Expressed sequence tags from the oomycete fish pathogen Saprolegnia parasitica reveal putative virulence factors

Abstract

Background: The oomycete Saprolegnia parasitica is one of the most economically important fish pathogens. There is a dramatic recrudescence of Saprolegnia infections in aquaculture since the use of the toxic organic dye malachite green was banned in 2002. Little is known about the molecular mechanisms underlying pathogenicity in S. parasitica and other animal pathogenic oomycetes. In this study we used a genomics approach to gain a first insight into the transcriptome of S. parasitica.

Results: We generated 1510 expressed sequence tags (ESTs) from a mycelial cDNA library of S. parasitica. A total of 1279 consensus sequences corresponding to 525944 base pairs were assembled. About half of the unigenes showed similarities to known protein sequences or motifs. The S. parasitica sequences tended to be relatively divergent from Phytophthora sequences. Based on the sequence alignments of 18 conserved proteins, the average amino acid identity between S. parasitica and three Phytophthora species was 77% compared to 93% within Phytophthora. Several S. parasitica cDNAs, such as those with similarity to fungal type I cellulose binding domain proteins, PAN/Apple module proteins, glycosyl hydrolases, proteases, as well as serine and cysteine protease inhibitors, were predicted to encode secreted proteins that could function in virulence. Some of these cDNAs were more similar to fungal proteins than to other eukaryotic proteins confirming that oomycetes and fungi share some virulence components despite their evolutionary distance

Conclusion: We provide a first glimpse into the gene content of S. parasitica, a reemerging oomycete fish pathogen. These resources will greatly accelerate research on this important pathogen. The data is available online through the Oomycete Genomics Database.

Figure 1

Phylogenetic relationships between Saprolegnia parasitica and four other stramenopiles. The phylogenetic tree was constructed using the neighbor joining method based on concatenated alignments from 18 conserved proteins (2533 amino acids). Percentile bootstrap values based on 1000 replications and obtained with the neighbor joining methods/ maximum parsimony methods are indicated at the nodes. The scale bar represents 5% weighted amino acid sequence divergence.

Figure 2

Fungal type I cellulose binding domain (CBD) is widespread in oomycetes. (A) Multiple sequence alignment of 18 oomycete type I CBDs with the domain of Trichoderma resei Cel6A (Tr_Cel6A). The four conserved cysteine residues are marked with asterisks. The glutamine and three aromatic residues that are known to be important for binding the carbohydrate substrate are shown by arrows. Sequence names refer to the Saprolegnia parasitica unigene (Sp_), Phytophthora infestans (Pi_), Phytophthora sojae (Ps_), and Phytophthora ramorum (Pr_) followed by the OGD accession number, GenBank accession number, or NCBI Trace Archive identifier (Ti number). Multiple domains originating from the same sequence are marked with the letters a or b at the end of the sequence name. (B) Consensus sequence pattern of the oomycete type I CBD. Consensus sequence was calculated using WebLogo. The bigger the letter, the more conserved the amino acid site. The positions of amino acids in the consensus sequence correspond to the positions in the sequence alignment in panel A.

Figure 3

PAN module/Apple domain is widespread in oomycetes. (A) Multiple sequence alignment of 52 oomycete PAN module/Apple domain including the two domains of the previously described CBEL protein of Phytophthora parasitica (Pp_CBELa and Pp_CBELb). The six conserved cysteine residues are marked with asterisks. Sequence names refer to the Saprolegnia parasitica unigene (Sp_), Phytophthora infestans (Pi_), Phytophthora sojae (Ps_), and Phytophthora ramorum (Pr_) followed by the OGD accession number, GenBank accession number, or NCBI Trace Archive identifier (Ti number). Multiple domains originating from the same sequence are marked with the letters a or b at the end of the sequence name. (B) Consensus sequence pattern of the oomycete oomycete PAN module/Apple domain. Consensus sequence was calculated using WebLogo. The bigger the letter, the more conserved the amino acid site. The positions of amino acids in the consensus sequence correspond to the positions in the sequence alignment in panel A.

Figure 4

Protease inhibitors in Saprolegnia parasitica. (A) Sequence alignment of Sp_001_01027 predicted amino acid sequence with representative Kazal family inhibitor domains. Protein names correspond to protease inhibitors of Saprolegnia parasitica Sp_001_01027, Phytophthora infestans EPI1 (EPI1a-b, AY586273) and EPI10 (EPI10a-c, AY586282), the crayfish Pacifastacus leniusculus (PAPI-1a-d, CAA56043), and the apicomplexan Toxoplasma gondii (TgPI-1a-d, AF121778). The conserved cysteine residues that define the Kazal family protease inhibitor domain are marked with asterisks. The putative disulfide linkages formed by cysteine residues within the predicted Kazal domains are shown. The position of the predicted P1 residues is shown by an arrow. (B) Sequence alignment of Saprolegnia parasitica Sp_001_01374 (N-terminal fragment of the mature protein), chicken egg white cystatin (CHKCYS, P01038, mature protein), rice oryzacystatin-I (Oryzacystatin-I, P09229), human cystatin A (CYTA_human, P01040) and cystatin B (CYTB_human, P04080). The proposed active-site residues in cystatins, forming the N-terminal trunk (NT) and first binding loop (L1), are indicated.