BSPP Presidential Meeting 2003
Plant Pathogen Genomics - From Sequence To Application
A. Tavanti1, A. Davidson1, N.A.R. Gow1, M. J. Maiden2, F.C. Odds1
1 School of Medical Sciences, Institute of Medical Sciences, University
of Aberdeen, Aberdeen AB25 2ZD, UK
2 The Peter Medawar Building for Pathogen Research and Department of Zoology, University of Oxford, UK.
Multilocus sequence typing (MLST) is an unambiguous procedure for characterising microbial species by the sequences of internal fragments of seven housekeeping genes. Approximately 400-700 bp internal fragments of each gene are used, as these can be accurately sequenced on both strands with an automated DNA sequencer. For each housekeeping gene, the different sequences present within a microbial species are assigned as distinct alleles and, for each isolate, the genotypes observed at each of the seven loci define the allelic profile or sequence type (ST). Each isolate of a species is therefore unambiguously characterised by a series of seven integers, which correspond to the alleles at the seven housekeeping loci. We applied multi-locus sequence typing (MLST) to a panel of 300 Candida albicans strains selected to include two that were expected to be identical. The rest came from diverse geographical and clinical sources. We sequenced 7 DNA fragments encoding housekeeping genes, including CaVPS13, CaADP1, CaSYA1 CaACC1, AAT1a, MPI and ZWF1. A total of 104 polymorphic loci was found among the 300 strains. PCR for variation at the transcribed spacer region of the 25S rRNA gene and for homozygosity at the Mating-Type-Like (MTL) locus provided additional typing information. MLST of additional sets of consecutive isolates from single patients showed they were identical or highly related. The sequences of seven fragments, obtained from hundreds of C. albicans isolates, provide data that can be used to address the population genetics and evolutionary biology of this opportunistic pathogen. The BURST (Based Upon Related Sequence Types) algorithm, which specifically examines the relationships between very closely related genotypes within clonal complexes, makes it possible to identify the most likely "ancestral" genotype of each clonal complex, from which the clonal variants have descended. MLST provides a portable, unambiguous, accurate, and highly discriminating typing system that can be used for Candida albicans molecular epidemiology.
David W. Denning
University of Manchester
Aspergillus is a genus of moulds found world-wide; over 180 species are officially recognised, some of which impact positively or negatively in medicine and industry. The genomes of A. fumigatus, nidulans and oryzae are nearing completion. Aspergillus fumigatus is the most common mould pathogen of humans, causing both life-threatening invasive disease in immunocompromised patients, aspergillomas in those with cavities left over from tuberculosis or other cystic lung disease and allergic disease in patients with atopic immune systems. Aspergillus nidulans, an occasional human pathogen (fifth most common Aspergillus spp.), is a model organism that has contributed to our understanding of genetics, gene regulation and cellular biology, while Aspergillus niger and Aspergillus oryzae are both used in industrial processes. Several other Aspergilli are thought to be significant allergens and/or to be responsible for mycotoxin production in food. To gain a better insight into the pathogenicity of this organism, an international consortium was established in 1998 to sequence the small (~30Mb) A. fumigatus genome. Sequencing is almost complete, and first pass computational annotation is being carried out by the Wellcome Trust Sanger Institute (UK) and The Institute for Genomic Research (USA). The A. nidulans genome has been sequenced to 13x by the Whitehead Institute and an automated annotation has been made available. The A. oryzae genome has been sequenced almost to completion by a Japanese consortium lead by the National Institute of Advanced Industrial Science and Technology, Tokyo. After detailed discussions it has been decided to share the genomic data between the sequencing centres to improve the quality of gene finding and annotation and to allow comparative genomics amongst the species. This process is underway. The genome sizes appear to be similar for A. fumigatus and A. nidulans, but A. oryzae is larger (37Mb). Gene finders in A. fumigatus have identified approximately 9,500 genes, in A. nidulans about 10,000 genes, and in A. oryzae about 11,500, without the information from genome comparisons. Long term curatorship has been funded by the Wellcome Trust [Central Aspergillus Data Repository (CADRE)] in 2002 to manage the information produced by the sequencing efforts, to contribute secondary annotation, and to facilitate future comparative studies by incorporating other genomic data from Aspergillus nidulans and other species as they become available.
Denning DW, Anderson MJ, Turner G, Latg JP, Bennett JW. Sequencing of the Aspergillus fumigatus genome. Lancet Infect Dis 2002;2:251-3.
Rothamsted Research, Harpenden, UK
The first systemic fungicides were introduced on to the market in the early 1970s. Because these fungicides have a very specific mode of action, often targeting a single site, they are very effective in controlling diseases, and have with less deleterious effects on plants and other non-target organisms. In contrast, with the older multi-site fungicides, active ingredients of systemic fungicides are absorbed/translocated within plant tissues and remain biologically active for long periods. While older fungicides have mainly preventative properties, most systemic fungicides have (additional) curative/eradicant properties allowing more targeted disease control after infection.
Unfortunately, more resistance problems have been encountered with the use of systemic site-specific inhibitors. This can be explained by the higher selection pressures associated with the use of systemic fungicides due to increased efficacy and eradicant properties. Additionally, for single site-specific inhibitors less genetic changes might be needed to compensate/overcome their actions. The evolution and speed of resistance development is different for each pathogen and depend on many factors such as mutation rate, generation time, sporulation, population size and occurrence of a sexual stage. Besides detoxification, reduced uptake and enhanced efflux, resistance mechanisms based on metabolic compensation, increased production of the target enzyme and target site alteration have been reported. High resistance levels affecting disease control are often associated with target-site mutations. With optimal sampling procedures and clear genotype-to-phenotype relationships, recent developments in DNA diagnostics allow us to detect and monitor resistance development much earlier so that measures can be taken in time to avoid product failures and to prolong the life time of (new) fungicides. This paper will show some practical applications of novel DNA diagnostic assays in order to track down resistance development over time (benzimidazole resistance in Mycosphaerella graminicola) and how diagnostics can be used to evaluate anti-resistance strategies (strobilurin resistance in Blumeria graminis and M. graminicola).
Pedro W. Crous
Centraalbureau voor Schimmelcultures, Uppsalalaan 8, 3584 CT Utrecht, The Netherlands
Genomics, which I define broadly to include comparative studies of genes as well as gene expression studies, is revolutionizing all aspects of mycology, including taxonomy and systematics. For the purpose of this talk, I will address phytomycology, which is the systematics of plant pathogenic fungal organisms. Although genomics is having a major impact in plant pathology, it has within a few short years completely rejuvenated phytomycology. Genomics has not only brought phytomycology to a crossroads it has actually redefined the science. Unfortunately, the synergism between plant pathology and phytomycology has largely been lost, and hence plant pathology as a science finds itself in a serious predicament. Most plant pathologists work with names that relate to outdated concepts. Few actually work with the organisms named in their grant proposals. In this talk I will present data to address various issues related to: (a) genomic data vs. the Saccardoan system and the anamorph names it gave rise to; (b) Index Fungorum, GenBank, and MycoBank, and an expos of how your virtual laboratory is being hampered by bad scientific journals; (c) pathogen diagnostics and the value of epitypification; (d) genomic data that will indicate that many of the pathogen names we are currently using need to change; (e) the need of plant pathologists to ensure that they are represented in ATOL initiatives; (f) the understanding that clonality, sex and variation mean we have to think about studying populations rather than random strains; (g) the proposition that old-fashioned voucher specimens and cultures will determine the success of genomic initiatives in systematics; (h) and the poignant reality that the day of morphological and biological species concepts has now passed, while the ecological species concept, incorporating phylogenetic data, holds the key to the future. Although the pros and cons of various proposed changes remain debatable, the mycological dogma we were taught is changing due to genomics. The biggest advantage of genomics to systematics is that it promises eventual stability to a science which underpins plant pathology.
John E. Hamer
Paradigm Genetics Inc. RTP, North Carolina 27709-4528.
The vast majority of genome initiatives over the past decade have been rationalized to funding agencies and governments as having direct impacts on human health. For agriculture the rationale has been less compelling. With a heavily subsidized northern hemisphere farm industry and continued European angst over genetically-modified crops, one may question the value of funding large scale genomic research for northern hemisphere agriculture. With European approval of commercial scale planting of GMO crops unlikely in the foreseeable future, crop protection chemicals will continue to be the major way that pests are controlled on crops. Although a $27B industry, there has been a surprising shortage of innovation in the form of newer and safer active ingredients. For example less that 30 modes of action are known for fungicides and herbicides combined. Genomic scale approaches to crop protection have the potential to revolutionize this industry while providing important tools to the academic community. By example genomic discovery platforms were developed by Paradigm Genetics Inc. for crop protection discovery. These platforms can be judged successful based on commercial partnerships and revenue generation, intellectual property generated, new chemical licenses and publications. These platforms also point the way to still newer innovations in combining recent advances in chemistry with rapid high throughput screens. For a modest up-front investment these chemical-genetic approaches could provide a set superior research tools with which to probe the complex reactions of host and parasite and provide new development opportunities for crop protection companies.
Pierre J.G.M. de Wit
Laboratory of Phytopathology, Department of Plant Sciences, Wageningen
University, Binnenhaven 5, 6709 PD Wageningen, The Netherlands.
Since their discovery by Flor in the early forties of last century, gene-for-gene systems have intrigued many plant pathologists and disease resistance breeders. More than a decade ago the first Cladosporium fulvum avirulence gene (Avr9) and the matching tomato resistance gene (Cf-9) were cloned. So far, we have cloned three additional Avr and four Ecp genes that all encode small cysteine-rich peptides that are secreted by C. fulvum in the host apoplast after penetration of tomato leaves. Expression of some of these Avr genes is also induced in vitro, when the fungus is cultured under nutrient-limiting conditions. Recognition of the Avr and Ecp genes-encoded cysteine-rich peptide elicitors is mediated by Cf resistance genes and eventually leads to a hypersensitive response (HR). C. fulvum can avoid recognition and subsequent induction of HR by various mechanisms: the Avr gene (i) is absent (Avr9; Avr4E), (ii) contains point mutations in the ORF leading to protease-sensitive elicitors or frame shift mutations leading to truncated non-active elicitor proteins (Avr2, Avr4), (iii) contains point mutations in the ORF leading to production of stable non-active elicitors (Avr4E) or (iv) contains transposon insertions in the Avr gene leading to a lack of Avr protein production (Avr2). The biochemical basis of the gene-for-gene system would imply that the Avr gene product directly interacts with the matching Cf gene product leading to HR. However, so far, no physical interaction between Avr and Cf proteins could be shown and the mechanism of Avr perception by Cf proteins is still unknown.
In contrast to Avr genes, the Ecp genes are present in all strains of C. fulvum and all encode active elicitors that can be recognized by some accessions of the wild tomato species Lycopersicon pimpinellifolium. The Cf genes occurring in those accessions have not yet been introduced in commercially-grown tomato cultivars and as a result the Ecp genes have not yet been under natural selection. All Avr and Ecp genes are supposed to represent virulence functions, but deletion of single genes did show significant effects on fungal growth in susceptible tomato cultivars. Probably, single Avrs and Ecps play only a minor role in virulence. Presently, we try to simultaneously knock down several Avr and Ecp genes in C. fulvum by RNAi. For two Avr proteins we have some indications for their biological function. The Avr4 elicitor appears to be a chitin-binding protein that can protect fungi against basic plant chitinases. Avr4 proteins encoded by virulent alleles in strains of C. fulvum are no longer recognised by Cf-4 plants, but still bind to chitin, suggesting that chitin-binding by Avr4 could represent a defensive virulence function. The Avr2 elicitor appears to be a cysteine protease inhibitor. For recognition of the Avr2 elicitor, in addition to Cf-2 protein, the tomato Rcr3 cysteine protease is required. Avr2 also inhibits Rcr3, but whether inhibition of Rcr3 by Avr2 itself or the modulation of Rcr3 by Avr2 is required for Cf-2-mediated HR is not known yet. Although we envisage that Avr and Ecp proteins of C. fulvum most probably interact with virulence targets in tomato plants that are guarded by the Cf proteins, it is not clear whether Rcr3 represents a virulence target of Avr2.
Presently we are also studying downstream responses induced by C. fulvum elicitors both in susceptible (to identify virulence targets) and resistant (to identify defence-related genes) tomato plants, and in Arabidopsis. Induction of HR in Cf-4 tomato leaves by the Avr4 elicitor is temperature-sensitive. Tomato seedlings expressing both the Avr4 gene and the Cf-4 gene quickly (within minutes) develop a synchronous systemic HR at 230C, but grow normally at 33OC. By cDNA-AFLP analysis we have identified, ca. 200 mRNAs that are either up- or down-regulated during the synchronous cell death program. By virus-induced gene silencing (VIGS) of the upregulated genes we discovered several candidates that play a role in mounting a HR in Nicotiana benthamiana expressing the Cf-4/Avr4 gene pair. The role of these genes both in HR and resistance to C. fulvum will also be tested by VIGS in tomato.Plans for future work on gene-for-gene systems will be discussed.