogous to that found in OY-M was present in both isolates, although all examples in PAa appear to be fragments. Comparative analysis identified highly conserved 5' and 3' UTR regions of ltrA, which may indicate how the gene is excised and inserted.

Conclusions: Comparison of two assembled 'Ca. Phytoplasma australiense' genomes has shown they possess a high level of plasticity. This comparative analysis has yielded clues as to how rearrangements occur, and the identification of sets of genes that appear to be associated with these events.

Figure 1

Circular representation of the chromosome of SLY isolate of ‘Candidatus Phytoplasma australiense’. The genome sequence was annotated by BASys (Bacterial Annotation System). The protein coding genes on the forward (red arrows) and reverse (blue arrows) strands are shown. The Clusters of Orthologous Groups (COG) functional categories of the protein coding genes in the forward (outermost circle) and reverse (innermost circle) directions are colour-coded as designated in the inset. Allocation to a particular COG functional group was determined by BLAST searches against protein databases as described in van Domselaar et al.[44].

Figure 2

Graphic representation of linearised genome sequences of ‘Candidatus Phytoplasma australiense’ isolates PAa and SLY. Coloured blocks represent unambiguous orthologous Open Reading Frames (ORFs) using PAa as the reference strain. Seventeen junctions where the two sequences diverged either by insertions, deletions or rearrangements are labelled below the representation of the PAa genome.

Figure 3

Conserved core genes of PMUs. Gene order (5′-3′) of three and four highly conserved genes found in the Potential Mobile Units (PMUs) of ‘Ca. Phytoplasma australiense’ and ‘Ca. Phytoplasma asteris’. The “AY-WB PMU” arrangement is the only one found in AY-WB, the “SLY PMU” arrangement is the only type found in SLY, whereas OY-M and PAa have examples of both types.

Figure 4

5′ UTR of rpoD. DNA sequence alignments of the 5′ Untranslated Region (UTR) of (A)rpoD Group A and (B)rpoD Group B genes found in ‘Ca. Phytoplasma australiense’. (C) Alignment of 5′ UTR region of Group B without SLY256. Although SLY256 has the best match with Group B, its protein sequence indicates that it is the most divergent of examples within that group. There is greater conservation of the 5′ UTR regions of the remaining members of Group B, although they seem to be more divergent that those of Group A. Yellow colouring indicates exact nucleotide match and blue colouring is greatest consensus match.

Figure 5

Graphical depiction of xorIIM and CHP_xap in SLY. Graphical depiction of genes and fragments of xorIIM and CHP_xap from three locations on the SLY genome. SLY164 and SLY575 are truncations with the Open Reading Frames interrupted by putative phage related protein (gepA).