Category: Publications

Stavrinides J, McCann HC, Guttman DS

Cell. Microbiol. 2008 Feb;10(2):285-92

PubMed PMID: 18034865

Abstract

Many bacterial pathogens require a type III secretion system (T3SS) and suite of type III secreted effectors (T3SEs) to successfully colonize their hosts, extract nutrients and consequently cause disease. T3SEs, in particular, are key components of the bacterial arsenal, as they function directly inside the host to disrupt or suppress critical components of the defence network. The development of host defence and surveillance systems imposes intense selective pressures on these bacterial virulence factors, resulting in a host-pathogen co-evolutionary arms race. This arms race leaves its genetic signature in the pattern and structure of natural genetic variation found in T3SEs, thereby permitting us to infer the specific evolutionary processes and pressures driving these interactions. In this review, we summarize our current knowledge of T3SS-mediated host-pathogen co-evolution. We examine the evolution of the T3SS and the T3SEs that traverse it, in both plant and animal pathosystems, and discuss the processes that maintain these important pathogenicity determinants within pathogen populations. We go on to examine the possible origins of T3SEs, the mechanisms that give rise to new T3SEs and the processes that underlie their evolution.

Lan H, Carson R, Provart NJ, Bonner AJ

BMC Bioinformatics 2007;8:358

PubMed PMID: 17888165

Abstract

BACKGROUND: Arabidopsis thaliana is the model species of current plant genomic research with a genome size of 125 Mb and approximately 28,000 genes. The function of half of these genes is currently unknown. The purpose of this study is to infer gene function in Arabidopsis using machine-learning algorithms applied to large-scale gene expression data sets, with the goal of identifying genes that are potentially involved in plant response to abiotic stress.

RESULTS: Using in house and publicly available data, we assembled a large set of gene expression measurements for A. thaliana. Using those genes of known function, we first evaluated and compared the ability of basic machine-learning algorithms to predict which genes respond to stress. Predictive accuracy was measured using ROC50 and precision curves derived through cross validation. To improve accuracy, we developed a method for combining these classifiers using a weighted-voting scheme. The combined classifier was then trained on genes of known function and applied to genes of unknown function, identifying genes that potentially respond to stress. Visual evidence corroborating the predictions was obtained using electronic Northern analysis. Three of the predicted genes were chosen for biological validation. Gene knockout experiments confirmed that all three are involved in a variety of stress responses. The biological analysis of one of these genes (At1g16850) is presented here, where it is shown to be necessary for the normal response to temperature and NaCl.

CONCLUSION: Supervised learning methods applied to large-scale gene expression measurements can be used to predict gene function. However, the ability of basic learning methods to predict stress response varies widely and depends heavily on how much dimensionality reduction is used. Our method of combining classifiers can improve the accuracy of such predictions – in this case, predictions of genes involved in stress response in plants – and it effectively chooses the appropriate amount of dimensionality reduction automatically. The method provides a useful means of identifying genes in A. thaliana that potentially respond to stress, and we expect it would be useful in other organisms and for other gene functions.

Urquhart W, Gunawardena AH, Moeder W, Ali R, Berkowitz GA, Yoshioka K

Plant Mol. Biol. 2007 Dec;65(6):747-61

PubMed PMID: 17885810

Abstract

The hypersensitive response (HR) involves programmed cell death (PCD) in response to pathogen infection. To investigate the pathogen resistance signaling pathway, we previously identified the Arabidopsis mutant cpr22, which displays constitutive activation of multiple defense responses including HR like cell death. The cpr22 mutation has been identified as a 3 kb deletion that fuses two cyclic nucleotide-gated ion channel (CNGC)-encoding genes, ATCNGC11 and ATCNGC12, to generate a novel chimeric gene, ATCNGC11/12. In this study, we conducted a characterization of cell death induced by transient expression of ATCNGC11/12 in Nicotiana benthamiana. Electron microscopic analysis of this cell death showed similar characteristics to PCD, such as plasma membrane shrinkage and vesicle formation. The hallmark of animal PCD, fragmentation of nuclear DNA, was also observed in ATCNGC11/12-induced cell death. The development of cell death was significantly suppressed by caspase-1 inhibitors, suggesting the involvement of caspases in this process. Recently, vacuolar processing enzyme (VPE) was isolated as the first plant caspase-like protein, which is involved in HR development. In VPE-silenced plants development of cell death induced by ATCNGC11/12 was much slower and weaker compared to control plants, suggesting the involvement of VPE as a caspase in ATCNGC11/12-induced cell death. Complementation analysis using a Ca2+ uptake deficient yeast mutant demonstrated that the ATCNGC11/12 channel is permeable to Ca2+. Additionally, calcium channel blockers such as GdCl3 inhibited ATCNGC11/12-induced HR formation, whereas potassium channel blockers did not. Taken together, these results indicate that the cell death that develops in the cpr22 mutant is indeed PCD and that the chimeric channel, ATCNGC11/12, is at the point of, or up-stream of the calcium signal necessary for the development of HR.

Humphrey TV, Bonetta DT, Goring DR

New Phytol. 2007;176(1):7-21

PubMed PMID: 17803638

Abstract

The emerging view of the plant cell wall is of a dynamic and responsive structure that exists as part of a continuum with the plasma membrane and cytoskeleton. This continuum must be responsive and adaptable to normal processes of growth as well as to stresses such as wounding, attack from pathogens and mechanical stimuli. Cell expansion involving wall loosening, deposition of new materials, and subsequent rigidification must be tightly regulated to allow the maintenance of cell wall integrity and co-ordination of development. Similarly, sensing and feedback are necessary for the plant to respond to mechanical stress or pathogen attack. Currently, understanding of the sensing and feedback mechanisms utilized by plants to regulate these processes is limited, although we can learn from yeast, where the signalling pathways have been more clearly defined. Plant cell walls possess a unique and complicated structure, but it is the protein components of the wall that are likely to play a crucial role at the forefront of perception, and these are likely to include a variety of sensor and receptor systems. Recent plant research has yielded a number of interesting candidates for cell wall sensors and receptors, and we are beginning to understand the role that they may play in this crucial aspect of plant biology.

Winter D, Vinegar B, Nahal H, Ammar R, Wilson GV, Provart NJ

PLoS ONE 2007;2(8):e718

PubMed PMID: 17684564

Abstract

BACKGROUND: The exploration of microarray data and data from other high-throughput projects for hypothesis generation has become a vital aspect of post-genomic research. For the non-bioinformatics specialist, however, many of the currently available tools provide overwhelming amounts of data that are presented in a non-intuitive way.

METHODOLOGY/PRINCIPAL FINDINGS: In order to facilitate the interpretation and analysis of microarray data and data from other large-scale data sets, we have developed a tool, which we have dubbed the electronic Fluorescent Pictograph – or eFP – Browser, available at http://www.bar.utoronto.ca/, for exploring microarray and other data for hypothesis generation. This eFP Browser engine paints data from large-scale data sets onto pictographic representations of the experimental samples used to generate the data sets. We give examples of using the tool to present Arabidopsis gene expression data from the AtGenExpress Consortium (Arabidopsis eFP Browser), data for subcellular localization of Arabidopsis proteins (Cell eFP Browser), and mouse tissue atlas microarray data (Mouse eFP Browser).

CONCLUSIONS/SIGNIFICANCE: The eFP Browser software is easily adaptable to microarray or other large-scale data sets from any organism and thus should prove useful to a wide community for visualizing and interpreting these data sets for hypothesis generation.

Wang PW, Morgan RL, Scortichini M, Guttman DS

Microbiology (Reading, Engl.) 2007 Jul;153(Pt 7):2067-73

PubMed PMID: 17600051

Abstract

Pseudomonas syringae pv. avellanae (synonym: P. avellanae, Pav) is the causal agent of hazelnut decline in Greece and Italy. The population structure and evolutionary relationships of 22 strains from these two countries were examined by multilocus sequence typing (MLST) of four housekeeping genes (gapA, gltA, gyrB and rpoD). Neighbour-joining and maximum-likelihood phylogenetic analysis revealed that Greek strains isolated from the original 1976 outbreak of hazelnut decline through 1990 were very similar to Italian strains isolated from 2002 through 2004. Other Italian strains that were isolated during the 1990s were very homogeneous and clustered in a clade that was quite distinct from the Greek isolates and Italian isolates from the 2000s. A split decomposition analysis found evidence for recombination between these two highly divergent clades in two of the four MLST housekeeping genes. Incorporating these data into a broad MLST analysis of the P. syringae species complex showed that the Pav Greek and Italian strains from the 2000s clustered with P. syringae phylogroup 1, which is predominantly composed of pathogens of tomato and Brassicaceae hosts, while the Pav Italian strains from the 1990s clustered in P. syringae phylogroup 2 and are most closely related to pea (Pisum sativum L.) pathogens. These results clearly indicate that the ability to infect hazelnuts has arisen twice. This evolutionary process may be due to de novo adaptation to hazelnut by local P. syringae strains (such as the colonizers of Leguminosae crops), or the result of genetic exchange from the original Greek Pav clonal group into a phylogroup 2 strain. The latter explanation is intriguing since there is no exchange of hazelnut propagative material between Italy and Greece, which would be a likely vector for the movement of these pathogens.

Stinchcombe JR, Hoekstra HE

Heredity (Edinb) 2008 Feb;100(2):158-70

PubMed PMID: 17314923

Abstract

A central challenge in evolutionary biology is to identify genes underlying ecologically important traits and describe the fitness consequences of naturally occurring variation at these loci. To address this goal, several novel approaches have been developed, including ‘population genomics,’ where a large number of molecular markers are scored in individuals from different environments with the goal of identifying markers showing unusual patterns of variation, potentially due to selection at linked sites. Such approaches are appealing because of (1) the increasing ease of generating large numbers of genetic markers, (2) the ability to scan the genome without measuring phenotypes and (3) the simplicity of sampling individuals without knowledge of their breeding history. Although such approaches are inherently applicable to non-model systems, to date these studies have been limited in their ability to uncover functionally relevant genes. By contrast, quantitative genetics has a rich history, and more recently, quantitative trait locus (QTL) mapping has had some success in identifying genes underlying ecologically relevant variation even in novel systems. QTL mapping, however, requires (1) genetic markers that specifically differentiate parental forms, (2) a focus on a particular measurable phenotype and (3) controlled breeding and maintenance of large numbers of progeny. Here we present current advances and suggest future directions that take advantage of population genomics and quantitative genetic approaches – in both model and non-model systems. Specifically, we discuss advantages and limitations of each method and argue that a combination of the two provides a powerful approach to uncovering the molecular mechanisms responsible for adaptation.

Johnson MT, Stinchcombe JR

Trends Ecol. Evol. (Amst.) 2007 May;22(5):250-7

PubMed PMID: 17296244

Abstract

A synthesis between community ecology and evolutionary biology is emerging that identifies how genetic variation and evolution within one species can shape the ecological properties of entire communities and, in turn, how community context can govern evolutionary processes and patterns. This synthesis incorporates research on the ecology and evolution within communities over short timescales (community genetics and diffuse coevolution), as well as macroevolutionary timescales (community phylogenetics and co-diversification of communities). As we discuss here, preliminary evidence supports the hypothesis that there is a dynamic interplay between ecology and evolution within communities, yet researchers have not yet demonstrated convincingly whether, and under what circumstances, it is important for biologists to bridge community ecology and evolutionary biology. Answering this question will have important implications for both basic and applied problems in biology.

Weadick CJ, Chang BS

BMC Evol. Biol. 2007;7 Suppl 1:S11

PubMed PMID: 17288569

Abstract

BACKGROUND: The diversity of visual systems in fish has long been of interest for evolutionary biologists and neurophysiologists, and has recently begun to attract the attention of molecular evolutionary geneticists. Several recent studies on the copy number and genomic organization of visual pigment proteins, the opsins, have revealed an increased opsin diversity in fish relative to most vertebrates, brought about through recent instances of opsin duplication and divergence. However, for the subfamily of opsin genes that mediate vision at the long-wavelength end of the spectrum, the LWS opsins, it appears that most fishes possess only one or two loci, a value comparable to most other vertebrates. Here, we characterize the LWS opsins from cDNA of an individual guppy, Poecilia reticulata, a fish that is known exhibit variation in its long-wavelength sensitive visual system, mate preferences and colour patterns.

RESULTS: We identified six LWS opsins expressed within a single individual. Phylogenetic analysis revealed that these opsins descend from duplication events both pre-dating and following the divergence of the guppy lineage from that of the bluefin killifish, Lucania goodei, the closest species for which comparable data exists. Numerous amino acid substitutions exist among these different LWS opsins, many at sites known to be important for visual pigment function, including spectral sensitivity and G-protein activation. Likelihood analyses using codon-based models of evolution reveal significant changes in selective constraint along two of the guppy LWS opsin lineages.

CONCLUSION: The guppy displays an unusually high number of LWS opsins compared to other fish, and to vertebrates in general. Observing both substitutions at functionally important sites and the persistence of lineages across species boundaries suggests that these opsins might have functionally different roles, especially with regard to G-protein activation. The reasons why are currently unknown, but may relate to aspects of the guppy’s behavioural ecology, in which both male colour patterns and the female mate preferences for these colour patterns experience strong, highly variable selection pressures.

Ma W, Dong FF, Stavrinides J, Guttman DS

PLoS Genet. 2006 Dec;2(12):e209

PubMed PMID: 17194219

Abstract

The concept of the coevolutionary arms race holds a central position in our understanding of pathogen-host interactions. Here we identify the molecular mechanisms and follow the stepwise progression of an arms race in a natural system. We show how the evolution and function of the HopZ family of type III secreted effector proteins carried by the plant pathogen Pseudomonas syringae are influenced by a coevolutionary arms race between pathogen and host. We surveyed 96 isolates of P. syringae and identified three homologs (HopZ1, HopZ2, and HopZ3) distributed among approximately 45% of the strains. All alleles were sequenced and their expression was confirmed. Evolutionary analyses determined that the diverse HopZ1 homologs are ancestral to P. syringae, and have diverged via pathoadaptive mutational changes into three functional and two degenerate forms, while HopZ2 and HopZ3 have been brought into P. syringae via horizontal transfer from other ecologically similar bacteria. A PAML selection analysis revealed that the C terminus of HopZ1 is under strong positive selection. Despite the extensive genetic variation observed in this family, all three homologs have cysteine-protease activity, although their substrate specificity may vary. The introduction of the ancestral hopZ1 allele into strains harboring alternate alleles results in a resistance protein-mediated defense response in their respective hosts, which is not observed with the endogenous allele. These data indicate that the P. syringae HopZ family has undergone allelic diversification via both pathoadaptive mutational changes and horizontal transfer in response to selection imposed by the host defense system. This genetic diversity permits the pathogen to avoid host defenses while still maintaining a virulence-associated protease, thereby allowing it to thrive on its current host, while simultaneously impacting its host range.