B. Patrick Chapman and Mark Gijzen, Agriculture and Agri-Food Canada
To accelerate discovery on Chalara ash dieback disease, sequence data from the ‘interaction transcriptome’ has been made available for ‘crowdsource’ analysis. Two different samples of infected material, AT1 and AT2, have been released thus far. Previous posts describe the analysis of assembled transcripts and their likely origin, from the host plant (Fraxinus spp.) or fungus (Hymenoscyphus pseudoalbidus; syn. Chalara fraxinea) based upon best BLAST match. This work suggests that fungal transcripts may account for ~30% the AT1 and ~15% of the AT2 assembled transcripts (http://oadb.tsl.ac.uk/?p=196 and http://oadb.tsl.ac.uk/?p=284). A previous post suggested that the potentially high percentage of pathogen transcripts present in the AT1 sample is somewhat unusual for samples from infected plant tissues (How useful is the AT1 assembly? Posted 16 Dec http://oadb.tsl.ac.uk/?p=254).
Pathogen sequences can account for >50% of transcripts from diseased plant tissues
We do not believe that the high representation of fungal transcripts in the AT1 sample is necessarily unusual or alarming because past studies have shown that pathogen sequences can dominate interaction transcriptomes, especially during late infection stages. For example, 48 h after inoculation of soybean hypocotyls with the root rot pathogen Phytophthora sojae, 60 to 70% of the expressed sequence tags (ESTs) from the infected tissue were from P. sojae (Qutob et al. 2000). Likewise, 5 d after infection of Arabidopsis leaves with Hyaloperonospora arabidopsidis, 58% of non-redundant ESTs were determined to be of pathogen origin (Cabral et al. 2011). Analysis of large subunit ribosomal RNA (ls-rRNA) can also be used to estimate the fraction of host and pathogen RNA in mixed samples, in cases where the ls-rRNA is polymorphic in size for the two species. For the P. sojae and soybean interaction, this method showed that from 12 to 48 h after infection the proportion of pathogen RNA increased rapidly, from <10% (12 h) to 40% (24 h) and 70% (48 h) (Moy et al. 2004). For Arabidopsis leaves 5 d after infection with H. arabidopsidis, analysis of ls-rRNA showed that ~50% of RNA originated from the pathogen. Conclusion These examples demonstrate that the fraction of pathogen sequences in the interaction transcriptome may be large, at least for oomycete species such P. sojae and H. arabidopsidis infecting their natural hosts. Thus, we do not believe that it is particularly unusual or alarming that fungal sequences are estimated to account for 30% or more of the assembled transcripts in the AT1 sample. References Cabral A, Stassen JH, Seidl MF, Bautor J, Parker JE, Van den Ackerveken G (2011) Identification of Hyaloperonospora arabidopsidis transcript sequences expressed during infection reveals isolate-specific effectors. PLoS One 6:e19328 http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0019328 Moy P, Qutob D, Chapman BP, Atkinson I, Gijzen M (2004) Patterns of gene expression upon infection of soybean plants by Phytophthora sojae. Mol Plant Microbe Interact 17:1051-62 http://apsjournals.apsnet.org/doi/abs/10.1094/MPMI.2004.17.10.1051 Qutob D, Hraber PT, Sobral BW, Gijzen M (2000) Comparative analysis of expressed sequences in Phytophthora sojae. Plant Physiol 123:243-54 http://www.plantphysiol.org/content/123/1/243.long