RXLR effector reservoir in two Phytophthora species is dominated by a single rapidly evolving superfamily with more than 700 members

Jiang et al. 10.1073/pnas.0709303105.

Supporting Information

Files in this Data Supplement:

SI Figure 5
SI Figure 6
SI Figure 7
SI Figure 8
SI Table 1
SI Table 2
SI Figure 9
SI Table 3
SI Figure 10
SI Table 4
SI Figure 11
SI Figure 12
SI Figure 13
SI Figure 14
SI Table 5
SI Table 6
SI Figure 15

SI Figure 5

Fig. 5. Avh proteins discovered by recursive BLAST and HMM search. (A) Number of new Avh members found during recursive BLAST starting with Avr1b (E value <1e-5, protein identity >25%). Newly identified members were used for BLAST to search for members in the next round. The number of Avh proteins that were not found in these BLAST searches is shown by the gray bar. These Avh proteins were identified by the HMM search. (B) Family attributes of the BLAST- and HMM-identified Avh proteins.

SI Figure 6

Fig. 6. Coding potential of PsAvh genes and noncoding sequences. A set of 50 randomly selected promoter sequences of 500 bp were used as noncoding regions. The Avh genes were divided into two groups, one group lacking and one group having W, Y, or L motifs (154 and 178 members, respectively). The differences between all pairs of the three datasets were significant (P < 0.001) based on two tailed heteroscedastic t tests.

SI Figure 7

Fig. 7. Estimation of RXLR effector reservoir size although permutation tests. The searches were performed with the total set of translated genomic ORFs from P. sojae (18107) and P. ramorum (13920). The string RXLR was required between 30 and 60 amino acids after the signal-peptide cleavage site. The gray bars show the numbers derived from the permuted proteins. The whole-protein sequence except the N-terminal signal peptide was permuted. The string search for GXLG was used as a control. The residues G (glycine) and R (arginine) occur at a similar frequency in the proteome.

SI Figure 8

Fig. 8. Comparison of three datasets of RXLR predictions. (A) Flowchart of different prediction methods. (B) Venn diagram of three datasets of RXLR predictions. Prediction I refers to the dataset of Bhattacharjee et al. (1). This prediction was performed with the published gene models of P. sojae and P. ramorum. Prediction II refers to the dataset of Win et al. (2), who performed a search on the total set of translated genomic ORFs with N-terminal signal peptide. Prediction III is the dataset described in this study. The string search RXLR pool refers to proteins with a N-terminal signal peptide and the string RXLR within 30-60 amino acids after the cleavage site of the signal peptide that we identified in this study as the total set of candidates. The number of proteins in each dataset is listed in the brackets.

1. Bhattacharjee S, et al. (2006) The malarial host-targeting signal is conserved in the Irish potato famine pathogen. PLoS Pathog 2:e50.

2. Win J, et al. (2007) Adaptive evolution has targeted the C-terminal domain of the RXLR effectors of plant pathogenic oomycetes. Plant Cell 19:2349-2369.

SI Figure 9

Fig. 9. Groups of Avh genes within the superfamily. Groups containing more than three members are plotted.

SI Figure 10

Fig. 10. (A) The relationship between Avr1b and ipiO1 can be shown by several transition members. Two proteins were connected by thin lines if a pairwise BLASTP hit was found (E value <1e-5). The transition members between Avr1b and ipiO1 are connected by thick lines. (B) Sequence similarity between Avr1b and ipiO1 can be shown by eight pairs of alignments of homologous Avh proteins. Only the N-terminal parts of the proteins are shown in the alignments. The RXLR-dEER domain is shaded in black in all pairwise alignments. Similar or identical amino acid residues are shaded in gray.

SI Figure 11

Fig. 11. Transition pairs connecting Avr1b to ipiO1 via P. sojae Avh proteins. Each pair shows significant sequence similarity (E value <1e-3, sequence identity >25%). Identical residues are in black.

SI Figure 12

Fig. 12. (A) Connections between Avh groups. Avh proteins were chosen from each group to form random pairs. If the pair can be connected by transition members, a green square was plotted. If not, a gray square was plotted. The Avh families from left to right and from top to bottom are AG_1_1 to AG_1_9, AG_2 to AG_78, and AG_83. (B) Connections between permuted Avh proteins. (C) Connections between groups in a set of 1,000 randomly selected P. sojae proteins.

SI Figure 13

Fig. 13. Conserved W, Y, and L motifs in the C termini of Avh proteins. Three Avh proteins possessing different number of C-terminal motifs are shown. The start and end positions of these motifs in the amino acid sequence are shown to the left of the sequences. The shading is according to the motif profiles shown in Fig. 2C. The most frequently occurring amino acid residue is in black, and the second most frequent ones are in gray. The asterisks mark the sequence where one to two amino acid residues have been omitted in the alignment.

SI Figure 14

Fig. 14. Phylogram constructed with AG_53 and three related groups of Avh proteins. The unrooted phylogram is based on Bayesian inference analysis by the program MrBayes. Bayesian posterior probabilities are given next to the node. Two ortholog pairs are connected by dotted lines.

SI Figure 15

Fig. 15. Conserved synteny between P. sojae and P. ramorum and local genome rearrangements in the Avr1b-1 region. This same region was previously shown by Tyler et al. (1), and the conserved synteny and rearrangements are similar. The protein Pr75759 has an RXLR string, but is not an Avh member. The ortholog of Pr75759 does not have an RXLR string. Their gene positions are conserved between species. Ortholog pairs are connected by thin lines.

1. Tyler BM, et al. (2006) Phytophthora genome sequences uncover evolutionary origins and mechanisms of pathogenesis. Science 313:1261-1266.