Jul 20, 2007 - Tableau annexe 1a . ... Tableau annexe 1b . ...... specific to P. ramorum (Table 2) using the software Primer Premier 5.00 (Premier Biosoft.
GUILLAUME J. BILODEAU
DÉTECTION ET GÉNOMIQUE DE PHYTOPHTHORA RAMORUM AGENT CAUSAL DE LA MORT SUBITE DU CHÊNE (L’ENCRE DES CHÊNES ROUGES)
Thèse présentée à la Faculté des études supérieures de l’Université Laval dans le cadre du programme de doctorat en microbiologie pour l’obtention du grade de Philosophiæ doctor (Ph.D)
DÉPARTEMENT DE BIOCHIMIE ET MICROBIOLOGIE FACULTÉ DES SCIENCES ET DE GÉNIE UNIVERSITÉ LAVAL QUÉBEC
2008
© Guillaume J. Bilodeau, 2008
i
Résumé Le Phytophthora ramorum Werres est responsable de la mort de dizaines de milliers de chênes sur la côte ouest-californienne depuis 1995. On nomme cette maladie l’encre des chênes rouges ou la mort subite du chêne. Cet organisme appartient au même genre que le Phytophthora infestans, agent causal de la famine irlandaise de la pomme de terre du 19e siècle. Il fut découvert en 1993, infectant des rhododendrons et des viornes en Europe. Depuis, il a été démontré qu’il pouvait infecter plus d’une centaine d’espèces de plantes, non seulement sur la côte ouest américaine, mais également au Canada et dans plusieurs pays d’Europe. Des mesures de quarantaine ont été mises en place, essentiellement en pépinières où sa prolifération est la plus efficace, afin d’éviter sa propagation dans d’autres États américains ou d’autres pays. Il est donc nécessaire de développer des outils de détection, et d’identification. La détection moléculaire et le génotypage peuvent être des outils importants pour découvrir et mieux comprendre la biologie de ces populations et les mouvements de cet agent pathogène. La disponibilité des séquences complètes du génome du P. ramorum, depuis 2004, apporte une nouvelle ressource pour l’identification de gènes portant du polymorphisme et la conception d’outils pour les études de populations. Les objectifs de ce projet de recherche proposés sont : de développer des outils moléculaires pour identifier le P. ramorum et le différencier des autres Phytophthora; de découvrir des loci différenciant le polymorphisme intra-spécifique du P. ramorum; de réaliser des études de populations entre les populations européennes et nord-américaines connues avec les différents marqueurs développés. Par le biais d’analyses bioinformatiques et de séquençage d’ADN, la méthodologie employée consiste au développement d’outils de diagnostic tels que la PCR en temps réel utilisant des sondes et amorces spécifiques au P. ramorum sur trois loci et de les distinguer des autres espèces de Phytophthora connues. D’ailleurs, ces analyses ont permis d’identifier plusieurs polymorphismes de nucléotides simples (SNPs) (Single Nucleotide Polymorphisms), dans treize gènes totalisant 6.3 kb utilisant la notion de volatilité des
ii codons. Dans une collection d’isolats provenant d’Europe et d’Amérique du nord, des profils de SNPs distincts et fortement corrélés avec l’origine géographique ont été identifiés. Les populations du P. ramorum en Californie et Oregon, présentent généralement trois profils uniques de SNPs et semblent plus être dérivés d’un ou quelques clones, quelques individus nouvellement trouvés présentant des insertions et délétions. En Europe plusieurs génotypes ont été retrouvés et les gènes sélectionnés semblent avoir des homologies avec des protéines de la paroi cellulaire et donc pourraient jouer un rôle dans l’adaptation. Cette thèse présente donc l’utilisation et la découverte de nouveaux gènes présentant du polymorphisme permettant de détecter et différencier le P. ramorum des autres espèces de Phytophthora mais également de connaître les origines et une identification des individus par leurs polymorphismes.
iii
Abstract Phytophthora ramorum Werres is responsible of mortality of ten thousand of oak trees on the California coast since 1995. This disease is called sudden oak death, Ramorum blight, canker. This organism is on the same genus than Phytophthora infestans, the causal agent of Irish potato famine on the 19th century. P. ramorum was discovered in 1993 infecting Rhododendron and Viburnum in Europe. Since, it was demonstrated that it could infect more than one hundred plant species, not only on the American west coast but also in Canada and many European countries. Quarantine measures were placed in effect, principally in plant nurseries where the spread is more efficient to prevent propagation in other American States or other countries. That is necessary to develop detection and identification tools. The molecular detection and genotyping could be important tools to discover and understand le biology of the population and movement of this pathogen. The availability of complete sequences from the P. ramorum genome since 2004 permitted to use new resources for identification of gene sharing potential polymorphism and conception of tools for population studies. The objectives of this research project were: to develop molecular tools for identification of P. ramorum and differentiate it from other Phytophthora species; to discover loci differentiating intra-specific polymorphism of P. ramorum; and finally to realize population studies between European and North American population known with the different markers developed. With utilisation of bioinformatics and DNA sequencing, the method used was to develop diagnosis tool with real-time PCR using specific probes and primers for P. ramorum on three different loci and distinguish it from the other Phytophthora species known. Moreover, these analyses allowed identification of multiple single nucleotide polymorphisms (SNPs), in thirteen genes for a total of 6.3 kb based on codon volatility. In a collection of isolates from Europe and North America, distinct SNPs profiles and correlated with the geographic origin were identified. P. ramorum populations in California and Oregon present generally three unique SNPs profiles and seem to drift from one or few clones, new individual newly discovered with insertion-deletion mutation. In Europe, many
iv genotypes were found and the selected genes had homologies with proteins implicated in cell wall. This could have an implication on adaptation and evolution. This thesis presents utilisation and discovery of new genes sharing polymorphisms allowing to detect and differentiate P. ramorum form other Phytophthora species and moreover to known the origin and identification of individuals by their polymorphisms.
v
Avant-Propos
Les quatre chapitres présentés dans cette thèse sont présentés sous forme d’articles publiés, soumis ou qui seront soumis dans différents journaux scientifiques. L’emploi de la langue anglaise a donc été utilisé pour ces chapitres pour la diffusion dans des revues internationales. L’introduction, la discussion et la conclusion générale de la thèse, sont quant à elles rédigées en français. Un résumé en français de même que les références de chaque chapitre ont été ajoutés à chaque section afin de rendre la lecture plus facile. Chaque chapitre, issu d’un article scientifique, est également muni d’un résumé en anglais et français, ainsi que les noms des coauteurs. L’ordre des chapitres ne se représente pas nécessairement selon l’ordre chronologique de publication mais plutôt selon l’ordre chronologique du projet de recherche et des liens entree les différents chapitres. Des précisions sont fournies ainsi que la contribution des différents coauteurs des quatre articles formant le corps de la thèse : Le premier chapitre de cette thèse fut publié dans la revue Phytopathology, de la Société américaine de Phytopathologie, en mai 2007 : Bilodeau G.J., Levesque C.A., de Cock, A.W.A.M., Duchaine C., Brière S., Uribe P., Martin F. N., and R. C. Hamelin. (2007). Molecular detection of Phytophthora ramorum by real-time polymerase chain reaction using TaqMan, SYBR Green, and molecular beacons. Phytopathology 97, 632-642. Il présente le développement et la comparaison de différents tests moléculaires utilisant la PCR en temps réel et permettant la détection de P. ramorum. J’ai réalisé l’ensemble des travaux et la rédaction de cet article. Drs. Hamelin, Lévesque et Duchaine, ont supervisé l’ensemble des travaux et ont participé à la rédaction de cet article. Les quatre autres coauteurs ont récolté et distribué le matériel nécessaire aux expérimentations de même qu’à la rédaction de l’article. Plusieurs communications orales et sous forme d’affiches ont d’ailleurs été présentées dans des congrès nationaux et internationaux depuis
vi le début du projet. De plus, une note d’application utilisant ses sondes PCR, développées en lecture finale (endpoint), à été réalisée et est distribuée par la compagnie Thermo electron corporation. Le second chapitre se réfère toujours au diagnostic du P. ramorum mais cette fois ci dans le développement d’un test de détection PCR en temps réel, mais en multiplex. Plusieurs sondes ciblant différents gènes du P. ramorum ainsi qu’une sonde contrôle d’extraction d’ADN, Phytophthora, Oomycete et plante ont été développées. Ces travaux ont été soumis à la revue : Applied and Environmental Microbiology, sous : Bilodeau, G. J., Pelletier, G., Pelletier, F., Lévesque, C. A., and R. C. Hamelin. Multiplex Real-Time PCR for detection of Phytophthora ramorum, the causal agent of sudden oak death. J’ai réalisé la majorité des travaux et la rédaction de cet article. Drs. Richard C. Hamelin, et André Lévesque, ont supervisé les travaux et ont participé à la rédaction de cet article. Les deux autres coauteurs ont participé aux développements et expérimentations du projet, de même qu’à la rédaction de l’article. Les résultats de ce chapitre ont également été présentés lors de congrès internationaux. Le troisième chapitre de cette thèse fut publié dans la Revue canadienne de phytopathologie (Canadian Journal of Plant Pathology), de la Société canadienne de Phytopathologie, en décembre 2007 : Bilodeau, G. J., Lévesque, C. A., de Cock, A.W.A.M., Brière, S.C, and R. C. Hamelin. (2007). Differentiation of European and North-American genotype of Phytophthora ramorum by Real-time polymerase chain reaction primer extension. Can. J. Plant Pathol. 29, 408-420. Cet article présente un test en PCR en temps réel permettant le génotypage et la différenciation des isolats européens et nord-américains du P. ramorum. J’ai réalisé l’ensemble des travaux et la rédaction de cet article. Drs. Richard C. Hamelin, et André Lévesque, ont supervisé l’ensemble des travaux et ont participé à la rédaction de cet article.
vii Les deux autres coauteurs ont récolté et distribué du matériel nécessaire aux expérimentations et ont participé à la rédaction de l’article. Le dernier chapitre porte sur le génotypage et l’étude des populations du P. ramorum utilisant plusieurs gènes présentant du polymorphisme SNPs et sélectionnés utilisant la notion de volatilité du génome. Cet article porte le nom: Bilodeau, G. J., Lévesque, C. A., and R. C. Hamelin. SNP Discovery and Multilocus Strain Genotyping in Phytophthora ramorum. Il sera soumis après le dépôt de la thèse. Deux communications orales ont d’ailleurs été présentées sur ce chapitre. J’ai réalisé l’ensemble des travaux et la rédaction de cet article. Drs. Richard C. Hamelin, et C. André Lévesque, ont supervisé l’ensemble des travaux et ont participé à la rédaction de cet article.
viii Remerciements…
Cette thèse a été réalisée grâce au soutien de mon comité aviseur composé de Dr Caroline Duchaine, ma directrice, Dr Richard Hamelin, mon codirecteur et Dr André Lévesque. Merci pour leur soutien et leurs conseils tout au long de ce projet de doctorat et dans l’élaboration et la poursuite des travaux. Je tiens tout d’abord à remercier Dr Richard Hamelin pour m’avoir accueilli dans son laboratoire et proposé ce projet de doctorat. Ce travail avec lui et son équipe, m’a permis d’évoluer dans le monde de la recherche avec autonomie, esprit scientifique mais surtout plaisir et enthousiasme. Merci également au Dr André Lévesque pour le support et l’expertise et ses précieux conseils. Merci au Dr Caroline Duchaine pour son soutient ses conseils et son encadrement pour la réalisation de ce doctorat. Je souhaite à tout étudiant un comité aviseur aussi agréable et enrichissant.
Je remercie chaleureusement l’équipe du Dr Hamelin au Centre de Foresterie des Laurentides (CFL), particulièrement ceux qui ont mis main à la pâte au projet, Gervais Pelletier, Josée Grondin et Françoise Pelletier. Merci également à Dr Danny Rioux pour sont aide et collaboration au projet. Un grand merci aussi à l’équipe du Dr Lévesque d’Agriculture et Agroalimentaire Canada, Ottawa pour leur aide, conseils et surtout pour le plaisir que j’ai eu de travailler avec eux. Plus précisément, Nicole Desaulniers, Hélène Rocheleau et Tharcisse Barasubyiet pour leur aide si précieuse.
Un gros merci aussi aux nombreux collaborateurs qui ont su me fournir en matériel (ADN), équipement et expertise : Stéphan Brière et son équipe à l’ACIA, Dr Arthur deCock, Dr Frank Martin, Dr Micheal Coffey, Dr Claude Husson, Dr Sabine Werres, Dr Anne Chandelier, et Dr Alenka Munda. Merci à toutes les collaborations de près et de loin qui ont permis d’échanger sur le monde de la recherche, des Oomycetes mais également de développer des amitiés.
ix Merci aux fonds CBS (Stratégique en biotechnologie canadienne) et aux fonds IRTC 04-0045RD (Initiative en recherche et technologie du CBRNE) pour leur support financier au cours de ce projet de recherche.
Et finalement, un gros merci à mon épouse et ma famille pour leur support moral, encouragements qu’ils ont su me manifester avec fierté tout au long de mes études et qui m’ont toujours poussé à aller jusqu’au bout et plus loin …
x
Table des matières Résumé..................................................................................................................................i Abstract.............................................................................................................................. iii Avant-Propos .......................................................................................................................v Table des matières ...............................................................................................................x Liste des tableaux..............................................................................................................xiv Liste des figures ................................................................................................................xvi Introduction générale ...........................................................................................................1 Introduction générale ...........................................................................................................1 1. La maladie.......................................................................................................................2 1.1 La découverte......................................................................................................2 1.2 Les hôtes du P. ramorum....................................................................................2 1.3 Localisation.........................................................................................................4 1.4 Impact économique.............................................................................................5 1.5 Mesures...............................................................................................................6 2. Biologie...........................................................................................................................7 2.1 Phytophthora ............................................................................................................7 2.2 Phytophthora ramorum.............................................................................................8 2.2.1 Caractères morphologiques du P. ramorum ......................................................8 2.2.2 La pathogénicité et variabilité morphologique et moléculaire du P. ramorum ..................................................................................................................................10 2.2.3 Le génome du P. ramorum ..............................................................................10 3. Identification et détection du P. ramorum ....................................................................12 3.1 L’isolement par culture...........................................................................................12 3.2 Les méthodes moléculaires.....................................................................................12 3.2.1 L’ELISA ..........................................................................................................13 3.2.2 La technique de PCR ................................................................................14 3.2.3 La technique de PCR en temps réel .................................................................14 3.2.4 Génotypage des populations du P. ramorum...................................................15 4.0 Problématique (objectifs et hypothèses de recherche)................................................16 5.0 Références....................................................................................................................28 Chapitre I ...........................................................................................................................36 Molecular Detection of Phytophthora ramorum by Real Time- Polymerase Chain Reaction Using TaqMan, SYBR®Green and Molecular Beacons ....................................36 I.1 Résumé/Abstract...................................................................................................38 I.1.1 Résumé..............................................................................................................38 I.1.2 Abstract.............................................................................................................39 I.2 Introduction...........................................................................................................40 I.3 Materials and methods ..........................................................................................42 I.3.1 Isolates ..............................................................................................................42 I.3.2 DNA sequencing...............................................................................................43 I.3.3 Design of primers..............................................................................................44
xi I.3.4 Design of molecular beacon and TaqMan probes ............................................44 I.3.5 PCR amplification.............................................................................................45 I.3.5.a SYBR® Green ..........................................................................................46 I.3.5.b TaqMan® probe........................................................................................46 I.3.5.c Molecular beacon......................................................................................46 I.3.5.d Standard curves.........................................................................................47 I.3.6 DNA extraction and pathogen isolation from infected plant material.............47 I.4 Results...................................................................................................................48 I.4.1 Phytophthora sequence divergence ..................................................................48 I.4.2 β-tubulin PCR assays with TaqMan, molecular beacon, and SYBR® Green ..49 I.4.3 Comparison of ITS, β-tubulin and elicitin TaqMan probes using real-time PCR 50 I.4.4 Detection of Phytophthora ramorum in infected samples................................51 I.5 Discussion.............................................................................................................52 I.6 Acknowledgements...............................................................................................57 I.7 References.................................................................................................................67 Chapitre II ..........................................................................................................................72 Multiplex Real-Time PCR for detection of Phytophthora ramorum, the causal agent of sudden oak death................................................................................................................72 II.1 Résumé/Abstract...................................................................................................74 II.1.1 Résumé..........................................................................................................74 II.1.2 Abstract.........................................................................................................75 II.2 Introduction...........................................................................................................76 II.3 Materials and methods ..........................................................................................78 II.3.1 Isolates from culture collection.....................................................................78 II.3.2 Isolates from field .........................................................................................78 II.3.3 DNA isolation from ELISA lysates ..............................................................79 II.3.4 Design of primers and probes .......................................................................79 II.3.5 Multi-gene P. ramorum multiplex, targeting Phytophthora ramorum and Phytophthora genus. .....................................................................................................80 II.3.6 Hierarchical multiplex, targeting Phytophthora ramorum, Phytophthora genus, oomycetes and plants.........................................................................................81 II.3.7 Standard curves.............................................................................................82 II.4 Results...................................................................................................................82 II.4.1 Multiplex assay targeting P. ramorum and Phytophthora genus .................82 II.4.2 P. ramorum and Phytophthora genus multiplex assay on infected host material 84 II.4.3 P. ramorum and Phytophthora genus multiplex assay using ELISA lysate DNA 85 II.4.4 Hierarchical and RuBisCO multiplexing......................................................85 II.5 Discussion.............................................................................................................87 II.6 Acknowledgements...............................................................................................93 II.7 References...............................................................................................................108 II.8 Supplemental Tables...........................................................................................112
xii Chapitre III.......................................................................................................................116 Differentiation of European and North-American genotype of Phytophthora ramorum by Real-time polymerase chain reaction primer extension...................................................116 III.1 Résumé/Abstract.................................................................................................118 III.1.1 Résumé........................................................................................................118 III.1.2 Abstract.......................................................................................................119 III.2. Introduction.....................................................................................................120 III.3. Materials and methods ....................................................................................123 III.3.1 Isolates and DNA extraction.......................................................................123 III.3.2 Primer design for DNA sequencing............................................................124 III.3.3 DNA sequencing and ASO primer extension .............................................125 III.3.4 Allele-specific genotyping..........................................................................126 III.3.5 Allele-specific genotyping of field samples ...............................................127 III.4 Results.................................................................................................................127 III.4.1 DNA polymorphisms..................................................................................127 III.4.2 Genotyping..................................................................................................128 III.4.5 Genotyping of field samples .......................................................................128 III.5 Discussion...........................................................................................................130 III.6 Acknowledgements.............................................................................................134 III.7 References...............................................................................................................150 Chapitre IV ......................................................................................................................153 SNP Discovery and Multilocus Strain Genotyping in Phytophthora ramorum. .............153 IV.1 Résumé/Abstract.................................................................................................155 IV.1.1 Résumé........................................................................................................155 IV.1.2 Abstract.......................................................................................................156 IV.2 Introduction.........................................................................................................157 IV.3. Materials and methods ....................................................................................159 IV.3.1 Isolates and DNA extraction.......................................................................159 IV.3.2 Codon volatility genome determination .....................................................160 IV.3.3 Selection of genes with high and low codon volatility...................................160 IV.3.4 Primer design ..............................................................................................161 IV.3.5 PCR amplification.......................................................................................161 IV.3.6 PCR-Single-strand conformation polymorphism (SSCP) ..........................162 IV.3.7 Sequence analysis, SNP, and haplotype determination ..............................162 IV.3.8 Sequence and population analyses..............................................................163 IV.4 Results.................................................................................................................164 IV.4.1 Phytophthora ramorum codon volatility ....................................................164 IV.4.2 Primers and PCR amplification ..................................................................165 IV.4.3 Sequences....................................................................................................165 IV.5 Discussion...........................................................................................................168 IV.6 Acknowledgements.............................................................................................174 IV.7 References...............................................................................................................189 IV.8 Supplement material ...........................................................................................194 Discussion générale et conclusion ...................................................................................199 Discussion générale et conclusion ...................................................................................200
xiii Références (Discussion générale et conclusion)..............................................................209 Annexes ...........................................................................................................................211 Annexe 1 ..........................................................................................................................212 Tableaux des hôtes naturels confirmés et possibles du Phytophthora ramorum............212 Annexe 2 ..........................................................................................................................241 Cycle de vie probable de Phytophthora ramorum..........................................................241 Annexe 3 ..........................................................................................................................242 Note d’application du test diagnostic en lecture de fin de réaction (endpoint) : ............242
xiv
Liste des tableaux Introduction Boite 1.0 Boite 2.0
Milieu de culture PARP....................................................................................12 Qu’est-ce que la volatilité? ...............................................................................19 Chapitre I
Table 1. Isolates of Phytophthora species from different culture collections used in this study..............................................................................................................................58 Table 2. Primers and probes used for PCR assays targeting Phytophthora spp. and P. ramorum........................................................................................................................60 Table 3. Isolates of Phytophthora species sequenced in this study and GenBank accession numbers for β-tubulin and elicitin genes ......................................................................61 Table 4. Number of cycles before fluorescence is detected in Phytophthora ramorum isolates tested by real-time PCR using three different gene regions (ITS, β-tubulin and elicitin) with TaqMan ...................................................................................................62 Table 5. Detection of Phytophthora ramorum in infected plant material by real-time PCR using three gene regions with TaqMan and comparison with other methods ..............63 Chapitre II Table 1. Isolates of Phytophthora and other oomycete species used in this study...............94 Table 2. Primers and probes used in the multiplex PCR assays targeting Phytophthora ramorum, Phytophthora genus, oomycetes and plants.................................................95 Table 3a. Result of the multiplex real-time PCR assay comprising three P. ramorum specific TaqMan probes and one Phytophthora genus TaqMan tested on a collection of Phytophthora ramorum from pure cultures..............................................................96 Table 3b. Results of the multiplex real-time PCR assay comprising three P. ramorum specific TaqMan probes and one Phytophthora genus TaqMan tested on a collection of Phytophthora and Pythium from pure cultures ........................................................97 Table 4. Results of the multiplex real-time PCR assay comprising three P. ramorum specific TaqMan probes and one Phytophthora genus TaqMan tested with field samples from a ring trial ...............................................................................................98 Table 5. Results of the multiplex real-time PCR assay comprising three P. ramorum specific TaqMan probes and one Phytophthora genus TaqMan. tested with field samples from ELISA lysate, survey 2006 CFIA ........................................................101 Table 6. Multiplex real-time PCR assay combining TaqMan probes targeting plant, oomycete, Phytophthora and P. ramorum tested with samples from different hosts infected by various species of Phytophthora..............................................................103 Supplemental Table 1. Alignment of the internal transcribed spacer 1, 5.8S ribosomal RNA gene, and internal transcribed spacer 2 sequences with oomycetes, ascomycetes and basidiomycetes............................................................................................................113
xv Supplemental Table 2. Alignment of RuBisCO (ribulose 1,5-bisphosphate carboxylase) sequences with different kind of the chloroplast plant species...................................115 Chapitre III Table 1. Isolates of Phytophthora ramorum from different culture collections used in this study............................................................................................................................135 Table 2. Primers used for PCR assays targeting Phytophthora spp. and P. ramorum. ......137 Table 3. ASO genotyping of β-tubulin gene in a selection of Phytophthora ramorum from different geographic origins and mating types. ..........................................................138 Table 4. ASO genotyping of CBEL gene in a selection of Phytophthora ramorum from different geographic origins and mating types. ..........................................................140 Table 5. Allele-specific oligonucleotide genotyping of field samples (nurseries). ............142 Table 5 (suite). Allele-specific oligonucleotide genotyping of field samples (nurseries). .145 Chapitre IV Table 1. Isolates of Phytophthora ramorum from different culture collections used in this study with genotypes bases on SNP profiles. .............................................................175 Table 2. Primers used for amplification and DNA sequencing ..........................................178 Table 3. Characteristics of Phytopthora ramorum genes sequenced..................................180 Table 4. Comparison of divergence in European and North American populations of P. ramorum......................................................................................................................181 Table 5. Polymorphism comparison between North American and European populations of P. ramorum .................................................................................................................182 Table 6: Multilocus genotypes of P. ramorum from Europe and North America at 13 loci ....................................................................................................................................183 Table 7: Genotype frequencies and heterozygosis at 13 different loci in and between European and North-American population.................................................................184 Supplement material 1. Isolates of Phytophthora ramorum used in this study with the Genbank accession number. ……………………………………………………..197 Annexe I Tableau annexe 1a ..............................................................................................................212 Tableau annexe 1b ..............................................................................................................221
xvi
Liste des figures Introduction Figure 1. Distribution du Phytophthora ramorum en Californie et en Oregon le 15 février 2008. .............................................................................................................................20 Figure 2 (a et b). Aperçu des dommages causés par l’encre des chênes rouges dans les collines du « China Camp State Park », en Californie. Les arbres en gris sont morts ou atteints par la maladie. ..................................................................................................21 Figure 3. Symptômes de la maladie de l’encre des chênes rouges .......................................23 Figure 4. Localisation du P. ramorum en Europe.................................................................24 Figure 5. Localisation du P. ramorum en Amérique du Nord..............................................25 Figure 6. Structures sexuées et asexuées chez Phytophthora ..............................................26 Figure 7. Arbre phylogénétique du monde du vivant basé sur des séquences d’ADN ribosomiaux. .................................................................................................................27 Figure 8. Deux exemples de calculs de la volatilité des codons...........................................19 Chapitre I Figure 1. Real-time PCR of Phytophthora species assayed with TaqMan (■), molecular beacons (●) and SYBR® Green (▲), targeting the β-tubulin gene. ............................65 Figure 2. Standard curves based on dilution of Phytophthora ramorum DNA for TaqMan assays targeting ITS, β-tubulin, and elicitin. ................................................................66 Chapitre II Figure 1. Standard curves based on dilution of Phytophthora ramorum DNA for TaqMan assays targeting P. ramorum on ITS, β-tubulin, and elicitin and Phytophthora species on β-tubulin.................................................................................................................106 Figure 2. Standard curves shape based on dilution of Phytophthora ramorum DNA for multiplex and simplex TaqMan assays targeting P. ramorum. ..................................107 Chapitre III Figure 1. SNPs observed on the chromatogram sequence of β-tubulin and CBEL:...........148 Figure 2. Allele-specific oligonucleotide (ASO) genotyping using real-time PCR amplification with SYBR® Green on SNP at position 279 of β-tubulin....................149 Chapitre IV Figure 1. Distribution of volatility of 10252 genes of P. ramorum....................................185 Figure 2. SSCP profiles of eight P. ramorum isolates, four from Europe and four from North America, with different migration profiles.......................................................186
xvii Figure 3. Agarose gel electrophoresis of the PCR amplicon of the gene 83987 with primers 83987-87U and 83987-594L on 1.5% agarose gel, TAE buffer 1X, stained with ethidium bromide........................................................................................................187 Figure 4. Unweighted pair group dendrogram presenting the different genotypes. ...........188 Annexe II Figure 1 (Annexe II). Cycle de vie probable de Phytophthora ramorum...........................241
1
Introduction générale
1
Introduction générale Les organismes pathogènes des plantes du genre Phytophthora ont depuis longtemps occupé une place de choix dans les priorités de recherche des phytopathologistes. Au XIXe siècle, la famine irlandaise, causée par la destruction des cultures de pommes de terre par le Phytophthora infestans (Mont.) de Bary, a provoqué de grands changements économiques et sociologiques dans le pays ainsi que l’émigration de près d’un quart de la population vers les États-Unis d’Amérique. Le Phytophthora cinnamomi Rands a d’autre part causé de sérieux problèmes dans les forêts australiennes. Une épidémie causée par cet agent pathogène, qui peut s’attaquer à plus de neuf cents plantes hôtes, a détruit une grande partie du couvert forestier et des arbustes dans les forêts d’eucalyptus (Eucalyptus marginata Sm.) de l’ouest du pays (Shearer et Tippett, 1989). Étant donné l’importance de l’industrie du bois d’eucalyptus, cette épidémie a eu des effets néfastes non seulement sur les écosystèmes, mais aussi sur ce secteur de l’économie (Stukely et al., 2007).
Plus récemment, le Phytophthora ramorum Werres, a causé « la mort subite du chêne » ou « l’encre des chênes rouges ». Cette maladie affecte principalement les espèces du groupe des chênes rouges ainsi que le lithocarpe de Californie (tanoak) sur la côte ouest californienne depuis les années 1990. Également problématique en pépinière, ce microorganisme affecte, tout comme le P. cinnamomi, une grande variété de plantes hôtes et cause de sérieux soucis aux exportateurs et importateurs de végétaux potentiellement sensibles à cet agent pathogène. Depuis sa découverte, plusieurs interceptions de cet organisme, qui fait l’objet de mesures de quarantaine, ont été rapportées en pépinière chez différentes plantes et dans plusieurs pays. Une revue des connaissances et des concepts de cette maladie et des méthodes d’identification et de détections sera discutée dans cette introduction afin d’exposer les problématiques présentées dans cette thèse.
2
1. La maladie
1.1
La découverte Depuis 1994-1995, on dénombre la mortalité de milliers de lithocarpes et de chênes
dans le comté Marin de la Californie (Fig. 1) (McPherson et al., 2001; Svihra, 1999), particulièrement le Lithocarpus densiflorus (Hook.et Arn.) Rehder et le Quercus agrifolia Née, un représentant appartenant à la famille des chênes rouges. Cette maladie appelée « encre des chênes rouges » ou « mort subite du chêne », traduction de l’anglais sudden oak death ou SOD, a décimé environ 80% des chênes de la Californie depuis son apparition (Hansard, 2003). Sujet d’actualité dans plusieurs journaux comme le New York Times, cette maladie fut classée à la 43e position des 100 meilleures histoires scientifiques en 2002 1. Elle cause des chancres saignants ou suintants sur le tronc des chênes et finit par mener à la mort de l’arbre (Fig. 2). Le nom « mort subite du chêne » fut donné par différents média américains après qu’on eut rapporté que la cime des arbres affectés semblait mourir très rapidement, la couleur du feuillage tournant d’un vert vigoureux à brun dans une période de quelques semaines (Rizzo et al., 2002; Svihra, 1999). Ce n’est qu’en 2000 qu’on réalise que l’organisme en cause appartient au genre Phytophthora. Il est nommé officiellement en 2001 Phytophthora ramorum (Werres et al., 2001).
1.2
Les hôtes du P. ramorum Jusqu’à maintenant, en faisant le décompte à partir de sites Web européens,
américains et canadiens, on dénombre plus de 120 espèces végétales, réparties dans quelque 70 genres et 35 familles, sensibles à cet agent pathogène. En plus des chênes rouges (Lithocarpus densiflorus, Quercus agrifolia, Q. kellogii Newberry), on retrouve le séquoia (Sequoia sempervirens (D.Don.)Endl.), le sapin de Douglas (Pseudotsuga menziesii var. 1
DISCOVER Vol.24 No.1 (Janvier 2003)
3 menziesii (Mirb.) Franko.), deux essences commerciales très importantes, ainsi que plusieurs arbres et arbustes ornementaux parmi les hôtes potentiels du P. ramorum. Une liste répertoriant les hôtes naturels et possibles de P. ramorum a été dressée dans le PRA (« Pest Risk Assesment ») de l’Agence canadienne d’inspection des aliments (ACIA) en 2006 (Rioux, et al., 2006) (Cf. Annexe 1). Quant au chêne des marais (Quercus palustris Münchh) et au chêne rouge (Q. rubra L.), que l’on retrouve plus au nord-est des États-Unis et également au Québec (spécifiquement le Q. rubra), ils ont été infectés artificiellement et ont présenté les mêmes symptômes que les chênes atteints en Californie, ce qui laisse croire à un risque potentiel de maladie chez ces hôtes (Tooley et al., 2004). Plus précisément au Québec, une étude de la sensibilité de différentes espèces d’arbres de l’est du Canada est en cours (Simard et al., 2007).
Les symptômes sont assez variables d’un hôte à l’autre. Sur les chênes de la Californie, les principaux symptômes se caractérisent par l’apparition de chancres accompagnée d’un saignement brun foncé à noir (Fig. 3a, b) qui peut survenir exceptionnellement jusqu’à une hauteur de 18,3 m (60 pieds) dans les arbres, ainsi que par le brunissement (Fig. 3c) et la perte des feuilles en plein été. Pour les autres hôtes, on peut retrouver des taches ou nécroses sur les feuilles (Fig 3d), les aiguilles ou les tiges des plantes infectées (Garbelotto et al., 2002a). La dissémination des spores se fait majoritairement par le vent et les éclaboussures lors des pluies ou mouvements dans les flaques d’eau sur les hôtes environnants. Les symptômes varient énormément dépendamment du type d’hôte (le dépérissement, la flétrissure et des lésions) ce qui rend parfois le diagnostic de la maladie très difficile. De plus, sur certaines plantes, l’infection semble parfois négligeable bien que ces plantes puissent toutefois servir de vecteur (réservoir d’inoculum) pour l’inoculation sur un hôte adjacent, un bon exemple étant le laurier de Californie (Umbellularia californica (Hook et Arn.) Nutt.) (Davidson et al., 2002). De plus, ils peuvent contribuer grandement à la propagation de l'agent pathogène, par les mouvements naturels dans l'environnement et par le commerce des végétaux provenant de pépinières infestées dont l’infection serait passée inaperçue.
4
1.3
Localisation Jusqu’à présent, l’agent pathogène est retrouvé en pépinière en Europe (Fig. 4) dans
les pays suivants : Allemagne, Belgique, Danemark, Espagne, Finlande, France, Irlande, Italie, Norvège, Pays-Bas, Pologne, Royaume-Uni, Slovénie, Suède, République Tchèque et Suisse (Brasier, 2004; De Merlier, 2003; Garbelotto et al, 2002a; Hansen et al., 2003; Herrero et al., 2006; Husson et al., 2007; Lane et al., 2003; Lane et al., 2004; Lilja et al., 2007; Moralejo et Werres, 2002; Werres, 2002; Werres et al., 2001). En Amérique du Nord (Fig. 5), la maladie est aussi présente en pépinière au Canada ainsi qu’aux États-Unis (dans 23 États américains en 2004 (Cave et al., 2007)). En décembre 2007, on répertoriait un total de 21 pépinières situées dans 6 États américains dans lesquelles le P. ramorum avait pu être détecté 2. Au Canada, le P. ramorum a seulement été retrouvé en pépinière en ColombieBritannique près de Vancouver depuis 2003 (CFIA, 2003). Il est resté confiné à ces endroits depuis ce temps là.
Il a été également retrouvé en milieu naturel en Californie et en Oregon ainsi qu’au Royaume-Uni et aux Pays-Bas, où il s’attaque à certains chênes ainsi qu’à d’autres espèces [par exemple le hêtre (Fagus sylvatica L.) et le châtaignier (Castanea sativa Mill.)] depuis janvier 2004 (DEFRA, 2004a; DEFRA, 2004b). Le RAPRA (« Risk Analysis for Phytophthora ramorum ») estime que le P. ramorum a été retrouvé hors pépinière (outdoor) dans 7 pays (RAPRA, 2008).
2
www.suddenaokdeath.org ou http://nature.berkeley.edu/comtf/html/chronology1.html
5
1.4
Impact économique L’impact économique du P. ramorum n’est pas facile à déterminer, mais semble être
majeur. Les conséquences pour l’industrie forestière pourraient être considérables si les forêts canadiennes venaient à être infectées par cette maladie comme c’est le cas en Californie et en Oregon. Si l’exportation de bois de sciage venait à être réduite, des sommes importantes seraient en jeu puisque le Canada en exporte annuellement pour plus de 20 milliards de dollars 3.
Ce sont les impacts indirects envers les pépinières et les entreprises paysagistes dans les zones infectées ou à risque qui sont pour l’instant les plus problématiques. Ainsi, le Canada a fermé son marché à la plupart des cultures végétales des États de l'Oregon et de la Californie en 2001. Si l'accès au marché n'avait pas été rétabli, les pépinières de l'Oregon auraient à elles seules perdu des exportations au Canada d'une valeur de 15 à 20 millions de dollars. En Colombie-Britannique, la valeur estimée du secteur des pépinières et de la floriculture est de 500 millions de dollars et les exportations vers les États-Unis sont estimées à 170 millions de dollars. De plus, si l’exportation de conifères vers l’Asie venait à être interrompue, c’est une industrie de 150 millions de dollars qui serait affectée 4. Depuis 2003 et jusqu’en 2006, les coûts en Colombie-Britannique ont été estimés à plus de 3 millions de dollars pour la perte de plants et les ventes perdues et à environ 500 000 $ pour le nettoyage et l’élimination des zones infestées (Kristjansson et Miller, 2008). Seulement en 2007, une enquête à travers différentes pépinières du Canada a permis de tester plus de 46 000 échantillons ce qui a conduit à la détection de 164 échantillons positifs en Colombie-Britannique entre le 1er avril et le 14 décembre 2007 (communication personnelle Stéphan Brière, ACIA, Ottawa). 3
Statistique Canada, CANSIM, Exportations de biens sur la base de la balance des paiements, selon le produit.2006, 228-0003 ; and BC STATS (exportation by province, 1997-2006, Exports to all countries 2006)
4
La Gazette du Canada, Vol. 141, no 14, 7 avril 2007, Règlement sur l'indemnisation relative au Phytophthora ramorum, http://canadagazette.gc.ca/partI/2007/20070407/html/regle1-f.html
6
1.5
Mesures Des mesures de quarantaine ont été mises en place dans plusieurs pays pour
éviter une propagation et/ou une introduction de l’agent pathogène. Pour déterminer la présence ou l’absence de cet organisme dans la gamme d’hôtes qui ne cesse de s’élargir, des méthodes de détection doivent être mises en place. Lorsqu'il faut confirmer que les arbres d’un secteur donné sont atteints par la maladie, il est très important de s’assurer que l’agent pathogène en cause est vraiment celui recherché. Pour cela on utilise différentes méthodes de détection comme la mise en culture de l’agent pathogène sur des milieux spécifiques ou des méthodes moléculaires ciblant l’ADN de l’organisme (Cf. introduction section 3). De tels outils permettent alors de mieux contrôler les plants et les zones infestées et ainsi éviter la propagation.
Lors des mesures de quarantaine, un protocole d’éradication 5 est enclenché. Tous les plants de la pépinière infectée qui sont des hôtes connus du P. ramorum sont mis en quarantaine. Une zone d’échantillonnage est déterminée selon la nature des hôtes et ces derniers se retrouvant dans les zones infestées (block) sont alors détruits (par brûlage et par fumigations (sols)). Des échantillons d’eau et de sol sont ensuite prélevés pour analyse de la présence du P. ramorum (infestation). Les zones sélectionnées mais non détruites sont mises en quarantaines pour 90 jours. Les plants sont ensuite ré-inspectés et testés de nouveau pour confirmer que le P. ramorum n’est plus présent. Si c’est le cas, les pépinières sont libérées, sinon une autre ronde d’éradication et quarantaine est amorcée. Les pépinières libérées sont toujours surveillées, mais ne sont plus sous quarantaine.
5
Official Regulatory Protocol for Wholesale and Production Nurseries Containing Plants Infected with Phytophthora ramorum, Confirmed Nursery Protocol: Version 8.0, Revised: July 20, 2007; United States Department of Agriculture (USDA), Animal Plant Health Inspection Service (APHIS), Plant Protection and Quarantine (PPQ), Center for Plant Health Science and Technology (CPHST), Emergency and Domestic Programs (EDP), Eastern Region (ER), Western Region (WR) http://www.aphis.usda.gov/plant_health/plant_pest_info/pram/protocols.shtml
7
2. Biologie
2.1 Phytophthora Le nom Phytophthora dérivant du grec phyto (plante) et phthora (destructeur), caractérise bien ce genre car les différentes espèces qu’il regroupe s’attaquent uniquement aux plantes et causent beaucoup de dégâts tant dans les milieux agricoles que forestiers (Erwin et Ribeiro, 1996). C’est d’ailleurs l’apparition d’une maladie causée par le P. infestans (Mont.) De Bary, entraînant une importante famine en Irlande au milieu du XIXe siècle, qui marqua la naissance de la phytopathologie (Large, 1940), et ce 20 ans avant que Louis Pasteur étudie la théorie des germes comme agents infectieux (Cooke, 2007). Depuis plus de 150 ans, les scientifiques étudient avec intérêt les organismes appartenant à ce genre, sans cesser de découvrir de nouvelles espèces : à ce jour, plus de 80 espèces ont été identifiées. Les Phytophthora sont les agents pathogènes les plus dévastateurs chez les dicotylédones (Erwin et Ribeiro, 1996). Ils font partie des Oomycètes car ils produisent une spore sexuée appelée oospore (Fig 6a), qui résulte de la fusion entre les deux gamètes [oogone (femelle) et anthéridie (mâle)]. Des structures asexuées sont aussi présentes chez cet organisme, telles les chlamydospores (Fig 6b) et les zoospores (Fig. 6c). La chlamydospore possède une paroi épaisse qui permet de résister longtemps à des conditions difficiles (c’est donc une structure de survie). Les zoospores sont des spores asexuées et biflagellées, qui nagent dans des milieux aqueux (Moore-Landeker, 1982) et sont attirées par les champs électromagnétiques (électrotaxi) des racines végétales. On retrouve chez le genre Phytophthora des espèces homothalliques et hétérothalliques. Autrefois, les Oomycètes étaient classés dans le règne des Fungi. Il font maintenant partie des Straménopiles qui se rapprochent beaucoup plus, par la présence de spores motiles à deux flagelles inégaux et par phylogénie de l’ADN ribosomal, aux diatomées et aux algues brunes (Baldauf et al., 2000; Cavalier-Smith et al., 1994; Van der Auwera et De Wachter, 1997) (Fig. 7). Les Straménopiles, (Stramenopila : « organismes plumeux », ou hétérokontes (Heterokonta : « organismes à deux flagelles différents ») (Fig. 6c) constituent
8 un sous-règne des Eucaryotes et sont parfois classés dans un règne supérieur, les Chromistes (Cavalier-Smith, 1998; Cavalier-Smith et al., 1994). Les Straménopiles auraient donc perdu leur capacité à faire la photosynthèse. En taxinomie, on peut classer le P. ramorum comme étant : Eukaryota; Stramenopiles; Oomycetes; Peronosporales; Phytophthora.
On les classait autrefois, parmi les Fungi, car ils présentent des similitudes avec ce règne notamment la formation d’un thalle mycélien. Par contre, les Oomycètes ne possèdent pas ou très peu de chitine dans leur paroi alors que l’on retrouve de bonnes quantités de cellulose et de β-glucanes dans celles-ci. Finalement, ils sont généralement diploïdes, dans leur phase végétative contrairement aux champignons vrais comme les Ascomycètes ou aux Basidiomycètes qui sont généralement haploïdes (Rossman, 2006.).
2.2 Phytophthora ramorum C’est en 2001 que le nom Phytophthora ramorum (Werres et al, 2001) est donné à l’agent pathogène infectant les branches des rhododendrons et des viornes (Viburnum spp.). Le nom d’espèce ramorum vient du mot ramus qui signifie en latin « branche » et fait référence à la prédilection de cette espèce à s’attaquer aux branches et aux pousses de ses hôtes.
2.2.1 Caractères morphologiques du P. ramorum
Chez le P. ramorum, l’oogone et l’anthéridie ne sont pas produites en culture simple, ce qui suggère qu’il serait autostérile et hétérothallique. Actuellement, comme la reproduction sexuée est peu probable, le cycle d’infection de P. ramorum (C.F. Annexe 2) est possible grâce à la reproduction asexuée. Les différentes populations échantillonnées ne semblent pas, pour l’instant, indiquer de signes de recombinaison. Le P. ramorum peut alors produire des oospores (reproduction sexuée) seulement lorsque les deux types de
9 compatibilité sexuelle (mating type) (A1 et A2) sont présents. Les deux types sexuels du P. ramorum ne se retrouvent pas souvent en nature dans un même territoire. La production d’oospores peut aussi survenir lorsque des espèces différentes, de types sexuels opposés (de sexes différents), croissent ensemble sur un même milieu (Erwin et Ribeiro, 1996). Par exemple, lorsqu’on met en culture le Phytophthora cryptogea Pethybridge et Lafferty, ou alors le Phytophthora cambivora (Petri) Buisman, avec un type sexuel différent du P. ramorum, l’obtention des structures sexuées est possible (Werres et al., 2001). Des études de croisements (Brasier et Kirk, 2004; Werres, et Zielke, 2003) ont montré que les populations du P. ramorum provenant d’Europe (EU) présentaient un type sexuel A1 et que celles provenant de l’Amérique du Nord (NA) étaient du type A2 (Brasier, 2003; Werres, Zielke, 2003). Ainsi, initialement, le type sexuel semble être lié à l’origine géographique. Cependant, des études récentes rapportent pour la première fois la présence d’individus de type sexuel A2 en Europe (Belgique). Ce type sexuel semblait jusque là être exclusivement lié aux populations d’Amérique du Nord (Werres et De Merlier, 2003). De même, le type A1 a été retrouvé récemment en Amérique du Nord, plus précisément au Canada et dans les états de Washington et de l’Oregon aux États-Unis (Hansen et al., 2003). Cette nouvelle découverte est importante car des phénomènes de croisements et de recombinaisons entre les deux types sexuels pourraient avoir lieu. Une nouvelle descendance présentant des variations de profils génétiques et de la pathogénicité pourrait changer dramatiquement l’épidémiologie de cet agent pathogène (Brasier, 2003; Erwin et Ribeiro, 1996). Pour s’en convaincre, on peut penser au P. infestans qui, de cette façon, a engendré une plus grande diversité génétique et une augmentation de sa virulence et de sa résistance aux fongicides (Drenth et Goodwin, 1999). Quand la formation d’oogones se produit chez le P. ramorum, elles sont, la plupart du temps, en position terminale et quelquefois latéralement sessiles, d’un diamètre variant de 24 - 40 µm (moyenne de 29.8 - 33µm). Les oospores plérotiques sont de 20 à 36 µm de diamètre.
L’anthéridie est de forme arrondie et sa taille
approximative est de 12-22 X 15-18 µm (Werres et al., 2001).
10
2.2.2 La pathogénicité et variabilité morphologique et moléculaire du P. ramorum
La pathogénicité des isolats du P. ramorum provenant d’Amérique du Nord et d’Europe a été étudiée et comparée (Brasier et al., 2003; Pogoda et Werres, 2002). L’analyse des empreintes d’ADN (AFLP) et microsatellites des populations de P. ramorum en Amérique du Nord et en Europe semble démontrer des profils assez différents (Garbelotto et al., 2002; Ivors et al., 2006; Ivors et al., 2004). Cependant, ces analyses indiquent que celles-ci seraient plus variables en Europe qu’en Amérique du Nord et ne permettent pas de cerner la source potentielle de l’épidémie. Par contre, les populations nord-américaines semblent démontrer une plus grande variabilité dans leur morphologie (Werres et Kaminski, 2005). Cela peut suggérer que les populations nord-américaines et européennes sont différentes sur le plan adaptatif (Brasier, 2003). Néanmoins, une étude de prédiction des origines et risques de l’encre des chênes rouges (Kluza et al., 2007) complémente les études réalisées précédemment (Guo et al., 2007; Guo et al., 2005; Meentemeyer et al., 2004). Cette étude basée sur des modèles de niches écologiques, tenant compte des climats (les niveaux de la température et de l’humidité sont très important pour l’établissement de la maladie), des topographies, des types d’hôtes (surtout d’origine asiatique), permet aux auteurs de prédire que l’origine du P. ramorum pourrait se situer dans l’est de l’Asie.
2.2.3 Le génome du P. ramorum
En avril 2004, le JGI (Joint Genome Institute) rendait disponible la séquence complète
du
génome
du
P.
ramorum
(http://genome.jgi-
psf.org/Phyra1_1/Phyra1_1.home.html) (Tyler et al., 2006). Dans la version 1.0 de 2004, les séquences de ce génome ont pu être obtenues grâce à la technique de séquençage à haut débit « shotgun » et l’assemblage a été réalisé en utilisant le logiciel JGI assembler dénommé « Jazz » pour une couverture de sept fois le génome. Après nettoyage des séquences vectrices et de faibles qualités, plus de 1 million de lectures ont été assemblées
11 en 2576 structures (scaffolds), pour une couverture totalisant 66,6 Mpb. La moitié de la séquence du génome brut de P. ramorum est contenue dans 63 structures d’au moins 308 Kpb de longueur. Ces assemblages ont été annotés en utilisant la plateforme d’annotation de JGI. Des modèles de gènes, les prédictions associant différents transcrits et/ou protéines ainsi que la création de cartes, ont été réalisés avec des méthodes utilisant les ADNc et des homologies de protéines. De plus, le génome du Phytophthora sojae (86 Mpb), séquencé en 2004, http://genome.jgi-psf.org/Physo1_1/Physo1_1.info.html, a montré une certaine conservation avec celui du P. ramorum apportant un support additionnel à son annotation. Les fonctions des gènes ont été assignées automatiquement basées sur les homologies des gènes connus. En 2006, l’annotation version v1.1, a pu déterminer que le génome contenait 15743 modèles de gènes. L’annotation est toujours en cours, et une partie est disponible sur la base de données du NCBI (GenBank) depuis la fin 2006. Depuis, le génome du Phytophthora infestans (190 Mpb) est disponible via le site internet du Broad Institute (http://www.broad.mit.edu/annotation/genome/phytophthora_infestans). Le génome du P. capsici sera également disponible sous peu après utilisation d’un mélange des technique de séquençage
Sanger
et
454
(454
Life
Sciences,
Branford,
(http://www.jgi.doe.gov/sequencing/why/CSP2006/Pcapsici.html).
CT) Celui
par d’un
le
JGI autre
Oomycète, le Pythium ultimum, est en cours de séquençage par le consortium TIGR (http://cpgr.tigr.org).
12
3. Identification et détection du P. ramorum
3.1 L’isolement par culture Pour confirmer que des plants sont infestés par la maladie il est très important de s’assurer que l’organisme en cause est vraiment celui recherché. L’approche traditionnelle (classique) pour identifier cet agent pathogène est sa mise en culture (Garbelotto et al., 2001; Werres et al., 2001a). Il s’agit de faire croître l’agent pathogène sur des milieux sélectifs, comme le PARP-V8 (Jeffers et Martin, 1986) permettant sa croissance et l’identification par ses traits morphologiques caractéristiques.
Boite 1.0
Milieu de culture PARP
Le PARP (pimaricine-ampicilline-rifampicine-pentachloronitrobenzène) est un milieu sélectif pour les oomycètes. La pimaricine et le pentachloronitrobenzène (PCNB) (Erwin et Ribeiro, 1996; Jeffers et Martin, 1986) sont des fongicides et l’ampicilline et la rifampicine sont des antibiotiques contre les bactéries. On rajoute ces composés généralement à des milieux V8, CMA (corn meal agar) ou
PDA (potato dextrose agar). Les Phytophthora et les Pythium y poussent très bien mais l’ajout de hémexazol permet généralement d’inhiber la croissance des Pythium.
3.2 Les méthodes moléculaires Dépendamment de la saison, du type d’hôte, de l’utilisation de fongicides, la culture des organismes pathogènes à partir des plantes n’est pas toujours facile. Elle demande beaucoup de temps, une bonne expertise et les risques de contamination sont élevés. Ces
13 raisons expliquent pourquoi des méthodes moléculaires ont été développées au cours des dernières années (Martin et al., 2000). Des méthodes de détection à partir d’ADN sont possibles grâce à la réaction de polymérisation en chaîne (PCR) qui permet d’amplifier des copies multiples d’une portion ciblée de l’ADN. Cette technique offre la possibilité d’être utilisée directement sur les tissus de l’hôte pour y détecter la présence de l’agent pathogène. De plus, ces méthodes permettent d’obtenir des résultats en quelques heures, contrairement à plusieurs jours pour la croissance sur milieux de culture (Garbelotto et al., 2001; Levesque, 1997).
3.2.1 L’ELISA
La première méthode utilisée dans le cas du P. ramorum est l’immunodosage. La technique ELISA ("enzyme-linked immunosorbent assay") (Gaastra, 1984) fait appel aux anticorps spécifiques aux protéines du Phytophthora. Ces anticorps sont couplés à des marqueurs chémiluminescents. En raison de leur très grande spécificité, les anticorps sont souvent utilisés pour détecter et mesurer une protéine spécifique. L’intensité lumineuse de l’enzyme peut être évaluée en fonction de l’absorbance. Cette méthode peu coûteuse et rapide est utilisée dans le cas du P. ramorum. Utilisée en première analyse, un résultat positif avec ce procédé permet de confirmer la présence d’un Phytophthora dans l’échantillon visé. L’ACIA (l’Agence canadienne d’inspection des aliments) utilise plus spécifiquement
l’enzyme
alcaline
phosphatase
conjuguée
à
l’anticorps
« anti-
Phytophthora » (Phytophthora Reagent Set Instructions, numéro de catalogue SRA 92600, Agdia Inc. Elkhart, IN). Il est à noter que cette méthode présente certaines limites. Elle détecte seulement le genre Phytophthora et non l’espèce P. ramorum. De plus, elle donne parfois des faux négatifs (Osterbauer et Trippe, 2005) selon le seuil de détection fixé par les expérimentateurs (Stéphan Brière communication personnelle). Une fois des résultats positifs obtenus (pour éliminer ainsi, le plus possible d’échantillons qui seraient négatifs pour la présence de Phytophthora), des tests faisant appel à la technique de PCR spécifique au P. ramorum sont utilisés sur ces échantillons pour confirmer la présence de l’agent pathogène.
14
3.2.2 La technique de PCR
Un protocole d’amplification d’une région de l’ADN ribosomal ITS (« Internal Transcribed Spacer ») a été développé par PCR imbriquée (« nested ») au début de la découverte de l’agent pathogène (Davidson et al., 2003). La PCR imbriquée consiste à réamplifier un amplicon. L’avantage de cette technique est qu’elle permet un meilleur rendement d’amplification et offre une plus grande sensibilité. Elle est toutefois loin d’être parfaite. Les risques de contamination par de l’ADN (amplicons) sont très grands, ce qui demande de manipuler avec la plus grande précaution. De plus, l’utilisation de la région ITS de l’ADN ribosomal nucléaire, ne différencie pas toujours très bien le P. ramorum du P. lateralis Tucker et Milbrath, l’espèce de Phytophthora qui montre le plus de ressemblance phylogénétiquement (Garbelotto et al., 2002) et qui est aussi retrouvée sur la côte ouest américaine. Il a donc fallu trouver d’autres régions ou gènes présentant plusieurs sites polymorphes ou SNP (« Single nucleotide polymorphism ») entre ces deux espèces et les autres espèces appartenant au genre Phytophthora. De récentes phylogénies basées sur des régions de l’ADN mitochondrial (la Cytochrome oxydase II) (Martin et Tooley, 2003) et nucléaire (la β-tubuline et le facteur d’élongation Iα ) (Kroon et al., 2004) ont permis de concevoir des amplifications PCR pour la détection de Phytophthora, incluant le P. ramorum (Ioos et al., 2006; Martin et Tooley, 2004; Martin et al., 2004).
3.2.3 La technique de PCR en temps réel
La PCR en temps réel (Real Time-PCR) ou quantitative (qPCR), une autre technique de biologie moléculaire a été développée ces dernières années afin de faire la distinction entre des polymorphismes (variations de séquences) et de quantifier des molécules (Schaad, Frederick, 2002). Elle utilise le principe d’hybridation d’un allèle spécifique (ASH) (Jenkins, 2001), par la détection d’acides nucléiques par SYBRGreen ou par des sondes fluorescentes internes. Le SYBRGreen est une molécule fluorescente, agent intercalant non spécifique, qui lorsque l’ADN est double brin, s’insère entre ceux-ci. Il est donc possible de
15 lire la fluorescence durant l’extension des brins d’ADN par l’ADN polymérase à chaque cycle (Giglio et al., 2003). Cependant, cette molécule est non spécifique et toutes fausses amplifications ou dimérisations d’amorces peuvent produire un signal. Néanmoins, des courbes de dissociation sont disponibles afin de vérifier si d’autres produits d’amplification ont été formés. Le phare moléculaire (« molecular beacon ») (Marras et al., 2003; Tyagi et Kramer, 1996; Tyagi et al., 2000) et le TaqMan (Holland et al., 1991; Lee et al., 1993; Livak, 2003) sont des sondes internes d’acides nucléiques qui s’hybrident avec l’ADNcible (séquence) lorsqu’elles sont complémentaires, émettant alors une fluorescence. La lecture du signal émis par le phare moléculaire se fait durant l’appariement des amorces et la molécule n’est pas dégradée. Le TaqMan quant à lui se lit durant l’élongation du brin, dégradant ainsi la molécule par l’activité exonucléase de la polymérase et conduisant à l’émission d’une fluorescence en séparant le fluorochrome de l’absorbeur (quencher). Ces techniques peuvent donc s’avérer très utiles pour une détection rapide des agents pathogènes en autant que des régions polymorphiques soient connues (Böhm et al., 1999; Schaad et Frederick, 2002; Weller et al., 2000). Dans la détection du P. ramorum, certaines variations de la technique de la PCR en temps réel ont été développées au cours de ces dernières années (Hayden et al., 2006; Hughes et al., 2006a; Hughes et al., 2006b; Schena et al., 2006; Tomlinson et al., 2007; Tomlinson et al., 2005; Tooley et al., 2006).
3.2.4 Génotypage des populations du P. ramorum
Comme mentionné plus tôt à la section 2.2.2, les analyses des populations du P. ramorum au moyen de marqueurs AFLP et microsatellites semblent démontrer des profils européens et américains assez différents (Garbelotto et al., 2002; Ivors, 2006; Ivors et al., 2004). Les variations semblent plus fréquentes en Europe qu’en Amérique du Nord, ce qui suggère que les populations ont évolué en goulot d’étranglement. Les populations étant séparées, ont donc subi une réduction de la taille effective ainsi qu’une réduction importante de la diversité allélique. Ces analyses démontrent également qu’un autre génotype unique fut découvert en Amérique. D’autres études de populations ont été réalisées dernièrement utilisant les microsatellites en Californie et en Oregon (Cooke, 2007;
16 Prospero et al., 2004; Prospero et al., 2007). Dans un premier temps, ces études montraient que les populations de l’Oregon en forêt et en pépinière semblaient provenir d’origines indépendantes. De plus, les isolats prélevés en forêt provenaient principalement d’un seul génotype qui s’est diversifié avec le temps.
4.0 Problématique (objectifs et hypothèses de recherche) Les différents concepts qui viennent d’être évoqués permettent de réaliser la problématique et l’ampleur de la mort subite du chêne et de l’organisme qu’est le Phytophthora ramorum. Mon projet de doctorat commença en 2002, peu après l’identification de l’agent pathogène causant l’encre des chênes rouges comme étant le P. ramorum (Werres et al., 2001). À cette époque, seule la méthode de culture et la technique de PCR imbriquée (Garbelotto et al., 2002b) permettaient d’identifier l’organisme en cause. Cependant, comme discuté dans l’introduction, ces méthodes ont leurs limites et ne permettent pas toujours de bien distinguer les deux espèces très apparentées se retrouvant dans la même niche écologique. Le P. ramorum avait déjà été retrouvé en Californie et en Oregon, ainsi que dans quelques pays Européens. Comme la maladie n’était toujours pas rapportée au Canada, son introduction à la suite d’échanges commerciaux était crainte par les responsables de l’ACIA. La maladie a été détectée pour la première fois au Canada en mars 2003 et elle est toujours présente aujourd’hui dans certaines pépinières de la Colombie-Britannique. Si l’organisme venait à s’échapper et parvenait à s’établir dans des milieux naturels, les coûts financiers seraient considérables et persistants. Pour éviter ce risque, une demande fut faite par l’ACIA visant l’obtention d’un outil de détection qui permettrait de révéler rapidement et efficacement la présence de l’agent pathogène. Pour bien répondre à cette demande, il est de mise de considérer les deux hypothèses suivantes : 1- le P. ramorum présente des différences inter-spécifiques par rapport aux autres Phytophthora spp. et 2- il existe du polymorphisme intra-spécifique à l’intérieur même de l’espèce P. ramorum.
17 Les différents chapitres de cette thèse permettent de répondre aux différents objectifs que nous nous étions fixés. Tout d’abord, les deux premiers chapitres visent à développer des outils moléculaires pour identifier le P. ramorum et le différencier des autres Phytophthora. Le chapitre 1 traite de la conception d’un outil de diagnostic permettant de détecter rapidement et efficacement le P. ramorum. Comme le génome du P. ramorum n’était pas connu au début de ce doctorat (Cf. introduction 2.2.3), les régions disponibles sur Genbank concernant les Phytophthora (par exemple celles des ITS, βtubuline et élicitine) ont donc été utilisées pour la création d’amorces amplifiant ces régions du génome des Phytophthora spp. Par la suite, après séquençage des espèces se rapprochant le plus du P. ramorum, les séquences ont été alignées pour déterminer le polymorphisme du P. ramorum en comparaison avec les autres Phytophthora. Des amorces et sondes spécifiques ont ainsi été conçues.
Le deuxième chapitre est une suite logique du premier chapitre. Il partage le même objectif et utilise les mêmes régions génomiques. Par contre, ce chapitre décrit essentiellement la création d’un test de détection mais ici avec une réaction multiplexe, c’est-à-dire que toutes les réactions se réalisent dans le même tube. Les résultats obtenus démontrent que l’utilisation de plus d’une région du génome augmentait la fiabilité du test par rapport à un test utilisant seulement une région. Cette réaction multiplexe permet de gagner du temps mais également de réduire les coûts de détection, deux éléments non négligeables puisque des dizaines de milliers d’échantillons doivent être analysés. De plus, afin d’assurer une plus grande fiabilité du test, une autre sonde et des amorces spécifiques au genre Phytophthora sont incluses dans la réaction permettant d’avoir un contrôle supplémentaire.
Les chapitres 3 et 4 ont pour objectifs de découvrir des loci différenciant le polymorphisme intra-spécifique du P. ramorum, de mener des études visant à différencier les populations européennes et nord-américaines connues tout en développant des outils permettant de détecter ces polymorphismes. Dans le chapitre 3, comme la séquence du génome du P. ramorum n’était pas encore disponible, des différences entre les séquences des souches du P. ramorum provenant d’Europe et d’Amérique du Nord ont pu être mises
18 en évidence en utilisant les séquences du gène de la β-tubuline et de CBEL (« cellulose binding elicitor lectin »). Ces différences (SNPs) ont donc permis de distinguer les populations du P. ramorum entre les deux continents. La détermination de la provenance (Europe vs Amérique du nord) du P. ramorum permet de mieux connaître l’ampleur de cette épidémie et d’en prévenir la propagation. De plus, afin de détecter plus rapidement ces types, des amorces spécifiques ASO (« Allele Specific Oligonucleotide ») ont été développées pour génotyper les individus rapidement et sans séquençage. Cette méthode permet ainsi de génotyper des individus du P. ramorum à partir d’échantillons provenant de cultures pures mais aussi d’échantillons environnementaux, ce qui est particulièrement intéressant lors d’enquêtes en pépinière visant à mieux cibler le génotype et l’origine de l’agent pathogène.
Finalement, le quatrième chapitre vise au génotypage plus exhaustif des différentes populations de P. ramorum. Des études préliminaires avaient montré que plusieurs populations étaient distinctes à la suite d’analyses avec les AFLP ou microsatellites (Cf. introduction 2.2.2). Comme vu au chapitre précédent, deux gènes présentant du polymorphisme intraspécifique (SNPs) avaient été identifiés. Cependant, depuis 2004, la séquence complète du génome du P. ramorum est disponible avec plus de 16000 régions transcrites rendues publiques. Dans ce chapitre, nous avons tenté de répondre à plusieurs questions comme : Quels gènes portent des variations intra-spécifiques? Est-ce que ces polymorphismes occasionnent des changements dans la séquence de la protéine? Quelle est l’ampleur de ces changements et dans quelles populations? Pour déterminer le choix des gènes sélectionnés pour le génotypage (plus de 16000), nous avons utilisé le critère de volatilité du génome (Cf. Boite 2.0) (Plotkin et al., 2004). Cette volatilité a permis de sélectionner les gènes à étudier parmi les 16000 disponibles, ayant comme hypothèse que les gènes les plus volatiles présenteraient plus de polymorphismes que les moins volatiles. De plus, plusieurs nouveaux gènes présentant du polymorphisme intraspécifique ont alors pu être détectés et utilisés pour l’étude des populations du P. ramorum en Amérique du Nord et en Europe.
19
Boite 2.0
Qu’est-ce que la volatilité?
La volatilité est un concept utilisé pour comparer la sélection des gènes par rapport à son génome entier. Cette méthode suggère une façon de détecter rapidement une différence de pression sélective sur les gènes par inspection des séquences du génome, pour une empreinte des substitutions non synonymes. Si une séquence nucléotidique pour une région codant une protéine subit un nombre excessif de substitutions d’acides aminés, la région va en moyenne contenir une surabondance de codons volatiles comparés avec le génome entier. Pour chacun des 61 codons, on définit la volatilité par une proportion de ses points de mutations voisins qui encodent différents acides aminés. La volatilité de ce codon sera donc utilisée pour quantifier la probabilité que la plus récente mutation sur ce codon cause une substitution d’acide aminé (Fig 8). Ainsi, la probabilité pour chaque codon est calculée pour chaque séquence du génome. Cette probabilité P-value permet donc de classer les différentes séquences selon leur probabilité de volatilité, soit les plus et les moins volatiles.
Figure 8 : Deux exemples de calculs de la volatilité des codons. La volatilité de chaque codon dépend de la structure du code génétique. Le codon CGA, codant pour l’arginine, possède huit codons ancestraux potentiels, ainsi qu’un codon-stop, qui diffère de CGA par un seul point de mutation. Quatre de ces codons ancêtres de CGA encodent pour des acides aminés autres que l’arginine (R). Ainsi, la volatilité de CGA est de 4/8. Le codon AGA aussi encode pour l’arginine mais avec une volatilité de 6/8. Il y a 22 codons qui ont au moins un synonyme avec une volatilité différente. La volatilité est utilisée pour quantifier la probabilité que la plus récente mutation acceptée à un site cause un changement d’acide aminé.
Les commentaires et la figure sont adaptés de (Plotkin et al., 2004).
20
Figure 1. Distribution du Phytophthora ramorum en Californie et en Oregon le 15 février 2008. Les différents comtés infectés sont représentés. Le comté de Marin est identifié par une flèche hachurée. Les données provenant de pépinières ne sont pas illustrées ici. Les données proviennent des pathologistes du CDFA et de l’UC Davis/UC Berkeley (adapté d’une carte produite par le UCB GIIF : http://kellylab.berkeley.edu/SODmonitoring/).
21
a
b
Figure 2 (a et b). Aperçu des dommages causés par l’encre des chênes rouges dans les collines du « China Camp State Park », en Californie. Les arbres en gris sont morts ou atteints par la maladie. Photos provenant de Shane Sela, ACIA.
22
a
b
23 c
d
Figure 3. Symptômes de la maladie de l’encre des chênes rouges : a) chancre rouge suintant sur le lithocarpe de Californie (tanoak); b) l’écorce est prélevée sur ce chancre et on observe la marge de progression de l’agent pathogène; c) brunissement des feuilles d’un lithocarpe de Californie malade et d) exemple de tâches foliaires que l’on peut observer sur les autres hôtes que le chêne, ici un rhododendron. Photos a et c, provenant de Guillaume Bilodeau et Danny Rioux, SCF et photos b et d, de Shane Sela, ACIA.
24
Figure 4. Localisation du P. ramorum en Europe. Les zones (pays) où le P. ramorum a été retrouvé en pépinière sont en gris. Les pays où le P. ramorum a été détecté en nature ont un cercle noir.
25
CB
Figure 5. Localisation du P. ramorum en Amérique du Nord. Les zones (pays) où le P. ramorum a été retrouvé en pépinière sont en gris. Les États où le P. ramorum a été détecté en nature ont un cercle noir.
26 a
b
c
Figure 6. Structures sexuées et asexuées chez Phytophthora : a) oospore (fusion de l’anthéridie et de l’oogone); b) chlamydospores sur mycélium et c) schéma d’une zoospore. Figure
adaptée
de
a)
http://www.apsnet.org/education/illustratedglossary/
http://www.livingharbour.net/partners/popups/dieback04.htm,
,
et
http://www.botany.hawaii.edu/faculty/wong/Bot201/Myxomycota/Flagellated_fungi.htm
b) c)
27
Figure 7. Arbre phylogénétique du monde du vivant basé sur des séquences d’ADN ribosomiaux. Les Straménopiles sont encerclés en rouge. On peut voir que les Straménopiles, auxquels appartiennent les Oomycètes (encadré jaune), dont les Phytophthora spp., sont assez éloignés des Fungi (adapté de (Baldauf, 2000)).
28
5.0 Références Baldauf S.L., A. J. Roger, I. Wenk-Siefert, and W. F. Doolittle (2000) A kingdom-level phylogeny of eukaryotes based on combined protein data. Science 290, 972–977. Böhm J., A. Hahn, R. Schubert, G. Bahnweg, N. Adler, J. Nechwatal, R. Oehlmann, and W.Osswald (1999) Real-time quantitative PCR: DNA determination in isolated spores of the mycorrhizal fungus Glomus mosseae and monitoring of Phytophthora infestans and Phytophthora citricola in their respective host plants. J. Phytopathol. 147, 409-416. Brasier C. (2003) Sudden oak death: Phytophthora ramorum exhibits transatlantic differences. Mycol. Res. 107, 258-259. Brasier C., and S. Kirk (2004) Production of gametangia by Phytophthora ramorum in vitro. Mycol. Res. 108, 823-827. Brasier C.M., S. Denman, J. Rose, S. A. Kirk, K. J. D. Hughes, R. L. Griffin, C. R. Lane, A. J. Inman, and J. F. Webber (2004) First report of ramorum bleeding canker on Quercus falcata, caused by Phytophthora ramorum. Plant Pathol. 53, 804-804. Brasier C.M., J. Rose, S.A. Kirk, and J.F. Webber (2002) Pathogenicity of Phytophthora ramorum isolates from North America and Europe to bark of European Fagaceae, American Quercus rubra and other forest trees. In Sudden Oak Death Science Symposium. Monterey, California. December 15 - 18, 2002. Cavalier-Smith T. (1998) A revised six-kingdom system of life. Biol. Rev. Camb. Philos. Soc. 73, 203-266. Cavalier-Smith T., M.T.E.P. Allsopp, and E.E. Chao (1994) Thraustochytrids are Chromists, not Fungi: 18s rRNA Signatures of Heterokonta. Philos. Trans. R. Soc. Lond., B 346, 387-397. Cave G.L., B. Randall-Schadel, and S.C. Redlin (2007) Risk Analysis for Phytophthora ramorum Werres, de Cock & In’t Veld, Causal Agent of Phytophthora Canker (Sudden Oak Death), Ramorum Leaf Blight, and Ramorum Dieback, p. 92. United States Department of Agriculture, Animal and Plant Health Inspection Service, Plant Protection and Quarantine, Center for Plant Health Science and Technology, Plant Epidemiology and Risk Analysis Laboratory. CFIA (2003) Hosts of Phytophthora ramorum (with notes on geographical distribution and mating type. July. http://www.cnr.berkeley.edu/comtf/pdf/P.ramorum.hosts.June.2003.pdf.
29 Cooke D.E.L. (2007) Tracking the sudden oak death pathogen. Mol. Ecol. 16, 3735-3736. Davidson J.M., D. M., Rizzo, M. Garbelotto, S. Tjosvold, and G. W. Slaughter (2002) Phytophthora ramorum and sudden oak death in California: II. Transmission and survival. . USDA For. Serv. Gen. Tech. Rep. PSW-GTR-184, 741-749. Davidson J.M., S. Werres, M. Garbelotto, E.M. Hansen, Rizzo D.M. (2003) Sudden oak death and associated diseases caused by Phytophthora ramorum. Plant Health Prog. 1-21. De Merlier D., A. Chandelier, and M. Cavelier (2003) First report of Phytophthora ramorum on Viburnum bodnantense in Belgium. Plant Dis. 87, 203. DEFRA (2004a) First infected oak found in Sussex. Department for Environment, Food And Rural Affairs October 16 http://www.gnn.gov.uk/content/detail.asp?ReleaseID=107094&NewsAreaID=2&Na vigatedFromSearch=True. DEFRA (2004b) Native oaks susceptible to new tree disease. Department for Environment, Food And Rural Affairs.October 16 http://www.gnn.gov.uk/content/detail.asp?ReleaseID=134884&NewsAreaID=2&Na vigatedFromSearch=True. Drenth A.D., and Goodwin S.B. (1999) Population structure of oomycetes. In: Structure and Dynamics of Fungal Populations (ed. Worrall JJ), p. 356. SUNY, College of Environmental Science and Forestry, Syracuse, USA. Erwin D.C., and O.K. Ribeiro (1996) Phytophthora Diseases Worldwide, APS Press edn. American Phytopathological Society, St. Paul, MN. Gaastra W. (1984) Enzyme-linked Immuno-sorbent Assay in Methods in Molecular Biology (part 2) (ed. Walker J), pp. 349-355. Humana Press, Clifton (U.S.A.). Garbelotto M., D.M. Rizzo, J.M. Davidson, and S.J. Frankel (2002a) How to recognize the symptoms of the diseases caused by Phytophthora ramorum, causal agent of Sudden Oak Death. USDA Forest Service, Pacific Southwest Region publication, 15 pp. Garbelotto M., D.M. Rizzo, K. Hayden, M. Meija-Chang, J. M. Davidson, and S. Tjosvold (2002b) Phytophthora ramorum and sudden oak death in California: III. Preliminary studies in pathogen genetics. USDA For. Serv. Gen. Tech. Rep. PSW-GTR-184, 765-774. Garbelotto M., P. Svihra, and D.M. Rizzo (2001) Sudden oak death syndrome fells three oak species. Calif. Agric. 55, 1-19.
30 Giglio S., P.T. Monis, and C.P. Saint (2003) Demonstration of preferential binding of SYBR Green I to specific DNA fragments in real-time multiplex PCR. Nucleic Acids Res. 31. e136. Guo Q.H., M. Kelly, P. Gong, and D.S. Liu (2007) An object-based classification approach in mapping tree mortality using high spatial resolution imagery. GISci. Rem. Sens. 44, 24-47. Guo Q.H., M. Kelly, and C.H. Graham (2005) Support vector machines for predicting distribution of sudden oak death in California. Ecol. Mod. 182, 75-90. Hansard C. (2003) Introduced Species Summary Project Sudden Oak Death (Phytophthora ramorum). Columbia University http://www.columbia.edu/itc/cerc/danoffburg/invasion_bio/inv_spp_summ/Phytophthora_ramorum.htm. Hansen E.M., P.W. Reeser, W. Sutton, and L.M. Winton (2003) First report of A1 mating type of Phytophthora ramorum in North America. Plant Dis. 87, 1267-1267. Hayden K., K. Ivors, C. Wilkinson, and M. Garbelotto (2006) TaqMan chemistry for Phytophthora ramorum detection and quantification, with a comparison of diagnostic methods. Phytopathology 96, 846-854. Herrero M.L., B. Toppe, S.S. Klemsdal, and A. Stensvand (2006) First report of Phytophthora ramorum in ornamental plants in Norway. Plant Dis. 90, 1458-1458. Husson C., C. Delatour, P. Frey, and B. Marcais (2007) First report of Phytophthora ramorum on ornamental plants in France. Plant Dis. 91, 1359-1359. Holland P.M., R.D. Abramson, R. Watson, and D.H. Gelfand (1991) Detection of Specific Polymerase Chain Reaction Product by Utilizing the 5' {rightarrow} 3' Exonuclease Activity of Thermus aquaticus DNA Polymerase. Proc. Natl. Acad. Sci. USA 88, 7276-7280. Hughes K.J.D., P.M. Giltrap, V.C. Barton, E. Hobden, J.A. Tomlinson, and P. Barber (2006a) On-site real-time PCR detection of Phytophthora ramorum causing dieback of Parrotia persica in the UK. Plant Pathol. 55, 813-813. Hughes K.J.D., J.A. Tomlinson, R.L. Griffin, N. Boonham, A.J. Inman and C.R. Lane (2006b) Development of a one-step real-time polymerase chain reaction assay for diagnosis of Phytophthora ramorum. Phytopathology 96, 975-981. Ioos R, L. Laugustin, N. Schenck, S. Rose, C. Husson and P. Frey (2006) Usefulness of single copy genes containing introns in Phytophthora for the development of detection tools for the regulated species P-ramorum and P-fragariae. Eur. J. Plant Pathol. 116, 171-176.
31 Ivors K., M. Garbelotto, I.D.E. Vries, C. Ruyter-Spira, B. T. Hekkert, N. Rosenzweig, and P. Bonants (2006) Microsatellite markers identify three lineages of Phytophthora ramorum in US nurseries, yet single lineages in US forest and European nursery populations. Mol. Ecol. 15, 1493. Ivors K.L., K.J. Hayden, P.J.M. Bonants, D.M. Rizzo, and M. Garbelotto (2004) AFLP and phylogenetic analyses of North American and European populations of Phytophthora ramorum. Mycol. Res. 108, 378-392. Jeffers S.N., and S.B. Martin (1986) Comparison of 2 media selective for Phytophthora and Pythium species. Plant Dis. 70, 1038-1043. Jenkins S., and and N. Gibson (2001) High-throughput SNP genotyping. Comp. Funct. Genomics 3, 57-66. Kluza D.A., D.A. Vieglais, J.K. Andreasen, and A.T. Peterson (2007) Sudden oak death: geographic risk estimates and predictions of origins. Plant Pathol. 56, 580-587. Kristjansson G.T., and S.J., Miller (2008) Plant Health Risk Assessment, Phytophthora ramorum Werres, de Cock & Man in't Veld, Causal agent of Ramorum Blight, Ramorum Bleeding Canker, Ramorum (shoot) Dieback and Sudden Oak Death. Plant Health Risk Assessment Unit, Science Advice Division, Science Branch, Canadian Food Inspection Agency, Nepean, Ontario. Kroon L.P.N.M., F.T. Bakker, G.B.M. van den Bosch, P.J.M. Bonants, and W.G. Flier (2004) Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences. Fungal Genet. Biol. 41, 766-782. Lane C.R., P. A. Beales, K. J. D. Hughes, R. L. Griffin, D. Munro, C. M. Brasier, and J. F. Webber (2003) First outbreak of Phytophthora ramorum in England, on Viburnum tinus. Plant Pathol. 52, 414. Lane C.R., P. A. Beales, K. J. D. Hughes, J. A.Tomlinson, A. J. Inman, and K. Warwick (2004) First report of ramorum dieback (Phytophthora ramorum) on containergrown English yew (Taxus baccata) in England. Plant Pathol. 53, 522. Large E.C. (1940) Advance of the fungi (ed. Cape J), London. Lee L.G., C.R. Connell, and W. Bloch (1993) Allelic discrimination by nick-translation PCR with fluorgenic probes. Nucleic Acids Res. 21, 3761-3766. Levesque C.A. (1997) Molecular detection tools in integrated disease management: overcoming current limitations. Phytoparasitica 25, 3-7. Lilja A., A. Rytkonen, M. Kokkola, P. Parikka, and J. Hantula (2007) First Report of Phytophthora ramorum and P. inflata in Ornamental Rhododendrons in Finland. Plant Dis. 91, 1055-1055.
32
Livak K.J. (2003) 9. SNP Genotyping by the 5'-Nuclease Reaction. In: Single Nucleotide Polymorphisms, Methods and Protocols, (ed. Kwok P-YUoC, San Francisco, CA)), p. 269. Humana press, Toyowa, NJ, (U.S.A.). Marras S.A.E., F.R. Kramer, and S. Tyagi, (2003) 8. Genotyping SNPs With Molecular Beacons. In: Single Nucleotide Polymorphisms, Methods and Protocols, (ed. Kwok P-YUoC, San Francisco, CA)), p. 269. Humana press, Toyowa, NJ, (U.S.A.). Martin F.N., and P.W. Tooley (2003) Phylogenetic relationships among Phytophthora species inferred from sequence analysis of mitochondrially encoded cytochrome oxidase I and II genes. Mycologia 95, 269-284. Martin F.N., and P.W. Tooley (2004) Identification of Phytophthora isolates to species level using restriction fragment length polymorphism analysis of a polymerase chain reaction-amplified region of mitochondrial DNA. Phytopathology 94, 983991. Martin F.N., P.W. Tooley, and C. Blomquist (2004) Molecular detection of Phytophthora ramorum, the causal agent of sudden oak death in California, and two additional species commonly recovered from diseased plant material. Phytopathology 94, 621631. Martin R., D. James, and C. Levesque (2000) Impacts of molecular diagnostic technologies on plant disease management. Annu. Rev. Phytopathol. 38, 207-239. McPherson B.A., D. L .Wood, A. J. Storer, N. M. Kelly, and R. B. Standiford (2001) Sudden Oak Death, a New Forest Disease in California. Integr. Pest Manag. Rev. 6, 243-246. Meentemeyer R., D. Rizzo, W. Mark, and E. Lotz (2004) Mapping the risk of establishment and spread of sudden oak death in California. For. Ecol. Manag. 200, 195-214. Moore-Landeker E. (1982) Fundamentals of fungi, 2nd edition, edition Prentice-Hall inc. edn., New Jersey. Moralejo E., and S.Werres (2002) First report of Phytophthora ramorum on Rhododendron sp. in Spain. Plant Dis. 86, 1052. Osterbauer N., and A. Trippe (2005) Comparing diagnostic protocols for Phytophthora ramorum in rhododendron leaves. Plant Health Prog. http://www,plantmanagementnetwork.org/pub/php/brief/2005/Pramorum Plotkin J.B., J. Dushoff, and H.B.Fraser (2004) Detecting selection using a single genome sequence of M-tuberculosis and P-falciparum. Nature 428, 942-945.
33 Pogoda, F., and S. Werres (2002) Pathogenicity of European and American P. ramorum isolates to rhododendron. In Sudden Oak Death Science Symposium. Monterey, California. December 15 - 18, 2002. Prospero S., J.A. Black, and L.M. Winton (2004) Isolation and characterization of microsatellite markers in Phytophthora ramorum, the causal agent of sudden oak death. Mol. Ecol. Notes 4, 672-674. Prospero S., E.M. Hansen, N.J. Grunwald, and L.M. Winton (2007) Population dynamics of the sudden oak death pathogen Phytophthora ramorum in Oregon from 2001 to 2004. Mol. Ecol. 16, 2958-2973. RAPRA. (2008). Distribution : geographical distribution of naturally infected host. http://rapra.csl.gov.uk/objectives/wp1/naturalhostresults.cfm. Rioux D., B Callan, and D. McKenney (2006) Phytophthora ramorum: causal agent of sudden oak death, ramorum blight, ramorum bleeding canker, ramorum (shoot) dieback. . In: Pest Risk Assessment #00-39 p. 126 pp. CFIA & CFS. Rizzo D.M., M. Garbelotto, J.M. Davidson, G.W. Slaughter, S.T. Koike (2002) Phytophthora ramorum as the cause of extensive mortality of Quercus spp. and Lithocarpus densiflorus in California. Plant Dis. 86, 205-214. Rossman A.Y. and M.E. Palm (2006) Why are Phytophthora and other Oomycota not true Fungi? Outlooks Pest Manag. 17, 193–240. Schaad N., and R. Frederick (2002) Real-time PCR and its application for rapid plant disease diagnostics. Can. J. Plant Pathol. 24, 250-258. Schena L., Hughes K.J.D., and D.E.L.Cooke (2006) Detection and quantification of Phytophthora ramorum, P-kernoviae, P-citricola and P-quercina in symptomatic leaves by multiplex real-time PCR. Mol. Plant Pathol. 7, 365-379. Shearer B.L., and J.T.Tippett (1989) Jarrah dieback; the dynamics and management of Phytophthora cinnamomi in the jarrah (Eucalyptus marginata) forest of southwestern Australia. Department of Conservation and Land Management, Western Australia, Research Bulletin 3, pp. 1-76. Simard M., S.C. Brière, G.J. Bilodeau, R.C. Hamelin, A.K. Watson and D. Rioux (2007) Phytophthora ramorum: évaluation de sa capacité d’infecter six espèces d’arbres indigènes de l’Est canadien. Résumés des communications. Phytoprotection 88, 67. Stukely M.J.C., C.E. Crane, J.A. McComb, and I.J.Bennett (2007) Field survival and growth of clonal, micropropagated Eucalyptus marginata selected for resistance to Phytophthora cinnamomi. Forest Ecol. Manag. 238, 330-334.
34 Svihra P. (1999) Sudden death of tanoak, Lithocarpus densiflorus. UC Cooperative Extension. Pest Alert #1, June, 2p. Tomlinson J.A., I. Barker, and N.Boonham (2007) Faster, simpler, more-specific methods for improved molecular detection of Phytophthora ramorum in the field. Appl. Environ. Microbiol.73, 4040-4047. Tomlinson J.A., N. Boonham, K.J.D. Hughes, R.L. Griffen, and I. Barker (2005) On-site DNA extraction and real-time PCR for detection of Phytophthora ramorum in the field. Appl. Environ. Microbiol. 71, 6702. Tooley P.W., K.L. Kyde, and L. Englander (2004) Susceptibility of selected ericaceous ornamental host species to Phytophthora ramorum. Plant Dis. 88, 993-999. Tooley P.W., F.N. Martin, M.M. Carras, and R.D. Frederick (2006) Real-time fluorescent polymerase chain reaction detection of Phytophthora ramorum and Phytophthora pseudosyringae using mitochondrial gene regions. Phytopathology 96, 336. Tyagi S., and F. Kramer (1996) Molecular beacons: Probes that fluoresce upon hybridization. Nature Biotech. 14, 303-308. Tyagi S., S.A.E. Marras, J.A.M. Vet, and F.R. Kramer (2000) Molecular beacons: hybridization probes for detection of nucleic acids in homogeneous solutions. In: Nonradioactive Analysis of Biomolecules (ed. Springer-Verlag), pp. 606-616. C. Kessler, Berlin. Tyler, B.M., S. Tripathy, X.Zhang, P. Dehal, R.H.Y. Jiang, A. Aerts, F.D. Arredondo, L. Baxter, D. Bensasson, J.L. Beynon, J. Chapman, C.M.B. Damasceno, A.E. Dorrance, D. Dou, A.W. Dickerman, I.L. Dubchak, M. Garbelotto, M. Gijzen, S.G. Gordon, F. Govers, N.J. Grunwald, W. Huang, K.L. Ivors, R.W. Jones, S. Kamoun, K. Krampis, K.H. Lamour, M.-K. Lee, W.H. McDonald, M. Medina, H.J.G. Meijer, E.K. Nordberg, D.J. Maclean, M.D. Ospina-Giraldo, P.F. Morris, V. Phuntumart, N.H. Putnam, S. Rash, J.K.C. Rose, Y. Sakihama, A.A. Salamov, A. Savidor, C.F. Scheuring, B.M. Smith, B.W.S. Sobral, A. Terry, T.A. Torto-Alalibo, J. Win, Z. Xu, H. Zhang, I.V. Grigoriev, D.S. Rokhsar, and J.L. Boore, (2006) Phytophthora genome sequences uncover evolutionary origins and mechanisms of pathogenesis. Science 313, 1261-1266. Van der Auwera G., and R. De Wachter (1997) Complete large subunit ribosomal RNA sequences from the heterokont algae Ochromonas danica, Nannochloropsis salina, and Tribonema aequale, and phylogenetic analysis. J. mol. evol. 45, 84-90. Weller S., J. Elphinstone, N. Smith, N. Boonham, and D. Stead (2000) Detection of Ralstonia solanacearum strains with a quantitative, multiplex, real-time, fluorogenic PCR (TaqMan) assay. Appl. Environ. Microbiol. 66, 2853-2858.
35 Werres S. (2002) A new species of Phytophthora causes Rhododendron dieback. Rhododendron und immergrune Laubgeholze Jahrbuch, 61-67. Werres S., and D. De Merlier (2003) First detection of Phytophthora ramorum mating type A2 in Europe. Plant Dis. 87. Werres S., and K. Kaminski (2005) Characterisation of European and North American Phytophthora ramorum isolates due to their morphology and mating behaviour in vitro with heterothallic Phytophthora species. Mycol. Res. 109, 860-871. Werres S., Marwitz R., Man-in'-t-Veld W.A., de Cock A.W.A.M., and P.J. Bonants (2001) Phytophthora ramorum sp. nov., a new pathogen on Rhododendron and Viburnum. Mycol. Res. 105, 1155-1165. Werres S., and B. Zielke (2003) First studies on the pairing of Phytophthora ramorum. Zeitschrift Fur Pflanzenkrankheiten Und Pflanzenschutz-J. Plant Dis. Protect. 110, 129-130.
36
Chapitre I
Molecular Detection of Phytophthora ramorum by Real Time- Polymerase Chain Reaction Using TaqMan, SYBR Green and Molecular Beacons
37
Molecular Detection of Phytophthora ramorum by Real TimePolymerase Chain Reaction Using TaqMan, SYBR Green and Molecular Beacons G. J. Bilodeau1, C. A. Lévesque2, A.W.A.M. de Cock3, C. Duchaine4, S. Brière5, P. Uribe6, F. N. Martin6, and R. C. Hamelin1 1
Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, P.O. Box
3800, 1055 du P.E.P.S., Sainte-Foy, Quebec, G1V 4C7, Canada; 2
Agriculture and Agri-Food Canada, National Program on Environmental Health -
Biodiversity, 960 Carling Avenue, Ottawa, Ontario, K1A 0C6, Canada; 3
Centraalbureau voor Schimmelcultures, P.O. Box 85167, NL-3508 AD Utrecht, The
Netherlands; 4
Département de biochimie et de microbiologie, Université Laval, Pavillon Alexandre-
Vachon, Quebec, Canada; Centre de recherche de l’Hôpital Laval, 2725 chemin SainteFoy, Sainte-Foy, Quebec, G1V 4G5, Canada; 5
Pest DNA Diagnostics Laboratory, Centre for Plant Quarantine Pests, CFIA, Ottawa,
Ontario, K2H 8P9, Canada; 6
U.S. Department of Agriculture-Agricultural Research Service (USDA-ARS), 1636 East
Alisal St., Salinas, California, 93905, USA; Ce manuscrit fut publié dans la revue Phytopathology, Vol 97 p. 632-642, en mai 2007 : doi:10.1094/PHYTO-97-5-0632.
38
I.1
Résumé/Abstract
I.1.1 Résumé
L’encre des chênes rouge (la mort subite du chêne) causée par l’organisme Phytophthora ramorum, est une maladie affectant plusieurs espèces d’arbres et d’arbustes. Cet agent pathogène se propage très rapidement et des mesures de quarantaines ont été mises en place afin d’éviter sa dissémination dans des endroits ou il était absent. Les outils de diagnostic moléculaire permettent de détecter rapidement et d’identifier le P. ramorum. Cependant, fréquemment, plusieurs de ces tests n’arrivent pas à différencier le P. ramorum des autres espèces de Phytophthora, surtout celles qui sont très proches génétiquement. Pour régler ce problème, et obtenir des tests augmentant le niveau de confiance, les régions de l’ITS (l’espace interne transcritt), le gène de la β-tubuline, et de l’élicitine ont été séquencés et été comparés afin d’identifier du polymorphisme dans une collection d’espèces de Phytophthora. Trois différentes technologies utilisant la PCR en temps réel ont été comparées : les sondes spécifiques (le phare moléculaire et le TaqMan) et le SYBR Green. Les tests ont permis de différencier le P. ramorum de 65 espèces de Phytophthora testées. Les sondes développées ont aussi été vérifiées avec de l’ADN extrait de 48 échantillons de plantes infectées et non infectées. Tous les échantillons environnementaux à partir desquels le P. ramorum a été isolé sur milieu PARP-V8 ont été détectés utilisant les trois tests de PCR en temps réel. Cependant, 24% des échantillons apparaissaient positifs en PCR en temps réel, mais pas en culture du P. ramorum. L’analyse des séquences des régions coxI et coxII, ont confirmé la présence de l’agent pathogène dans plusieurs échantillons. Les tests basés sur la détection des régions de l’ITS et de l’élicitine utilisant la sonde TaqMan, semblent avoir une valeur de limite de détection des cycles plus basse que celle utilisée avec la sonde de la β-tubuline et donc, semblent être plus sensibles.
39
I.1.2 Abstract Sudden oak death, caused by Phytophthora ramorum, is a severe disease that affects many species of trees and shrubs. This pathogen is spreading rapidly and quarantine measures are currently in place to prevent dissemination to areas that were previously free of the pathogen. Molecular assays that rapidly detect and identify P. ramorum frequently fail to reliably distinguish between P. ramorum and closely related species. To overcome this problem and to provide additional assays to increase confidence, internal transcribed spacer (ITS), β-tubulin, and elicitin gene regions were sequenced and searched for polymorphisms in a collection of Phytophthora spp. Three different reporter technologies were compared: molecular beacons, TaqMan, and SYBR Green. The assays differentiated P. ramorum from the 65 species of Phytophthora tested. The assays developed were also used with DNA extracts from 48 infected and uninfected plant samples. All environmental samples from which P. ramorum was isolated by PARP-V8 were detected using all three real-time PCR assays. However, 24% of the samples yielded positive real-time PCR assays but no P. ramorum cultures, but sequence analysis of the coxI and II spacer region confirmed the presence of the pathogen in most samples. The assays based on detection of the ITS and elicitin regions using TaqMan tended to have lower cycle threshold values than those using β-tubulin and seemed to be more sensitive.
40
I.2
Introduction Since 1995, extensive oak mortality, referred to as sudden oak death, has been
reported in California. In 2001, a new pathogen called Phytophthora ramorum Werres, de Cock & Man in’t Veld, was described (Werres et al. 2001b) and identified as the causal agent of sudden oak death (SOD) (Davidson et al. 2003; Davidson 2002a; Garbelotto et al. 2001; Maloney et al. 2002; Murphy and Rizzo 2003; Rizzo et al. 2002). This pathogen has since spread to several counties in California (14 in 2004) and was discovered in Oregon in 2001 (Goheen 2002; Hamelin et al. 1996), but is now increasing its range. It was found again in 2003 in nurseries in Oregon, Washington, and British Columbia (Hansen et al. 2003a), and was subsequently eradicated from infested sites. In 2004, the pathogen was discovered in nurseries in several U.S. states and one Canadian province after a nursery in California shipped infected material across the continent. P. ramorum is also present in Europe where it has been reported in 14 countries, mostly in nurseries, but also in public parks, where it causes leaf blight and shoot dieback mainly in Rhododendron sp. and Viburnum sp. (De Merlier 2003; Jullien 2003; Lane 2003; Moralejo and Werres 2002; Orlikowski 2002; Werres 2002). Recently, it was found on southern red oak, Quercus falcata, in England where the bleeding canker symptoms typical of sudden oak death were observed. Quercus rubra is susceptible to P. ramorum in artificial inoculations and two trees have been found naturally infected in the Netherlands (Brasier and Kirk 2004; Brasier 2004; Denman 2005).
Phytophthora ramorum infects a broad range of hosts and causes different symptoms, including cankers, leaf spots, blights and diebacks (Garbelotto 2003b). Inspection and proper diagnosis of nursery material can therefore be challenging especially on hosts for which the etiology has not been described. Experience in Europe and North America indicates that P. ramorum spreads readily on infected nursery material by zoospore release and that it may also be spread via chlamydospores in soil debris and rain splash (Davidson 2002b).
41 In order to prevent the movement of material from infested to disease-free locations, state and federal quarantine measures were put into place in the U.S.A. and in Canada. Despite these measures, the disease spread to new locations along the coast of the Pacific Northwest in 2003 and 2004. Intensive surveys have been conducted in several countries, including Canada, where this pathogen could represent a risk, to determine whether it is present and to assess the extent of its distribution. Depending on jurisdiction and circumstances such as time of year, survey samples are tested by direct isolation on selective medium, or by some combination of isolation, enzyme-linked immunosorbent assay (ELISA), and polymerase chain reaction (PCR). One of the difficulties encountered with identifications based solely on morphology is that it requires mycological expertise and it does not always allow the distinction of variants within species. This is important in the case of P. ramorum since close relatives of P. ramorum exist in North America and isolates of P. ramorum from North America and Europe are genetically distinct (Brasier 2003; Ivors 2006; Ivors et al. 2004; Kroon et al. 2004a). Furthermore, seasonal variation in the ability to recover P. ramorum from infected field-collected tissue has been reported (Hayden et al. 2004). Molecular diagnostic tools can allow rapid and sensitive pathogen detection and provide an increased confidence level in the identification, even when the pathogen cannot be isolated. Molecular assays have the advantage that they can be used on most types of material, from pure to mixed cultures of the pathogen or symptomatic to asymptomatic host tissue samples. However, identification of P. ramorum based solely on molecular methods can yield false positive results (Garbelotto 2003a; Osterbauer and Trippe 2005; Osterbauer et al. 2004; Tomlinson et al. 2005b) or questionable results that cannot be confirmed by isolation or other methods. Thus, identification of P. ramorum for regulatory purposes is most reliable when using a combination of molecular and other methods.
Recent phylogenetic analyses using internal transcribed spacer (ITS) (Cooke et al. 2000) sequences of the rRNA cistron revealed a close relationship between P. ramorum and P. lateralis Tucker & Milbrath, with a total of 12 single nucleotide polymorphisms between them (Garbelotto 2003a; Rizzo et al. 2002). In addition, other phylogenetic analyses of Phytophthora species conducted with mitochondrial and nuclear DNA
42 sequences have revealed a close relationship with P. hibernalis (Kroon et al. 2004c; Martin and Tooley 2004). Given the close phylogenetic relationships among these species, one PCR assay, developed from the ITS region, may not always reliably discriminate P. ramorum from P. lateralis or P. hibernalis (Garbelotto 2003a; Osterbauer and Trippe 2005). This could be important since P. lateralis, the causal agent of Port Orford cedar root rot, is commonly found on the west coast of North America (Garbelotto 2003a; Garbelotto 2003b; Winton and Hansen 2001). Other real-time PCR protocols were developed to detect P. ramorum and were using single gene detection (Hayden et al. 2006; Hayden et al. 2004; Tomlinson et al. 2005b; Tooley et al. 2006).
The objectives and approach of this investigation were the following: (i) to sequence two nuclear genes (β-tubulin and elicitin “ramorumin”) of P. ramorum and related species and use existing ITS sequences to identify polymorphisms specific to P. ramorum; (Dodd et al.) to design and test real-time (RT)-PCR assays to specifically detect P. ramorum, based on the polymorphisms identified using three reporter technologies (SYBR Green, TaqMan probes and molecular beacons targeting the three gene regions); and (iii) to validate the real-time PCR assays using DNA from pure cultures of a comprehensive collection of Phytophthora species and from infected plant samples.
I.3
Materials and methods
I.3.1 Isolates All isolates of Phytophthora used in this study are listed in Table 1. The P. ramorum collection comprised 38 isolates from European and North American origin isolated from different hosts. The Phytophthora collection comprised isolates of 65 different species and varieties, representing most of the recognized Phytophthora species currently available in pure culture. Mycelium was cultivated and DNA was extracted
43 following the procedures described in De Cock et al. (De Cock et al. 1992a) or Möller et al. (Moller et al. 1992).
I.3.2 DNA sequencing The primers listed in Table 2 were used to amplify three gene regions of the nuclear DNA of P. ramorum and related Phytophthora species by PCR (Table 1). Primers for the βtubulin and the elicitin genes were designed based on sequences of these genes obtained from Phytophthora parasitica (GenBank accession number S67432) for elicitin, and Phytophthora cinnamomi (GenBank accession number U22050) for β-tubulin. ITS primers used were from Bakkeren et al. (Bakkeren et al. 2000) and Mazzola et al. (Mazzola et al. 2002), and the sequence AY038050 from GenBank for P. ramorum was used for alignment. These nuclear regions were selected because of the high level of sequence divergence among species were observed in preliminary results and large number of sequence entries available in public databases.
Genomic DNA from P. ramorum (CBS 101553, DAOM 230728) and other species was amplified using these genus specific primers. Most reactions yielded a single band. However, multiple bands were amplified for the elicitin gene of P. lateralis ATCC 201856, P. cactorum BR675, P. citricola BR 681, P. cinnamomi BR 680 and P. infestans CBS 366.51. A band of approximately 280 bp was cut and extracted from agarose gels with the QIAEX II agarose gel extraction kit (Qiagen, Valencia, CA). PCR products were then reamplified.
PCR products were purified with the QIAquick PCR purification Kit (Qiagen) using the microfuge method, quantified and sequenced using the same primers as for PCR. Sequencing reactions were performed using a Big Dye Terminator Sequencing kit on an ABI 310 automated sequencer (PE Applied Biosystems, Foster City, CA). Both strands were sequenced with the primers listed in Table 2. Sequences were aligned using Sequencher version 4.0.5 (Gene Codes Corporation, Ann Arbor, MI) and MegAlign version
44 5.08 (DNASTAR Inc., Madison, WI) using Clustal W. The sequences were deposited in GenBank and the accession numbers are listed in Table 3.
I.3.3 Design of primers
The alignments of sequences listed in Table 3 were used to design PCR primers specific to P. ramorum (Table 2) using the software Primer Premier 5.00 (Premier Biosoft International, Palo Alto, CA). The selection criteria were the following: Tm (melting temperature) 55-65°C, primer length 18-22 bp, and absence of secondary structure whenever possible. Specific primers were designed so that the nucleotides unique to the target were at the 3’ end position of the primer. In primers Phy_ram_482U_LNA F and Prameli259L R (Table 2), positioning the discriminating site at the 3’ end resulted in secondary structures and the primers were moved toward the 5’ end. The primer pairs were designed such that PCR products were shorter than 200 bp, an important parameter for RTPCR. In cases where only single nucleotide differences were present and unmodified primers did not allow specific amplifications, primers were synthesized with a Lock Nucleic Acid (LNA) (Braasch and Corey 2001) (Proligo LLC, Boulder, CO) to increase specificity. The ITS primers were different from other published primers (Garbelotto 2003a; Hayden et al. 2006; Hayden et al. 2004) and targeted instead positions 622-755 with mismatches at the 3’ site and the LNA modification.
I.3.4 Design of molecular beacon and TaqMan probes Molecular beacons were designed using Mfold version 3.1 (DNA mfold server: 1996-2003, Michael Zuker, Rensselaer Polytechnic Institute) and Beacon designer 3 software (Premier Biosoft International, Palo Alto, CA) to calculate the Tm and the structure of the molecule. The molecular beacon was labelled with fluorescein (6-FAMtm) at the 5’ end and with the quencher Dabcyl at the 3’ end (Bonnet et al. 1999; Tyagi and
45 Kramer 1996; Tyagi et al. 2000). TaqMan probes (Heid et al. 1996) were designed with Primer Premier 5.00. We used the following parameters for the design: Tm 10°C higher than the primers, 15-30 bp in length and the total number of G’s or C’s in the last five nucleotides at the 3' end of the primer not exceeding two. The mismatching nucleotide was positioned as close as possible to the middle of the probe rather than at the ends while avoiding positions with secondary structures. The TaqMan probes were labelled with fluorescein (6-FAMtm) at the 5’ end and with the quencher Black Hole Quenchertm-1 (BHQtm-1) at the 3’ end (Integrated DNA Technologies Inc., Coralville, IA).
I.3.5 PCR amplification Real-time PCR was performed with a DNA Engine Opticon® 2 Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA). Fluorescent molecules (SYBR Green, dual-labeled probe [TaqMan® or molecular beacons]) were included in the PCR master mix (QuantiTectTM, Qiagen). All reactions were performed in 25 µl volumes. The DNA concentration used in the reaction was determined with a Nanodrop spectrophotometer ND-1000 (Nanodrop Technologies, Wilmington, DE) and ranged from 0.05 to 60 ng/µl for P. ramorum and from 5 and 110 ng/ul for other Phytophthora sp. Molecular beacon, TaqMan and SYBR Green real-time PCR assays were used in a preliminary experiment with the three gene regions using a subset of isolates. The real-time PCR on P. ramorum tested on the three gene regions were made in triplicates, and the mean and standard error were calculated for each. Subsequently, TaqMan assays were used for further studies using 48 plant tissue samples from seven hosts. Phytophthora ramorum CBS 101553 was used as positive control, and P. lateralis CBS 168.42 and no template DNA were used as negative controls. The analysis software Opticon Monitor version 2.01.10 (Bio-Rad Laboratories, Hercules, CA) was used to analyze the data (cycle range set at 121). Data were exported as Ct values and analyzed for comparisons among samples in Excel spreadsheets. Statistical analyses were performed with Excel (Microsoft® Excel VER.9.0.3821 SR-1, Redmond, WA). The specific reaction conditions for each of the three detection technologies tested were set up as follows.
46
I.3.5.a SYBR Green
The PCR reaction contained 0.4 µM of each LNA primer or regular primer (Table 2), 1X of QuantiTectTM SYBR Green PCR Kit (Qiagen, Valencia, CA), and template DNA. PCR cycling conditions were set at 95°C for 15 min, 40 cycles at 94°C for 15 s, 60°C for 30 s, and 72°C for 30 s. Fluorescence was read during the extension at 72°C.
I.3.5.b TaqMan® probe
The PCR reaction contained 0.4 µM of each LNA primer or regular primer depending on the region used, 0.2 µM of TaqMan (dual-labeled probe) (Table 2), 1X of QuantiTectTM Probe PCR Master Mix (Qiagen, Valencia, CA), and template DNA. PCR cycling conditions were set at 95°C for 15 min, 36 cycles at 94°C for 15 s, and 65°C for 60 s (68°C for 60 s for elicitin). Cycle number was extended to 40-45 for β-tubulin and ITS on environmental samples to increase sensitivity. Fluorescence was read during the extension at 65-68°C.
I.3.5.c Molecular beacon
The PCR reaction contained 0.4 µM of each LNA primer for β-tubulin (Table 2), 0.2 µM of molecular beacon probe (Table 2), 1X of QuantiTectTM Probe PCR Master Mix, and template DNA. PCR cycling conditions were set at 95°C for 15 min, 36 cycles at 94°C for 15 s, 65°C for 30 s, and 72°C for 30 s. Fluorescence was read after the annealing at 65°C. This allows the molecular beacon to open and anneal with the target sequence.
47
I.3.5.d Standard curves
Serial dilutions of P. ramorum DNA were used to calculate amplification efficiency with TaqMan assays. DNA serial dilutions consisted of P. ramorum strain CBS 101327 with an estimated initial concentration of 10 ng/ul diluted in a series of 1:10 from 1 ng/ul to 1 fg/ul. Three dilution series were done with Tris buffer, 10 mM pH 8.0, Rhododendron sp. DNA (38.8 ng/ul) and P. monticola DNA (3.5ng/ ul). Standard curves were replicated for each marker used twice samples for each dilution. The results were analyzed by plotting the Log of template concentration against cycle threshold (Zeller et al.) values. PCR efficiency was calculated with the formula, E = (10
(-1/slope)
– 1) × 100, where E is the amplification
efficiency and the slope is derived from the plot of Log of template concentration vs Ct. A slope of 3.32 translates into one hundred percent efficiency of amplification.
I.3.6 material
DNA extraction and pathogen isolation from infected plant
To test sensitivity and specificity of the assays on infected plant material, environmental samples were obtained from different hosts either healthy or naturally infected by the pathogen. Samples were collected on 14 June 2005, at Pfeiffer Big Sur State Park, Monterey County, CA, and processed at the USDA/ARS station in Salinas, CA. Protocols for sample collection and handling followed California Oak Mortality Task Force (Pest alert 6, Diagnosis and Monitoring of Sudden Oak Death, March 2002) recommendations. Briefly, symptoms resembling SOD were sought in the field. Leaf blights and diseased twigs of plants belonging to host species such as bay laurel (Umbellularia californica), madrone (Arbutus menziesii), coffee berry (Frangula californica), christmas berry (Heteromeles arbutifolia) and tan oak (Lithocarpus densiflorus) were individually collected, annotated and stored in cold for transportation and processing. For comparison purposes samples of healthy hosts such as camellia (Camellia
48 spp.) and live oak (Quercus agrifolia) were collected at the USDA/ARS station in Salinas CA. Samples were superficially cleaned with sterile water from excess soil and debris and blot dried. For species in which the pathogen typically causes cankers such as tan oaks and live oaks, often necrosis of twigs and wilting of small branches signal ongoing infections. Therefore, instead of sampling the bark, wilting twigs were collected. This type of sample has been shown to correlate with P. ramorum infection and to subsequently produce symptoms such as shoot tip dieback, leaf flagging, or the formation of a Shepard’s crook (Davidson et al. 2003). Once in the laboratory, tan oak leaves including the petioles were used as starting material to isolate the pathogen. Particular care was given to use adjacent tissues of the infected sample to perform the in vitro culture and DNA isolation of the pathogen.
Host DNA was isolated using the USDA-APHIS protocol with the Qiagen DNeasy kit (APHIS 2004). For control purposes most of the healthy samples of live oak and camellia were spiked with P. ramorum CBS 101553 and P. pseudosyringae EH P96 DNA. Aliquots of the DNA samples were coded and sent blind to our laboratory for testing. Pathogen isolation was done on PARP (pimaricin-ampicillin-rifampicin-PCNB agar) media (Jeffers and Martin 1986).
In addition, environmental samples were tested with a coxI/II TaqMan in RT-PCR (Tooley et al. 2006). To confirm that P. ramorum was present in PARP (-)/RT-PCR (+) amplicons, Phytophthora genus-specific primers were used to amplify the spacer regions between the coxI and II genes and this amplicon was sequenced as previously described (Martin et al. 2004).
I.4
Results
I.4.1 Phytophthora sequence divergence
49 Approximately 2 kb for the two nuclear genes and ITS ribosomal DNA region were sequenced from several Phytophthora species (Table 3). Depending on the species, close to 866 bp, 260 bp, and 780 bp were sequenced for β-tubulin, elicitin, and the ITS, respectively. Divergence was the highest for β-tubulin and elicitin for comparisons between the two closely related species P. ramorum and P. lateralis. A total of 18/866 bases for βtubulin (2.2% divergence) and 4/93 bases for elicitin (4.5% divergence) were polymorphic between these two species. In contrast, the ITS contained only 13 polymorphisms out of 783 bases (1.4% divergence) between these two species. However, divergence among species was greater for comparisons with more distant Phytophthora species. For example, divergence between P. ramorum and P. cactorum was 6.0%, 11.4%, and 10.8% for βtubulin, elicitin, and ITS, respectively. Similarly, divergence between P. ramorum and P. cinnamomi was 7.5%, 9.8%, and 12.2% for these three genes. A similar level of sequence divergence was observed in comparisons with P. citricola, P. infestans, and P. nicotianae. However, the low number of polymorphisms in β-tubulin was still sufficiently high among the distant species to design specific primers and probes for P. ramorum.
I.4.2 β-tubulin PCR assays with TaqMan, molecular beacon, and SYBR Green Using the β-tubulin primers, the three PCR reporter technologies were compared. PCR primers specific to P. ramorum were designed for β-tubulin to amplify an amplicon of 171 bp. The primers contained several nucleotides polymorphic among Phytophthora species used in this study (Tables 2 and 3). However, because of the low divergence between P. ramorum and P. lateralis, and the constraints with designing primers to amplify amplicons less than 200 bp for real-time PCR assays, both β-tubulin primers contained only one polymorphic nucleotide and were modified with LNA to increase specificity. Specificity of the β-tubulin assays was compared for TaqMan, molecular beacon, and SYBR Green in real-time PCR initially on P. ramorum, P. lateralis, P. cactorum, P. cinnamomi, P. citricola, and P. infestans (Fig. 1), representing closely related as well as more diverged species.
50
All real-time PCR assays targeting the β-tubulin gene yielded Ct values for the P. ramorum samples, but the Ct varied according to the reporter technology used. The TaqMan, SYBR Green, and molecular beacon assays yielded Ct values of 22.04, 22.77, and 25.08, respectively. No amplifications were observed with the non-target Phytophthora species tested. Since PCR efficiencies were comparable among the three assays (results not shown) and based on the fact that TaqMan probes are easier to design than molecular beacon, we designed TaqMan reporter probe assays for the comparison of PCR efficiencies in the three genes targeted (Table 2).
I.4.3 Comparison of ITS, β-tubulin and elicitin TaqMan probes using real-time PCR To compare the specificity and sensitivity of probes in different gene regions, we focused on the TaqMan assay using DNA from 65 species of Phytophthora (Table 1) from pure cultures. Again, the Ct values for P. ramorum varied according to the gene region targeted. Ct values were 20.8 for the ITS-TaqMan and 21.4 for the elicitin-TaqMan realtime PCR assays. The β-tubulin TaqMan assay yielded the highest Ct values (Ct=24.4). None of the remaining Phytophthora spp. templates yielded Ct values with any of the TaqMan real-time PCR assays, indicating that the assays were highly specific, including species that frequently cross-react with P. ramorum such as P. pseudosyringae, P. nemorosa and P. kernoviae (not shown).
Standard curves with serial dilutions of P. ramorum DNA were generated to calculate amplification efficiency. Amplification efficiencies were 100%, 91% and 100%, in TaqMan assays targeting ITS, elicitin and β-tubulin diluted in Tris-HCL (Fig. 2). Addition of plant DNA from pine changed amplification efficiencies for these markers to 86%, 100%, and 91%, respectively, while rhododendron DNA changed them to 78%,
51 100%, and 91% (results not shown). The limit of detection was 1 fg, 10 fg and 100 fg for the ITS, elicitin and the β-tubulin TaqMan assays, respectively (Fig. 2).
To determine whether the developed assays were applicable to a broad range of P. ramorum sources, 30 isolates of P. ramorum from Europe and North America (Table 4) were tested with the TaqMan probes for the three different gene regions. All reactions with P. ramorum isolates resulted in fluorescent curves that rose above the threshold value, regardless of the gene targeted by the real-time PCR assay (Table 4). Again, the Ct values varied according to the gene targeted, with Ct values for P. ramorum averaging 20.02 (Standard error [S.E.] 2.97), 20.65 (S.E. 2.46), and 24.70 (S.E. 2.57) for the ITS, elicitin, and β-tubulin, respectively (Table 4). The relatively large S.E. reflects the variability in initial DNA concentration (ranging from 2-32 ng).
I.4.4 Detection of Phytophthora ramorum in infected samples. To determine the reliability of the assays to amplify P. ramorum DNA from infected plant samples and from samples spiked with Phytophthora DNA, the three TaqMan real-time PCR assays were used to test 48 different plant extracts from seven hosts in a blind test (Table 5). All samples were plated on PARP-V8 medium and tested using cox I/II RT-PCR (Tooley et al. 2006) at the USDA/ARS laboratory in Salinas, CA, and assayed in a blind test at the Laurentian Forestry Centre (LFC) laboratory. All samples spiked with P. ramorum were detected accurately using all of the assays developed in this study, and were concordant with the cox I/II assays. All of the environmental samples gave concordant results for the four molecular assays, including 63% negatives and 13% positives (excluding spiked samples). All environmental samples that were naturally infected and yielded P. ramorum cultures on PARP-V8 as well as positive PCR results with the coxI/II assay were positive with the ITS, β-tubulin and elicitin assays. There was some level of discordance among the results for the molecular assays and the results of the culture isolation; the pathogen was not recovered from 24% of the samples that were PCR positive. The presence of P. ramorum in all but four of these samples was confirmed by
52 sequence analysis of the spacer region between the coxI and II genes amplified using Phytophthora genus-specific primers (data not shown). The sample spiked with P. pseudosyringae was negative with all molecular assays. There were some cross reactions with P. lateralis at a high DNA concentration (8ng/ul) with the elicitin TaqMan assay with a Ct value that was higher than the target. However, the results of the ITS assay were negative and the Ct for the β-tubulin assay was 41, above what normally would be considered a positive sample.
Ct values in samples from infected material were higher than in the pure culture assays described in the previous sections. The Ct values ranged from 23-39 (TaqMan-ITS) to 28-40 (TaqMan-β-tubulin) and 21-34 (TaqMan-elicitin) in environmental samples that were tested positive (Table 5).
I.5
Discussion Real-time PCR has been used in plant pathology for DNA quantification and
diagnosis of several plant pathogens (Schaad and Frederick 2002). The potential of molecular detection in plant pathology has been shown in several studies (Martin et al. 2000; McCartney et al. 2003; Schaad and Frederick 2002). The advantages include speed of the assays and, the possibility of detecting non-culturable pathogens with high sensitivity. The disadvantages include cost, the need for well-equipped laboratories, and the inability to determine if the target organism is viable. Also, the risk of obtaining false positives and negatives can never be fully discounted due to the inability of fully assessing the biodiversity of target species and close relatives. To reduce the risk of incorrect identification (e.g. in regulatory applications), confirmation of morphometric properties (from culture isolation) should be included in the diagnostic protocol.
The real-time PCR assays described in this research resulted in the specific and sensitive detection of P. ramorum from cultures and plant samples. Tests with pure cultures
53 of Phytophthora species and P. ramorum from a worldwide collection demonstrated the specificity of the ITS, β-tubulin and elicitin TaqMan assays. No false positive or false negative results were observed with any of the assays consisting of dual probes and specific PCR primers, except for P. lateralis at high DNA concentration in the environmental samples experiment. However, Ct values were high even at those high DNA concentrations (41 for β-tubulin and 33-34 for elicitin) and values over 40 would be considered negative in an operational assay. In addition, the assays were tested using infected plant material from which the pathogen had been isolated, or with plant material spiked with P. ramorum DNA. This validation step was done as a blind test and confirmed the concordance between the various molecular assays and the increased confidence afforded by the gene region redundancy in the assays.
Our study compared molecular beacon, TaqMan and SYBR Green assays for more than one gene region against a single target microorganism. The use of multiple gene targets increases reliability and confidence in the assays. Although it should be possible to directly multiplex the internal probe assays into single reactions, we preferred to conduct assays in separate reactions for the different gene targets. Multiplexing generally results in some competition among the PCR products and could reduce the amplification efficiency and thus the sensitivity of the assays. Molecular beacons and TaqMan probes also have built-in redundancy of polymorphisms and give two potential levels of specificity in their design: the PCR primers and the internal dual-labelled probe. In the assays we developed, this provided several polymorphic nucleotides between the target P. ramorum and related taxa. Even for the closely related P. lateralis, which has only 1.4% divergence with P. ramorum in the ITS region, it was possible to design a TaqMan assay with one polymorphic nucleotide per PCR primer, and an additional three polymorphic nucleotides within the TaqMan probe. The extra specificity given by the internal probes could be significant in reducing the probability of false positives. As SYBR Green intercalates nonspecifically into all double-stranded DNA molecules during the reaction, the assay must rely entirely on the PCR primers for specificity (Giglio 2003). In addition, the inhibition of the Taq DNA polymerase by SYBR Green could result in lower specificity (Arezi et al. 2003). In our experiments, the annealing temperature was lowered in the reactions using
54 SYBR Green compared with those using internal probes to counter this inhibition. Since the same primers were used in the SYBR Green and the TaqMan and molecular beacon assays, this could have resulted in a lower specificity in SYBR Green.
The TaqMan assays developed here produced very high levels of amplification efficiency. The fluorescence continued to increase even after 40 cycles in some trials, suggesting the reaction components had not yet become limiting even if the reaction was very efficient. In a dilution series of P. ramorum DNA from 1 fg to 10 ng, the TaqMan system with elicitin, β-tubulin and ITS required on average 3.30, 3.30 and 3.60 cycles per 10-fold dilution to pass the threshold, which is very close to the theoretical value of 3.33 if the DNA molecules are doubled at every cycle (Fig. 2). The lower than 100% amplification efficiency could be explained by several factors, including the addition of plant DNA in our dilution series to reproduce amplification conditions encountered when processing field samples. In contrast, SYBR Green and molecular beacon fluorescence reached a plateau probably because fluorescence is generated “de novo” at every cycle by the intercalation of the SYBR Green dye in the double stranded DNA during extension or by the hybridization of the molecular beacon at annealing. Alternatively, it is possible that the dye used for the TaqMan probe was more fluorescent than in the other assays, although FAM was used in all those assays and different quenchers were used for molecular beacon and TaqMan. However, the Ct values and the overall intensity of fluorescence of each sample followed the same trends among the three types of DNA detection assays (Fig. 1).
DNA extracts from 65 of the 84 documented Phytophthora species were tested in our study with the ITS, β-tubulin and elicitin TaqMan assays. These P. ramorum-specific assays did not cross react with other Phytophthora species when tested at 5-110 ng, including the closely related P. hibernalis. However, at high concentrations (8ng/ul), we did get some cross-reaction with P. lateralis using the P. ramorum-specific assays for the elicitin marker. Similar cross-reactivity has been reported between P. ramorum and P. lateralis for an ITS TaqMan assay (Tomlinson et al. 2005b). The three TaqMan assays also correctly detected P. ramorum in the plant tissue extracts tested. The sensitivity of the TaqMan assay was approximately 4-7 cycles higher when used with the ITS and elicitin
55 probes compared with the mean Ct of β-tubulin with the same samples (Table 5). This would make the TaqMan-ITS and TaqMan-elicitin assays approximately 1 or 2 orders of magnitude more sensitive than the β-tubulin assay with any of the techniques used.
By comparing standard curves, we could determine that the ITS TaqMan assay was the most sensitive, followed by the elicitin and β-tubulin assays. The TaqMan ITS assay yielded Ct values that were on average 4.5 cycle lower than the β-tubulin assays and 1.94 cycles lower than the elicitin assay. This is somewhat expected since rDNA consists of a tandemly repeated gene cluster, and therefore a single cell contains several copies of the cistron (White et al. 1990). Elicitin is also known to have multiples copies (Ponchet 1999). By contrast, β-tubulin is usually found in fewer copies (Ayliffe et al. 2001; Weerakoon 1998). Using genes in multiple copies allows detection of smaller amounts of target DNA compared with assays based on single copy genes. Molecular diagnosis of P. ramorum based on nested PCR amplification of the ITS has been reported (Garbelotto 2003a; Garbelotto 2003b; Hayden et al. 2006). Although nested PCR can be very powerful and sensitive, it requires additional manipulations of the amplified DNA and can therefore result in higher risks of contamination, and therefore higher rates of false positives (Osterbauer and Trippe 2005). Our assay also uses the ITS region but with a single round of PCR in a closed tube assay, reducing the possibility of false positives due to carryover contamination. However, direct comparison of the nested PCR protocol with the assays described here would be necessary to determine the relative merits of each test for specificity and sensitivity.
Other PCR-based methods have been used to characterize P. ramorum at the DNA level. AFLP assays have been used for molecular characterization of P. ramorum isolates. These studies revealed that the European (EU) and North American (NA) strains of P. ramorum differ (Brasier 2003; Ivors 2006; Ivors et al. 2004; Kroon et al. 2004a; Prospero et al. 2004b; Werres and De Merlier 2003b) and show variation. Our sequence analysis also revealed some single nucleotide polymorphism (SNP) differences between the EU and NA isolates in the β-tubulin gene. Thus far, this polymorphism correlates completely with the
56 origin of the strains (e.g. European vs North American). We are currently developing a SNP genotyping assay that allows high throughput characterization for geographic origin (Bilodeau 2003) and are in concordance with microsatellites analysis (Ivors 2006). All field samples that yielded cultures of P. ramorum gave positive PCR results but clearly, the PCR assays also detected P. ramorum consistently in some samples that P. ramorum could not be recovered. In all but four cases, however, sequence analysis of the spacer region between the coxI and II genes confirmed the presence of the pathogen. Sequencing of this spacer region has been proven to be a valuable tool for organism identification in Phytophthora species (Martin and Tooley 2004) and prior evaluations with a larger number of field samples confirmed the ability of the coxI /II marker system to identify the pathogen when it could not be cultured (Martin et al. 2004; Tooley et al. 2006). For these four samples clean sequence data was not obtained, perhaps due to the presence of multiple Phytophthora species in the sample. The high Ct values obtained when using the markers developed in this submission indicated that the amount of P. ramorum was probably very low. The greater sensitivity of molecular assays compared with culture assays is not entirely unexpected as culture methods can be affected by the presence of contaminants, competitors or inhibitors (e.g. fungicides) in the tissues (Hamelin et al. 1996; Hamelin 2000), as well as seasonal variation (Hayden et al. 2004). Moreover, nonviable pathogen propagules would result in positive PCR assays, but cultures would not be obtained.
The real-time PCR assays developed here can be a useful tool to detect P. ramorum rapidly and sensitively from pure cultures and plant tissue samples. Our approach allows for redundancy both in the gene regions targeted by designing assays that take advantage of SNPs at priming sites as well as at internal sites for probes. This approach, combined with use of assays for multiple gene/spacers targets, should greatly increase the confidence level of these assays and reduce the potential for false positive results in molecular testing.
57
I.6
Acknowledgements We thank A. de Cock and R. Nijman for cultivation of the fungi and isolation of
most of the DNA samples as well as N. Desaulniers and T. Barasubiye for their help with DNA sequencing. Funding for this work was received from the Canadian Biotechnology Strategic Fund and the CBRN Research and Technology Initiative (CRTI grant 040045RD).
58
Table 1. Isolates of Phytophthora species from different culture collections used in this study Phytophthora Collection Phytophthora Collection Phytophthora Collection species number species number species numbera arecaeb CBS 305.62 katsuraeb CBS 587.85 ramorum CBS 101550 boehmeriaeb
CBS 291.29
lateralisbc
ATCC201856
ramorum
CBS 110544
botryosab
CBS 581.69
lateralis
CBS 168.42
ramorum
CBS 109278
brassicaeb
CBS 178.87
meadiib
CBS 219.88
ramorumc
CBS 101552
brassicaeb
CBS 686.95
megakaryab
CBS 238.83
ramorum
CBS 101549
cactorumb
CBS 108.09
CBS 402.72
ramorum
CBS 101331
cactorumb
CBS 108.09
megaspermab v. megasperma melonisb
CBS 582.69
ramorum
CBS 110535
cactorumc
mexicanab
CBS 554.88
ramorum
CBS 110547
cambivorab
DAOM BR 675 CBS 248.60
mirabilisb
CBS 678.85
ramorum
CBS 1110536
capsicib
CBS 128.23
multivesiculatab
CBS 545.96
ramorum
CBS 110542
cinnamomib
CBS 144.22
nicotianaeb
CBS 305.29
ramorum
CBS 110534
cinnamomibc
DAOM BR 680
CBS 109.17
ramorum
CBS 110900
cinnamomib v. parvispora cinnamomib var parvispora citricolab
CBS 413.96
nicotianae b (type of terrestris) operculatab
CBS 241.83
ramorum
CBS 110901
CBS 411.96
palmivorab
CBS 236.30
ramorum
adc 01.01
CBS 221.88
phaseolib
CBS 556.88
ramorum
adc 01.06
citricolac
DAOM BR 681 CBS 181.25
primulaeb
CBS 275.74
ramorum
pseudosyringae
EH P96
ramorum
CBS 950.87
pseudotsugaeb
CBS 444.84
ramorumc
clandestinab
CBS 347.86
psychrophilab
CBS 803.95
ramorum
colocasiaeb
CBS 955.87
quercinab
CBS 784.95
richardiaeb
DAOM 229466 DAOM 230729 DAOM 230728 DAOM 230727 CBS 240.30
cryptogeab
CBS 113.19
ramorum
CBS 101327
sinensisb
CBS 557.88
cryptogeaa f. sp. begoniae drechslerib
CBS 468.81
ramorum
CBS 101326
sojaeb
CBS 418.91
CBS 291.35
ramorum
CBS 110538
sojaeb
CBS 382.61
erythrosepticab
CBS 129.23
ramorum
CBS 110601
spb.(‘aquatica’)
CBS 363.79
erythrosepticab (type of
CBS 357.59
ramorum
CBS 110543
sp. b (‘marine’)
CBS 215.85
citricolab (type of P. pini) citrophthorab
59 P.himalayensis) erythrosepticab var pisi europeab
adc 99.69
ramorum
CBS 110537
syringaeb
CBS 132.23
CBS 109049
ramorum
CBS 110541
syringaeb
CBS 367.79
fragariaeb v. fragariae fragariaeb var rubi
CBS 209.46
ramorum
CBS 110539
tartareab
CBS 208.95
CBS 967.95
ramorumb
CBS 109279
tentaculatab
CBS 552.96
gonapoyidesb
CBS 554.67
ramorum
CBS 101554
tropicalisb
CBS 434.91
heveaeb
CBS 296.29
ramorum
CBS 110545
uliginosab
CBS 109054
hibernalisb
CBS 522.77
ramorum
CBS 110548
vignaeb
CBS 241.73
humicolab
CBS 200.81
ramorum
CBS 101553
idaeib
CBS 971.95
ramorum
CBS 101329
ilicisb
CBS 255.93
ramorum
CBS 101551
infestansbc
CBS 366.51
ramorumb
CBS 101332
insolitab
CBS 691.79
ramorum
CBS 101330
iranicab
CBS 374.72
ramorum
CBS 110546
kandeliib
CBS 111.91
ramorum
CBS 101548
a
CBS: Centraalbureau voor Schimmelcultures, DAOM: Canadian Agriculture and Agri-Food Canada, Ottawa, adc: Arthur W.A.M. de Cock, ATCC: American Type Culture Collection, EH: Everett Hansen collection b Samples used in the comparison of ITS, β-tubulin and elicitin TaqMan probes using real-time PCR tested with 65 species of Phytophthora c Sample used in Figure 1: Real-time PCR of Phytophthora samples with TaqMan, molecular beacons and SYBR Green (β-tubulin gene).
60 Table 2. Primers and probes used for PCR assays targeting Phytophthora spp. and P. ramorum. Namea
Primer sequence
Region
Notes
Βetatubulin
˜539bp
Βetatubulin
˜500bp
ITS
˜350bp
ITS
˜650bp
Elicitin
˜300bp
Betatubulin
Primers - LNA modified 171 bp
ITS
Primers- LNA modified 133bp
Universal to Phytophthora species Oom-Btub-up415 F Oom(Ph)-Btub-lo954 R Oom(Ph)-Btub-up901 F Oom-Btub-lo1401 R UN-UP18S42 F b OOM-LO5.8S47 R c OOM-UP5.8S-55 F c UN-LO28S22 R b Phy_elicitor1 F Phy_elicitor2 R
5’-CGCATCAACGTGTACTACAA-3’ 5’-GCACACCAGGTGGTTC-3’ 5’-TACGACATTTGCTTCCG-3’ 5’-CGCTTGAACATCTCCTGG-3’ 5’-CGTAACAAGGTTTCCGTAGGTGAAC-3’ 5’-ATTACGTATCGCAGTTCGCAG-3’ 5’-TGCGATACGTAATGCGAATT-3’ 5’-GTTTCTTTTCCTCCGCTTATTGATATG-3’ 5’-GCCCTCGTCGGCTCCAC-3’ 5’-GTGAACACGTTGAGTACCAGGC-3’
Specific to Phytophthora ramorum d Phy_ram_482U_LNA 5’-GGCGCTGTACGACATTTG-3’ F Phy_ram_653L_LNA 5’-ACGCGGGAACGGAATCAA-3’ R ITSPrim622U F 5’-AATGACTGGTGAACCGTAGCTG-3’ ITSPrimer755L R
5’-CGAAGCCGCCAACACAAG-3’
Prameli 102U F Prameli259L R
5’-TTCAACCAGTGCGCGACC-3’ 5’-GGCACAGTCAGCTCGCAGTC-3’
Elicitin Elicitin
157bp
Molecular beacon Beacon4. Beta tub F
[6~FAM]CGGGCTCGGACATAGCGGCGCACACC BetaAGCCCG[Dabcyl] tubulin
TaqMan probes [6~FAM]CGTGGTGATGCCGGACATAGCG[BHQ1 Betatubulin ~Q]e [6~FAM]AACACCGTCGATTCAAAAGCCAAGC[ TaqManITSGB651L R ITS BHQ1~Q] [6~FAM]CCGTGGACGCGCACATGAGCGAGTAC Taq195eli R Elicitin [BHQ1~Q] a F: Forward primers, R: Reverse primers b Bakkeren et al. 2000 c Mazzola et al. 2002 d Polymorphisms with the most closely related species are underlined. Bases with Locked Nucleic Acid modification are in bold box. e BHQ1~Q: Black hole quencher-Quencher. TaqMan581L R
61
Table 3. Isolates of Phytophthora species sequenced in this study and GenBank accession numbers for β-tubulin and elicitin genes Collection GenBank accession numbera b Species β-tubulin Elicitin number Phytophthora ramorum DAOM 229466 AY766201 _ Phytophthora ramorum DAOM 230729 AY766202 _ DAOM 230728 AY766203 AY766222 Phytophthora ramorumb Phytophthora ramorum DAOM 230727 AY766204 _ Phytophthora ramorum CBS 109279 AY766205 _ Phytophthora ramorum CBS 109278 AY766206 _ Phytophthora ramorum CBS 101554 AY766207 _ Phytophthora ramorum CBS 101551 AY766208 _ Phytophthora ramorum CBS 101550 AY766209 _ Phytophthora ramorum CBS 101548 AY766210 _ Phytophthora ramorum CBS 101332 AY766211 _ Phytophthora ramorum CBS 101330 AY766212 _ Phytophthora ramorum CBS 101327 AY766213 _ Phytophthora ramorum CBS 101552 AY766214 _ Phytophthora ramorum CBS 101326 AY766215 _ Phytophthora ramorum adc 01.01 AY766216 _ Phytophthora ramorum adc 01.06 AY766217 _ Phytophthora ramorum CBS 101553 AY766218 AY766223 Phytophthora ramorum CBS 110537 _ AY766231 Phytophthora ramorum CBS 110541 _ AY766232 Phytophthora ramorum CBS 110545 _ AY766233 Phytophthora ramorum CBS 110900 _ AY766234 Phytophthora ramorum CBS 110901 _ AY766235 Phytophthora lateralis CBS 168.42 AY766219 AY766225 Phytophthora cactorum CBS 108.09 AY766220 _ Phytophthora cinnamomi CBS 144.22 AY766221 _ Phytophthora lateralis ATCC 201856 _ AY766224 Phytophthora cactorum DAOM BR 675 _ AY766226 Phytophthora cinnamomi DAOM BR 680 _ AY766227 Phytophthora infestans CBS 366.51 _ AY766228 Phytophthora citricola DAOM BR 681 _ AY766229 a Alignments are available for those sequences as a popset on Genbank -Indicates no sequence data. b CBS: Centraalbureau voor Schimmelcultures; DAOM = Canadian Agriculture and Agri-Food Canada, Ottawa;adc = Arthur W.A.M. de Cock, ATCC = American Type Culture Collection
62
Table 4. Number of cycles before fluorescence is detected in Phytophthora ramorum isolates tested by real-time PCR using three different gene regions (ITS, β-tubulin and elicitin) with TaqMan Cta CBS number
Origin
ITS
Isolate number Mean Ctb
Standard error
Elicitin Mean Ctb
Standard error
β –tubulin Mean Ctb
Standard error
101327 Netherlands PD 93/56 17.71 0.58 19.05 0.04 23.41 0.57 101326 Netherlands PD 98/8/6933 19.61 0.69 20.54 0.04 24.90 0.72 110538 California Pr65 19.26 1.61 19.82 0.24 23.76 0.62 110601 California Pr84-sz 26.57 0.89 26.04 0.31 30.14 0.79 110543 Oregon Pr159 17.23 0.17 19.17 0.35 23.07 0.71 110537 California Pr52 18.92 0.51 21.29 0.23 24.75 0.56 110541 California Pr86 21.10 0.26 21.36 0.37 24.87 0.67 110539 California Pr70 18.27 0.68 18.86 1.50 23.35 0.19 109279 Germany BBA 13/99-1 18.80 0.55 19.29 0.18 23.98 0.49 101554 Germany BBA 2/4 19.52 0.94 19.84 0.20 24.21 1.19 110545 Poland Rh/2/00 19.67 1.02 19.80 0.44 23.83 0.97 110548 France adc 02.09 27.86 0.74 27.36 0.42 31.61 0.60 101553 Germany BBA 9/95 Type 20.66 0.51 21.05 0.17 25.18 0.61 101329 Netherlands PD 98/8/6285 16.91 0.57 18.71 0.24 22.73 0.48 101551 Germany BBA 12/98 19.39 0.42 19.38 0.38 23.66 0.72 101332 Netherlands PD 94/844 17.78 0.86 18.48 0.26 22.56 0.72 101330 Netherlands PD 98/8/5233 18.65 0.62 19.63 0.19 24.17 0.76 110546 Poland Rh/6/00 27.71 1.20 26.61 0.32 31.37 1.22 101548 Germany BBA 69082 18.44 1.24 19.00 0.33 22.74 1.06 101550 Germany BBA 14/98-a 19.93 0.89 20.87 0.24 24.72 0.83 110544 California PrPRJL3.5.3 18.19 1.17 20.13 0.29 23.72 0.87 109278 Germany BBA 16/99 19.67 0.70 20.40 0.39 24.59 0.61 101552 Germany BBA 9/3 18.97 0.61 17.94 0.15 22.04 0.60 101549 Germany BBA 104/5 19.80 1.05 19.81 0.19 24.31 0.88 101331 Netherlands PD 98/8/2627 17.55 0.70 18.44 0.15 22.90 0.84 110535 California Pr03 19.40 0.51 19.92 0.06 23.07 0.67 110547 Poland Rh/122/98 26.06 1.01 25.67 0.11 30.20 0.88 110536 California Pr04 19.04 0.75 20.75 0.76 23.58 0.79 110542 California Pr110 19.47 0.40 20.47 0.48 24.12 0.77 110534 California Pr01 18.39 0.91 19.84 0.46 23.39 0.70 Average Ct by TaqMan 20.02 2.97 20.65 2.460 24.70 2.57 a Ct Number of cycles before fluorescence threshold is reached. The threshold was never reached for P. lateralis and 64 other Phytophthora species (data not shown). b Mean Ct was calculated with triplicates and standard error was obtained.
63 Table 5. Detection of Phytophthora ramorum in infected plant material by real-time PCR using three gene regions with TaqMan and comparison with other methods Cta TaqMan TaqMan Samples Host C PARP-V8 isolation coxI/IIb βITS Elicitin tubulina PU2 Bay laurel * Negative Negative 39.58 None None PU3 Bay laurel * Negative Negative None None None PU4 Bay laurel * Negative Negative None None None PU5 Bay laurel * Negative Negative None None None PU6 Bay laurel * Negative Negative None None None PU8 Tan oak * Negative Negative None None None PU9 Tan oak * Negative Negative None None None PU10 Tan oak * Negative Negative None None None PU11 Tan oak * Negative Negative None None None California coffee PU12 Negative Negative None None None berry * California coffee PU13 Negative Negative None None None berry * California coffee PU14 Negative Negative None None None berry* PU15 Bay laurel * Negative Negative None None None PU21 Bay laurel * Negative Negative None None None PU25 Bay laurel * Negative Negative None None None PU26 Bay laurel * Negative Negative None None None PU27 Bay laurel * Negative Negative None None None PU29 Bay laurel * Negative Negative None None None PU30 Tan oak * Negative Negative None None None PU31 Tan oak * Negative Negative None None None PU33 Tan oak * Negative Negative None None None PU36 Tan oak * Negative Negative None None None PU38 Tan oak * Negative Negative None None None PU39 Christmas berry * Negative Negative None None None PU18 Bay laurel * Positive Positive 36.04 32.83 30.39 PU19 Bay laurel * Positive Positive 38.47 36.59 33.69 PU20 Bay laurel * Positive Positive 36.57 33.37 30.29 PU28 Bay laurel * 33.56 30.66 Positive Positive 38.37 PU32 Tan oak * Positive Positive 34.10 29.57 29.00 PU1 Madrone * Negative Positive 28.45 23.59 22.70 PU16 Bay laurel * Negative Positive None 39.21 33.76 PU17 Bay laurel * Negative Positive 37.43 36.01 31.54 PU22 Bay laurel * Negative Positive 39.56 39.52 34.45 PU23 Bay laurel * Negative Positive 38.80 34.68 32.27 PU24 Bay laurel * Negative Positive 36.27 28.26 30.57 PU34 Tan oak * Negative Positive 40.61 38.48 31.81 PU35 Tan oak * Negative Positive 39.96 39.64 31.28 PU37 Tan oak * Negative Positive 40.55 37.50 32.09 Controls and spiked samples PU40 Live oak ** Correct 28.29 22.77 21.40 Spiked with 1.3 ng/ul d PU41 Live oak ** Correct 30.33 24.11 22.48 Spiked with 0.65 ng/ul d PU44 Live oak ** Correct 33.52 27.64 25.64 Spiked with 121 pg/ul d PU43 Live oak ** Correct 32.24 24.56 25.35 Spiked with 62 pg/ul d
64 Live oak ** Spiked with 4.3 ng/ul e Correct None None None No Plant DNA ** P. ramorum CBS 101553 Correct 25.88 19.33 19.10 Camellia ** Correct 33.85 20.25 19.12 Spiked with 3.75 ng/ul d Camellia ** Correct 32.28 28.24 25.90 Spiked with 125 pg/ul d Camellia ** Not spiked Correct None None None Camellia ** Not spiked Correct None None None Positive control, PC1 P. ramorum CBS 29.05 25.13 22.27 101553 P. lateralis CBS Plat >41.00 None >33.00 168.42 Negative control, no NC None None None template DNA a Threshold was set at 0.009 in the Opticon Monitor software. b Host scientific names: bay laurel, Umbellularia californica; California coffee berry, Rhamnus californica; camellia, Camellia japonica; Christmas berry, Heteromeles arbutifolia; live oak, Quercus agrifolia; madrone, Abutus menziesii; and tan oak, Lithocarpus densiflorus. *, sample origin: Monterey County, CA, Big Sur area; **, sample origin: Monterrey County, CA, USDA-ARS Salinas station. c Bold positives samples, confirmed by sequence analysis of the spacer region between the coxI and coxII genes. d Spiked with P. ramorum CBS 101553; regression equations derived from plotting Ct versus Log(concentration) in spiked samples: β-tubulin, y = –3.46x + 29.64, r2 = 0.90; ITS, y = –3.60x + 28.82, r2 = 0.96; and elicitin, y = –3.41x + 21.71, r2 = 0.98. e Spiked with P. pseudosyringae EH P96. PU42 PU45 PU46 PU47 PU48 PU49
65
Figure 1. Real-time PCR of Phytophthora species assayed with TaqMan (■), molecular beacons (●) and SYBR Green (▲), targeting the β-tubulin gene. The Phytophthora species used were: P. ramorum 230928, P. ramorum 230991; all other species used are P. lateralis ATCC 201856, P. cactorum BR675, P. cinnamomi BR680, P. citricola BR681, P. infestans M0014 and a negative control (water) and are represented by the ×. Threshold was set at 0.009, with a cycle range of 1-15.
66
Figure 2. Standard curves based on dilution of Phytophthora ramorum DNA for TaqMan assays targeting ITS, β-tubulin, and elicitin. Plots represent Cycle threshold (Zeller et al.) versus the Log of DNA concentration. P. ramorum DNA was diluted from concentrations of 1 ng/ul to 1fg/ul in Tris buffer, pH 8.0, 10 mM. Assays were as follows: ■ β-tubulin TaqMan ▲ elicitin TaqMan, and
ITS TaqMan.
67
I.7 References APHIS. (2004). Invasive Species and Pest Management Programs (ISPM), USDA APHIS PPQ Approved Laboratory Diagnostic Protocols, PCR. Protocol. Published online http://www.aphis.usda.gov/ppq/ispm/pramorum/pdf_files/pcrprotocol4.pdf. Arezi B., W.M. Xing, J.A. Sorge, and H.H. Hogrefe (2003) Amplification efficiency of thermostable DNA polymerases. Anal. Biochem. 321, 226-235. Ayliffe M.A., P.N. Dodds, and G.J. Lawrence (2001) Characterisation of a [beta]-tubulin gene from Melampsora lini and comparison of fungal [beta]-tubulin genes. Mycol. Res. 105, 818-826. Bakkeren G., J.W. Kronstad, and C.A. Levesque (2000) Comparison of AFLP fingerprints and ITS sequences as phylogenetic markers in Ustilaginomycetes. Mycologia 92, 510. Bilodeau G., C. A. Lévesque, A. W. A. M. de Cock, G. Kristjansson, , J. McDonald, and R. C. Hamelin (2003) Detection and identification of Phytophthora ramorum, the causal agent of sudden oak death by real-time PCR. Phytopathology 93, S8. Bonnet G., S. Tyagi, A. Libchaber, and F.R. Kramer (1999) Thermodynamic basis of the enhanced specificity of structured DNA probes. Proc. Natl. Acad. Sci. USA 96, 6171. Braasch D.A., and D.R. Corey (2001) Locked nucleic acid (LNA): fine-tuning the recognition of DNA and RNA. Chem. Biol. 8, 1-7. Brasier C. (2003) Sudden oak death: Phytophthora ramorum exhibits transatlantic differences. Mycol. Res. 107, 258-259. Brasier C., and S. Kirk (2004) Production of gametangia by Phytophthora ramorum in vitro. Mycol. Res. 108, 823-827. Brasier C.M., S. Denman, J. Rose, S. A. Kirk, K. J. D. Hughes, R. L. Griffin, C. R. Lane, A. J. Inman, and J. F. Webber (2004) First report of ramorum bleeding canker on Quercus falcata, caused by Phytophthora ramorum. Plant Pathol. 53, 804. Cooke D.E.L., A. Drenth, J.M. Duncan, G. Wagels, and C.M. Brasier (2000) A molecular phylogeny of Phytophthora and related oomycetes. Fungal Genet. Biol. 30, 17. Davidson J.M., M.Garbelotto, S. T. Koike, and D. M. Rizzo (2002a) First report of Phytophthora ramorum on Douglas-fir in California. Plant Dis. 86, 1274.
68 Davidson J.M., D. M. Rizzo, M.Garbelotto, S. Tjosvold, and G. W. Slaughter (2002b) Phytophthora ramorum and sudden oak death in California: II. Transmission and survival. . USDA For. Serv. Gen. Tech. Rep. PSW-GTR-184, 741-749. Davidson J.M., S. Werres, M.Garbelotto, E.M. Hansen, and D.M. Rizzo (2003) Sudden oak death and associated diseases caused by Phytophthora ramorum. Plant Health Prog., 1-21. De Cock A.W.A.M., A. Neuvel, G. Bahnweg, J.C.J.M. de Cock, and H.H. Prell (1992) A comparison of morphology, pathogenicity and restriction fragment patterns of mitochondrial DNA among isolates of Phytophthora porri Foister. Neth. J. Plant Pathol. 98, 277-289. De Merlier D., A. Chandelier, and M. Cavelier (2003) First report of Phytophthora ramorum on Viburnum bodnantense in Belgium. Plant Dis. 87, 203. Denman S., S. A. Kirk, C. M. Brasier, V. C. Barton, K. J. D. Hughes, and J. F. Webber (2005) Phytophthora ramorum on Quercus ilex in the United Kingdom. Plant Dis. 89, 1241-1241. Garbelotto M. (2003a) Molecular diagnostics of Phytophthora ramorum, causal agent of sudden oak death. In: Sudden oak death: how concerned should you be? An international symposium, 21 April-12 May, 2003, Online Symposium. American Phytopathological Society Online symposium. Garbelotto M., J. M. Davidson, K. Ivors, P. E. Maloney, D. Huberli, S. T. Koike, and D.M. Rizzo (2003b) Non-oak native plants are main hosts for sudden oak death pathogen in California. Calif Agric. 57, 18-23. Garbelotto M., P. Svihra, and D.M. Rizzo (2001) Sudden oak death syndrome fells three oak species. Calif. Agric. 55, 1-19. Giglio S., P.T. Monis, and C.P. Saint (2003) Demonstration of preferential binding of SYBR Green I to specific DNA fragments in real-time multiplex PCR. Nucleic Acids Res. 31 (22), DOI: 10.1093/nar/gng135. Goheen E.M., E. M. Hansen, A. Kanaskie, M. J. McWilliams, N. Osterbauer, and W. Sutton (2002) Sudden oak death caused by Phytophthora ramorum in Oregon. Plant Dis. 86, 441. Hamelin R., P. Berube, M. Gignac, and M. Bourassa (1996) Identification of root rot fungi in nursery seedlings by nested multiplex PCR. Appl. Environ. Microbiol. 62, 40264031. Hamelin R.C., M. Bourassa, J. Rail, M. Dusabenyagasani, V. Jacobi, and G. Laflamme (2000) PCR detection of Gremmeniella abietina, the causal agent of scleroderris canker of pine. Mycol. Res. 104, 527-532.
69
Hansen E.M., P.W. Reeser, Sutton W., and L.M. Winton (2003) First report of A1 mating type of Phytophthora ramorum in North America. Plant Dis. 87, 1267-1267. Hayden K., K. Ivors, C. Wilkinson, and M.Garbelotto (2006) TaqMan chemistry for Phytophthora ramorum detection and quantification, with a comparison of diagnostic methods. Phytopathology 96, 846-854. Hayden K.J., D. Rizzo, J. Tse, M.Garbelotto (2004) Detection and quantification of Phytophthora ramorum from California forests using a real-time polymerase chain reaction assay. Phytopathology 94, 1075-1083. Heid C., J. Stevens, K. Livak, and P. Williams (1996) Real time quantitative PCR. Genome Res. 6, 986-994. Ivors K., M. Garbelotto, I. D. E. Vries, C. Ruyter-Spira, B. T. Hekkert, N. Rosenzweig, and P. Bonants (2006) Microsatellite markers identify three lineages of Phytophthora ramorum in US nurseries, yet single lineages in US forest and European nursery populations. Mol. Ecol. 15, 1493. Ivors K.L., K.J. Hayden, P.J.M. Bonants, D.M. Rizzo, and M.Garbelotto (2004) AFLP and phylogenetic analyses of North American and European populations of Phytophthora ramorum. Mycol. Res. 108, 378-392. Jeffers S.N., and S.B.Martin (1986) Comparison of 2 media selective for Phytophthora and Pythium species. Plant Dis. 70, 1038-1043. Jullien J., and C. Casset (2003) Phytophthora ramorum. Legitimate concern in France for several ornamental nursery species. Phytoma. No.561, 16-20. Kroon L., E.C.P. Verstappen, L.F.F. Kox, W.G. Flier, and P.J.M. Bonants (2004a) A rapid diagnostic test to distinguish between American and European Populations of Phytophthora ramorum. Phytopathology 94, 613. Kroon L.P.N.M., F.T. Bakker, G.B.M. van den Bosch, P.J.M. Bonants, and W.G. Flier (2004b) Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences. Fungal Genet. Biol. 41, 766-782. Lane C.R., P.A. Beales, K.J.D. Hughes, R. L. Griffin, D. Munro, C. M. Brasier, and J. F. Webber (2003) First outbreak of Phytophthora ramorum in England, on Viburnum tinus. Plant Pathol. 52, 414. Maloney P.E., D.M. Rizzo, S.T. Koike, T.Y. Harnik, and M. Garbelotto (2002) First report of Phytophthora ramorum on coast redwood in California. Plant Dis. 86, 1274.
70 Martin F.N., and P.W. Tooley (2004) Identification of Phytophthora isolates to species level using restriction fragment length polymorphism analysis of a polymerase chain reaction-amplified region of mitochondrial DNA. Phytopathology 94, 983991. Martin F.N., P.W. Tooley, and C. Blomquist (2004) Molecular detection of Phytophthora ramorum, the causal agent of sudden oak death in California, and two additional species commonly recovered from diseased plant material. Phytopathology 94, 621631. Martin R., D. James, and C. Levesque (2000) Impacts of molecular diagnostic technologies on plant disease management. Annu. Rev. Phytopathol. 38, 207-239. Mazzola M., P. Andrews, J. Reganold, and C. Levesque (2002) Frequency, virulence, and metalaxyl sensitivity of Pythium spp. isolated from apple roots under conventional and organic production systems. Plant Dis. 86, 669-675. McCartney H., S. Foster, B. Fraaije, and E. Ward (2003) Molecular diagnostics for fungal plant pathogens. Pest Manag. Sci. 59, 129-142. Moller E.M., G. Bahnweg, H. Sandermann, and H.H. Geiger (1992) A Simple and Efficient Protocol for Isolation of High-Molecular-Weight DNA from Filamentous Fungi, Fruit Bodies, and Infected-Plant Tissues. Nucleic Acids Res. 20, 6115-6116. Moralejo E., and S. Werres (2002) First report of Phytophthora ramorum on Rhododendron sp. in Spain. Plant Dis. 86, 1052. Murphy S.K., and D.M.Rizzo (2003) First report of Phytophthora ramorum on canyon live oak in California. Plant Dis. 87, 315. Orlikowski L.B., and G. Szkuta, (2002) First record of Phytophthora ramorum in Poland. Phytopathol. Polonica. 25, 69-79. Osterbauer N., and A. Trippe (2005) Comparing diagnostic protocols for Phytophthora ramorum in rhododendron leaves. Plant Health Prog. http://www,plantmanagementnetwork.org/pub/php/brief/2005/Pramorum Osterbauer N.K., J.A. Griesbach, and J. Hedberg (2004) Surveying for and eradicating Phytophthora ramorum in agricultural commodities. In: Plant Health Prog. Ponchet M., F. Panabières, M.-L.Milat, V. Mikes, J.-L. Montillet, L. Suty, C. Triantaphylides, Y. Tirilly, and J. P. Blein (1999) Are elicitins cryptograms in plantoomycete communications? Cell. Mol. Life Sci. 56, 1020-1047. Prospero S., J.A. Black, and L.M. Winton (2004) Isolation and characterization of microsatellite markers in Phytophthora ramorum, the causal agent of sudden oak death. Mol. Ecol. Notes 4, 672-674.
71
Rizzo D.M., M. Garbelotto, J.M. Davidson, G.W. Slaughter, and S.T. Koike (2002) Phytophthora ramorum as the cause of extensive mortality of Quercus spp. and Lithocarpus densiflorus in California. Plant Dis. 86, 205-214. Schaad N., Frederick R. (2002) Real-time PCR and its application for rapid plant disease diagnostics. Can. J. Plant Pathol. 24, 250-258. Tomlinson J.A., N. Boonham, K.J.D. Hughes, R.L. Griffen, and I. Barker (2005) On-site DNA extraction and real-time PCR for detection of Phytophthora ramorum in the field. Appl. Environ. Microbiol. 71, 6702. Tooley P.W., F.N. Martin, M.M. Carras, and R.D. Frederick (2006) Real-time fluorescent polymerase chain reaction detection of Phytophthora ramorum and Phytophthora pseudosyringae using mitochondrial gene regions. Phytopathology 96, 336. Tyagi S., and F. Kramer (1996) Molecular beacons: Probes that fluoresce upon hybridization. Nature biotech. 14, 303-308. Tyagi S., S.A.E. Marras, J.A.M. Vet, and F.R. Kramer (2000) Molecular beacons: hybridization probes for detection of nucleic acids in homogeneous solutions. In: Nonradioactive Analysis of Biomolecules (ed. Springer-Verlag), pp. 606-616. C. Kessler, Berlin. Weerakoon N.D., J.K. Roberts, L.P. Lehnen, J.M. Wilkinson, J.S. Marshall, and A.R. Hardham (1998) Isolation and characterization of the single beta-tubulin gene in Phytophthora cinnamomi. Mycologia 90, 85-95. Werres S. (2002) A new species of Phytophthora causes Rhododendron dieback. Rhododendron und immergrune Laubgeholze Jahrbuch, 61-67. Werres S., and D. De Merlier (2003) First detection of Phytophthora ramorum mating type A2 in Europe. Plant Dis. 87. Werres S., R. Marwitz, W.A.Man-in'-t-Veld, A.W.A.M. de Cock, P.J. Bonants (2001) Phytophthora ramorum sp. nov., a new pathogen on Rhododendron and Viburnum. Mycol. Res. 105, 1155-1165. White T.J., T. Bruns, S. Lee, and J. Taylor (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: PCR Protocols. A Guide to Methods and Applications. (ed. M. A. Innis DHG, J. J. Sninsky, and T. J. White), pp. p. 315-322. Academic Press, Inc. San Diego. Winton L.M., and E. M. Hansen (2001) Molecular diagnosis of Phytophthora lateralis in trees, water, and foliage baits using multiplex polymerase chain reaction. Forest Pathol. 31, 275-283.
72
Chapitre II
Multiplex Real-Time PCR for Detection of Phytophthora ramorum, the Causal Agent of Sudden Oak Death.
73
Multiplex Real-Time PCR for Detection of Phytophthora ramorum, the Causal Agent of Sudden Oak Death. G. J. Bilodeau1, G. Pelletier1, F. Pelletier1, C. A. Lévesque2, and R. C. Hamelin1 1
Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, 1055 du
P.E.P.S., P.O. Box 10380, Stn. Sainte-Foy, Québec, Qc G1V 4C7, Canada 2
Agriculture and Agri-Food Canada, Central Experimental Farm - Biodiversity, 960 Carling
Avenue, Ottawa, On K1A 0C6, Canada
Ces travaux ont été soumis à la revue : Applied and Environmental Microbiology en juin 2008
74
II.1 Résumé/Abstract
II.1.1 Résumé Depuis 1995, le Phytophthora ramorum cause une maladie appelée ; la mort subite du chêne (l’encre des chênes rouges) en Californie et en Oregon, affectant des dizaines de milliers d’arbres. Cet agent pathogène peut infecter jusqu’à maintenant plus d’une centaine d’espèces de plantes, particulièrement en pépinières. Des mesures de quarantaine sont mises en place, il est donc très important pour les agences réglementaires de détecter rapidement et efficacement cet organisme avant qu’il ne cause de sérieux problèmes. Des tests de détection moléculaire ont été développés au cours des dernières années. La redondance en utilisant plusieurs régions géniques du P. ramorum a démontré une augmentation de l’habileté à détecter l’agent pathogène. Cependant, ces tests nécessitent d’être effectués en trois réactions séparées à des temps différents. Afin de régler ce problème et d’améliorer le diagnostic du P. ramorum, les trois sondes TaqMans ont été multiplexées avec une quatrième sonde TaqMan, spécifique au genre Phytophthora en une seule réaction. Une deuxième réaction multiplexe pour détecter les oomycètes et les plantes a été conçue et testée en conjonction avec les sondes TaqMan P. ramorum ITS et β-tubuline au genre Phytophthora permettant ainsi un contrôle sur l’extraction de l’ADN sur l’ADN de plante ou d’oomycètes. Ces réactions multiplexes ont été testées sur différentes espèces de Phytophthora et sur deux sortes d’échantillons environnementaux avec préalablement testés par d’autres laboratoires. Les échantillons environnementaux testés proviennent d’ADN extraits de plusieurs plantes hôtes infectées par différents Phytophthora ainsi que par de l’ADN extraits de lysat d’ELISA. Tous les échantillons du P. ramorum ont été détectés utilisant ces tests multiplexes que ce soit par ADN provenant de cultures pures ou du terrain. En général, les sondes présentaient une plus faible sensibilité quand la PCR était effectuée en réaction multiplexe plutôt que dans des réactions séparées. Par contre, cette réaction permet de détecter avec confiance à plus faible coût, un plus haut débit d’analyses du P. ramorum.
75
II.1.2 Abstract Since 1995, Phytophthora ramorum has been causing desease called; sudden oak death in California and Oregon, affecting tens of thousands of oak trees over large areas and infecting more than one hundred other plant species. Quarantine measures are in effect, and regulatory agencies need to detect this organism in a timely and efficient matter. Various molecular assays have been developed over the past few years, and redundancy using multiple gene regions of P. ramorum was shown to increase the reliability in detecting the pathogen. However, such multi-gene assays require different PCR reactions to test a single sample. To improve P. ramorum detection, three different TaqMan assays were multiplexed with a fourth TaqMan specific to the Phytophthora genus in one single reaction. A second multiplex TaqMan PCR assay to detect oomycetes and give a positive PCR reaction in the presence of host DNA was also designed and tested in conjunction with the P. ramorum ITS and Phytophthora genus TaqMan assays. These assays were tested on different Phytophthora species, and were verified on two different sets of field samples previously assayed by other laboratories. These were obtained from multiple field hosts infected by various Phytophthora species and the DNA from one set was extracted from ELISA lysates. All known P. ramorum samples from pure cultures or field samples were detected using this multiplex real-time PCR assay. In general, TaqMan multiplex assays showed lower detection sensitivity than single separated reactions. However, the multiplex assays still detected P. ramorum accurately while decreasing the cost and increasing throughput.
76
II.2 Introduction Since the mid-1990s, Phytophthora ramorum has been found on a large number of herbaceous hosts and trees and is causing severe damage on red oaks in California and Oregon (Davidson et al. 2003; McPherson et al. 2005; Werres et al. 2001b). During the same period, the pathogen has been found to infect Rhododendron and Viburnum in nurseries in Europe (Werres et al. 2001b). Through pathogenicity tests it can infect more than 110 plant species, causing a variety of symptoms ranging from bleeding cankers to leaf spots, twig and leave blight, and (shoot) dieback (Rioux et al. 2006; USDA-APHIS 2008). After its discovery in California and Germany and The Netherlands, this pathogen was found in nurseries of several countries and quarantine measures were put into place to limit its spread. Enforcement of quarantine regulations relies on inspection and reliable diagnosis of the disease. However, diagnosis of the disease can be challenging partly because of the various symptoms induced on the different hosts and because of the lack of sensitivity of culture-based methods. This can lead to high levels of false negatives, the worse type of error form in assays on quarantined pathogens than can escape in the environment.
Although P. ramorum possesses some distinguishing features (e.g., its unique combination of semipapillate, deciduous sporangia with short pedicels and high length:width ratio; large chlamydospores abundant on agar; relatively slow growth and low cardinal temperatures for growth), identification of P. ramorum based solely on morphology can be too risky for quarantine regulation enforcement (Werres et al. 2001b). Some closely related species may have overlapping morphological characters that can make species identification challenging even for experts in a particular group. To aid and support diagnosis and identification, multiple methods have been developed and several are currently in use by the regulatory and research communities. Methods based on PCR and real-time PCR have been particularly helpful to detect Phytophthora ramorum rapidly and
77 with high sensitivity, thereby providing a higher confidence level in the detection and identification of the pathogen (Bilodeau et al. 2007b; Hayden et al. 2006; Hughes et al. 2006a; Hughes et al. 2006b; Schena et al. 2006; Tomlinson et al. 2007; Tomlinson et al. 2005b; Tooley et al. 2006). Some PCR assays target a single gene region of P. ramorum, while other complementary assays were developed against Phytophthora spp., oomycetes or against host plants (Hayden et al. 2006; Hughes et al. 2006a; Hughes et al. 2006b; Tomlinson et al. 2005b; Tooley et al. 2006; Winton and Hansen 2001) to increase the confidence in DNA extraction methods. TaqMan, molecular beacon and SYBRGreen realtime PCR assays that are P. ramorum-specific were developed to target three different gene regions (ITS, β-tubulin and elicitin). The combination of these assays resulted in increased redundancy and therefore in increased confidence in the detection and identification of the pathogen from cultures as well as from field samples (Bilodeau et al. 2007b).
To compare the performance of these and other methods, a blind ring-trial validation of culture and field samples was conducted for several P. ramorum detection assays (Martin et al. 2006; Martin et al. 2008). Several of the protocols were comparable in their rates of false positives and false negatives. However, combining multiple gene targets increased redundancy and made it possible to improve significantly reliability and confidence. Since these assays were used in simplex reactions (i.e., one probe-target at a time), each sample had to be processed multiple times. Multiplexing these assays in a single reaction would be advantageous because it would reduce the cost of reactions, while allowing high-throughput sample processing. TaqMan assays can be multiplexed because of the possibility to use different fluorophores with emission and detection spectra that are compatible. However, multiplexing is complicated by the fact that a large number of oligonucleotides (for PCR and probes) need to be included in the reaction (Ma and Michailides 2007).
Having a reliable, sensitive, specific, fast and low cost molecular diagnostic assay is highly desirable. In order to improve reliability and specificity, while not compromising speed and cost, we propose the development of Real Time PCR (RT-PCR) assays targeting multiple gene regions of P. ramorum as well as higher taxonomical levels and plant genes.
78 The objectives of this research was to create multiplex RT-PCR using TaqMan probes with different reporter dyes targeting three different gene regions of P. ramorum (ITS, β-tubulin, elicitin), Phytophthora genus (β-tubulin), oomycetes (ribosomal 5.8S subunit) and host plants (RuBisCO) allowing simultaneous permitting the simultaneous detection of P. ramorum, while verifying DNA extraction and the presence of other oomycetes in the DNA sample. The sensitivity of TaqMan assays in single and multiplex reactions was also compared.
II.3 Materials and methods
II.3.1 Isolates from culture collection All isolates of Phytophthora used in this study are listed in Table 1. Phytophthora ramorum isolates used as positive control were CBS 101328 and CBS 101332. The Phytophthora collection included isolates from 22 different host species and varieties (Martin et al. 2006; Martin et al. 2008), representing most of the recognized Phytophthora species currently available in pure culture. For CBS cultures, DNA was extracted following the procedures described by de Cock (de Cock et al. 1992b) or Möller (Möller et al. 1992).
II.3.2 Isolates from field To test sensitivity and the specificity of the assays on infected plant material, environmental samples were obtained from different hosts that were either healthy or naturally infected by the pathogen. Leaf disks that were diagnosed with P. ramorum or other Phytophthora species with PARP medium selective isolation, and by ELISA for the presence of Phytophthora spp., were tested. Cultures obtained were identified by DNA sequencing of the ITS region using primers ITS1 and ITS4 (White et al. 1990). For assay validation, a broad collection of samples was obtained from the World Phytophthora
79 Genetic Resource Collection (WPGRC) at the University of California, Riverside (http://phytophthora.ucr.edu), and processed as described by Martin et al. (Martin et al. 2006).
II.3.3 DNA isolation from ELISA lysates DNA samples extracted from ELISA lysates from the CFIA (Canadian Food Inspection Agency) 2006 P. ramorum survey were also used in these multiplex assays. The DNA extraction procedure for the ELISA lysates is described in Bilodeau et al. (Bilodeau et al. 2007a). Rhododendron DNA was extracted using the same procedure but was grinded using Christison M3 Mixermill (tungsten bead, 2X 2 min and 2X 30 sec) and Plant DNA easy mini kit (Qiagen, Valencia, CA), according to the manufacturer’s instructions.
II.3.4 Design of primers and probes Phytophthora ramorum primers and probes (Table 2) used in the multi-gene P. ramorum multiplex real-time PCR were the same as those used previously (Bilodeau et al. 2007b) excepted for ITS and elicitin TaqMan that have different reporter dyes and quenchers to improve compatibility. For the hierarchical multiplex, primers and probes targeting Phytophthora genus, oomycetes and plants were designed de novo.
The set of primers targeting Phytophthora spp. was designed using β-tubulin sequences (Kroon et al. 2004). The oomycete primers and probes were created using an alignment of 5.8S sequences from oomycetes along with some ascomycetes, basidiomycetes and other fungi obtained from GenBank (NCBI) (Supplemental Table 1). Primers and probes targeting plants were developed in the RuBisCO (Ribulose-1,5bisphosphate carboxylase/oxygenase large subunit) region with an alignment of different plant RuBisCO regions (Supplemental Table 2). RuBisCO is an enzyme present in plant
80 chloroplasts that allows carbon sequestration and fixation, and it is the most common protein on earth.
In cases where only single nucleotide differences were present and unmodified primers did not allow specific amplifications, primers were synthesized with a Lock Nucleic Acid (LNA) (Braasch and Corey 2001) (Integrated DNA Technologies, Coralville, IA) to increase specificity by using higher annealing temperatures. The TaqMan probes used were modified with different compatible quenchers and reporters to obtain a multiplex reaction (see Table 2 for details).
II.3.5 Multi-gene P. ramorum multiplex, targeting Phytophthora ramorum and Phytophthora genus. Real-time PCR was performed with a Chromo4 Real-Time PCR Detection System that possesses four channels (Bio-Rad Laboratories Inc., Hercules, CA). This multiplex reaction for P. ramorum was also tested on a Stratagene Mx3000P system (not shown), but amplifications with the β-tubulin and elicitin assays were inconsistent with those of ITS P. ramorum and β-tubulin-all Phytophthora genus (data not shown). Fluorescent molecules dual-labelled probes (TaqMan®) were added to the PCR master mix (QuantiTect Multiplex PCR KitsTM, without Rox, Qiagen Inc., Valencia, CA). All reactions were performed in 25 µl volumes. The DNA concentrations used in the reaction were determined by a Nanodrop spectrophotometer ND-1000 (Nanodrop Technologies, Wilmington, DE) and ranged from 0.3 to 3.0 ng/µl for P. ramorum and from 0.1 and 5.0 ng/ul for other Phytophthora species. Phytophthora ramorum CBS 101332 was used as positive control and reactions without template DNA were used as negative controls.
PCR reactions contained different concentrations of each of eight different LNA primers or regular primers: ITS 0.3 µM, β-tubulin_all_sp (β-tubulin-all Phytophthora genus) 0.3 µM, elicitin 0.3 µM, β-tubulin 0.6 µM; and 0.2 µM for the concentration of the
81 four TaqMan probes (Table 2), 1X of QuantiTect Multiplex PCR Master Mix (Qiagen, Valencia, CA), and template DNA. PCR cycling conditions were set at 95°C for 15 min for one cycle, 55 cycles at 94°C for 10 s, and 67°C for 2 min. Fluorescence was read during the extension.
The analysis software Opticon Monitor version 3.1.32 (Bio-Rad Laboratories, Hercules, CA) was used to analyze the data. Threshold was fixed at 0.016 for ITS, Elicitin, β-tubulin-all Phytophthora genus, and 0.036 for β-tubulin. Ct values were exported and analyzed for comparison among samples in Excel spreadsheets. Statistical analyses were performed with Excel (Microsoft® Excel VER.9.0.3821 SR-1, Redmond, WA).
II.3.6 Hierarchical multiplex, targeting Phytophthora ramorum, Phytophthora genus, oomycetes and plants Real-time PCR was performed with a Stratagene Mx3000P system (Agilent Technologies, Inc., Santa Clara, CA). Dual-labelled probes (TaqMan®) were added in the PCR master mix (QuantiTect Multiplex PCR KitsTM, without Rox, Qiagen Inc., Valencia, CA). All reactions were performed in 25 µl volumes. The DNA concentrations used in the reaction were also determined by a Nanodrop spectrophotometer ND-1000 and ranged from 0.3 to 3.0 ng/µl for P. ramorum, from 0.1 to 5.0ng/µl for other Phytophthora spp., from 0.5 to 6.3 ng/µl for oomycetes, while plant DNA concentration was 11ng/µl. Phytophthora ramorum CBS 101332 was used as positive control, and reactions without template DNA were used as negative control.
PCR reactions contained the same 0.3 µM concentrations of each LNA primers or regular primers, and the concentrations were 0.2 µM for the four TaqMan probes (Table 2), 1X of QuantiTect Multiplex PCR Master Mix, and template DNA. PCR cycling conditions were set at 95°C for 15 min for one cycle, 55 cycles at 94°C for 10 s, and 65°C for 2 min. A two-step PCR protocol with annealing and extension at 67°C was also tested with all the
82 same primers and probes (data not shown). Fluorescence was read during the extension at 3X reads. Threshold fluorescence values were (dR): CY5 (1000), ROX (1500), HEX (1000) and FAM (2000) to make sure that the baseline correction did not intercept the line twice. Filter gain settings for each of the four detection dyes tested were set up as follows: CY5 8X, ROX 1X, HEX 2X and FAM 8X to allow the best range of reading.
The Stratagene MxPro QPCR Software (Agilent Technologies Inc., Santa Clara, CA) was used to analyze the data. Ct values were exported and analyzed for comparisons among samples in Excel spreadsheets. Statistical analyses were performed with Excel.
II.3.7 Standard curves Serial dilutions of P. ramorum DNA were used to calculate amplification efficiency with TaqMan assays. DNA serial dilutions consisted of P. ramorum strain CBS 101332 with an estimated initial concentration of 11 ng/ul diluted in a series of 1:10 from 1.1 ng/ul to 1.1 fg/ul. Plant DNA extracted from rhododendron leaves was also diluted in series of 1:10 from 11 ng/ul to 1.1 fg/ul to test efficiency and limit of detection. Fluorescence threshold and filter gain setting were the same as previously described. Results were analyzed by plotting the Log of template concentration against cycle threshold values. PCR efficiency was calculated with the formula E = (10
(-1/slope)
– 1)*100, where E is the
amplification efficiency and the slope is derived from the plot of Log template concentration vs Ct. A slope of 3.32 translates into 100% efficiency of amplification.
II.4 Results
II.4.1 Multiplex assay targeting P. ramorum and Phytophthora genus
83 Three P. ramorum specific PCR primer pairs and TaqMan probes targeting βtubulin, elicitin and ITS were multiplexed with a Phytophthora genus (β-tubulin_all_sp) set of probes and primers (Table 3a and b). All annealing temperatures were harmonized to allow multiplexing of the various primers and probes. This multiplex assay was tested with different pure cultures including eight strains of P. ramorum, 19 Phytophthora spp. and two Pythium spp. (Table 3b). All P. ramorum samples tested were positive, with Ct values ranging from 18 to 35 for the P. ramorum specific gene regions (Table 3a). The Phytophthora genus assay was positive for all P. ramorum tested as well (Table 3a). All other Phytophthora and Pythium species tested here were negative for the three P. ramorum-specific probes. The Phytophthora genus TaqMan was positive with all Phytophthora tested but negative with Pythium and the negative control (Table 3b). An important result was that no false negatives were observed when the four assays were multiplexed.
As previously reported (Bilodeau et al. 2007b), the P. ramorum-specific ITSTaqMan did not yield false positives on other Phytophthora species or false negatives on P. ramorum. The elicitin TaqMan yielded some high Ct values for P. erythroseptica, P. fragariae var. fragariae and P. lateralis (Table 3b). Given the high sensitivity of the elicitin TaqMan assay, (Bilodeau et al. 2007b) the threshold limit for elicitin was fixed at Ct >34; therefore these high Ct values were considered negative. The P. ramorum-specific βtubulin TaqMan yielded high Ct values compared with the elicitin and ITS TaqMan in this assay, which translates into lower sensitivity. Four Phytophthora false positives were detected (P. erythroseptica, two P. nicotianae and P. syringae), with Ct values of 31.01, 36.53 and 37.62, and one Pythium was positive, with a Ct value of 28.92.
The Ct values varied in the four PCR assays tested. The ITS TaqMan assay had lower Ct values than the β-tubulin and elicitin TaqMan assays. In increasing order, the Ct values were lower for ITS, elicitin, β-tubulin for P. ramorum and β-tubulin_all_sp. The PCR efficiency of some TaqMan used in the multiplex reaction was reduced compared with single TaqMan reactions. PCR efficiency was 77, 88, 78 and 109% in multiplex reactions (Fig. 1a), and 109, 118, 75, and 89% in simplex (Fig. 1b) for β-tubulin, elicitin, ITS and β-
84 tubulin-Phytophthora spp., respectively. Efficiency was comparable for the ITS TaqMan. The β-tubulin Phytophthora genus TaqMan was apparently more efficient in multiplex than in simplex reactions. Not only did the Ct values change, but so did the appearance of the curves; in multiplex, the curves are not as steep and appear to have a lower plateau on the β-tubulin TaqMan and ITS TaqMan (Fig. 2a and c), and the same goes for the other two TaqMan assays, whether multiplex or simplex (Fig. 2b and d).
II.4.2 P. ramorum and Phytophthora genus multiplex assay on infected host material Multiple hosts infected by different Phytophthora species were analyzed using this multiplex assay (Table 4) to assess the reliability of the multiplex assay to detect P. ramorum in field samples and to determine the rates of false positive and negative results. All P. ramorum samples previously shown to be positive by cultures and DNA sequencing of the ITS region (n=9) were positive for all four TaqMan probes in the multiplex assay reported here, including the Phytophthora genus probe. P. ramorum was also detected in sample #17, which was from an oak with a mixed infection of P. ramorum and P. nemorosa. The two P. ramorum positive controls, including the serial dilution of one of them, were also positive. All plant samples infected with other Phytophthora or organisms as well as the negative control (NTC) yielded negative results for the P. ramorum-specific TaqMan probes. Only nine samples presented possible weak cross reactions with β-tubulin P. ramorum specific markers with Ct values 100 host plants that can be affected by this pathogen in nurseries and in the wild in Europe and North America. Molecular diagnosis and genotyping are important tools to monitor and prevent spread of this pathogen, to enforce quarantines and to support eradication programs. By amplifying and sequencing DNA in a worldwide panel of strains of P. ramorum, we discovered single-nucleotide polymorphisms (SNP) in two genes (β-tubulin and cellulose binding elicitor lectin (CBEL)) that differ in the European and the North American P. ramorum isolates. Using these SNPs in diagnostic assays and distinguishing between strains of different geographic origins, we developed primer extension assays in real-time polymerase chain reaction using allele specific oligonucleotides (ASO). All European isolates genotyped were homozygous at position 279 in the β-tubulin gene (β-tub279: G/G) but heterozygous at position 858 (βtub858: A/G). By contrast, all the California isolates were heterozygous for β-tub279 (T/G), but homozygous for β-tub858 (A/A). For CBEL, all European isolates were heterozygous for the SNPs at position 245 (CBEL245: G/C) and 412 (CBEL412: G/A), but the California samples were homozygous at both SNP loci CBEL245 (G/G) and CBEL412 (G/G). This assay was also used directly on DNA extracted from the ELISA lysates from field samples collected in British Columbia in 2005 by the Canadian Food Inspection Agency. Although most samples had the North American DNA profiles, there were several samples with the European profiles from the nurseries. This finding could provide information on the biology, epidemiology, etiology, host range and populations of P. ramorum from North America and Europe and provide useful tools in P. ramorum surveys.
120
III.2. Introduction Phytophthora ramorum Werres et al. (Werres et al. 2001a) is a plant pathogen that was identified for the first time in North America in 2001, where it is believed to have been introduced. It was observed for the first time in 1994-1995, infecting tanoaks (Lithocarpus densiflorus (Hook. & Arn.) Rehder) and coast live oaks (Quercus agrifolia Née) in Marin County, California (McPherson et al. 2000). It can infect a vast range of hosts, often causing leaf spots that rarely result in mortality (Garbelotto et al. 2003). However, it also causes cankers and extensive mortality on oak and tanoak trees in California and Oregon. In affected areas, mortality rates can be as high as 80% on tanoak in the western U.S. (Hansard 2003b). This pathogen has also been reported in Europe where it causes disease on several hosts mostly found in nurseries. Symptoms similar to those observed in California were observed on oaks (Quercus falcata Michx) in Britain and The Netherlands (Brasier 2003). The disease pattern in Europe and North America is different ─ it is most severe on rhododendrons (Rhododendron spp.) and other related hosts in Europe whereas a severe epidemic has developed on species of the red oak group along the west coast in the U.S., and the disease has been detected in several nurseries in North America (Garbelotto et al. 2003). The pathogenicity of European (EU1) and North American (NA1) isolates of P. ramorum has been studied and compared. Different growth rates and colony morphology were observed between EU1 and NA1 strains, and variable pathogenicity was observed among hosts and isolates (Brasier et al. 2002a; Pogoda and Werres 2002a). In addition, when analyzed, European and North American isolates had different DNA amplified fragment length polymorphism (AFLP) patterns (Ivors et al. 2004) and microsatellite profiles (Garbelotto et al. 2002; Ivors et al. 2006). Although the results did not allow the investigators to determine the origin of the pathogen, more DNA variation was found in the European than in the North American populations, suggesting that a population bottleneck occurred in the latter. In addition, multilocus microsatellite analysis revealed a single new multilocus genotype in the U.S. (Ivors et al. 2006). Multiple morphological variants were also observed in the North American population (Pogoda and Werres 2002a), and it was suggested that the North American and European populations are adaptively differentiated
121 (Brasier 2003). The fact that P. ramorum has been introduced in both North America and Europe and is showing different phenotypes on these continents could be due to different events or processes, namely the introduction of different founder populations or different adaptations or both.
Phytophthora ramorum is a heterothallic organism; it is considered self-sterile and produces oospores only when opposite mating types (A1 and A2) are grown together (Erwin and Ribeiro 1996; Werres and Zielke 2003; Werres and Kaminski 2005). This stimulation of oospore production could also be induced between isolates from different species when opposing mating types are plated together (Erwin and Ribeiro 1996). In the case of P. ramorum, the most successful mating partners tested were Phytophthora cryptogea Pethybr. & Laferty, followed by Phytophthora cinnamomi Rands, Phytophthora drechsleri Tucker and Phytophthora cambivora (Petri) Buisman (Werres and Kaminski 2005). Pairing and mating type studies of P. ramorum showed that all European P. ramorum isolates were of the A1 mating type whereas all North American isolates were of the A2 type (Werres and Zielke 2003; Werres and Kaminski 2005). Thus, initially, mating types and geographic origin were believed to be totally linked. However, a strain in Belgium with the A2 mating type was recently reported (Werres and De Merlier 2003a) and the A1 mating type of P. ramorum was reported in North America using microsatellite alleles of the European tester isolates (Hansen et al. 2003a). This discovery was important because of the risk of crossing and recombination between these two mating types. So far, population surveys have not found any indication of recombination. However, since both mating types have been found, recombination is theoretically possible. New progeny with genetic and pathogenic variation could be produced and could change dramatically the epidemiology of this pathogen (Erwin and Ribeiro 1996). Appearance of the alternate mating type in another oomycete, the potato late blight fungus caused by Phytophthora infestans (Mont.) de Bary, has resulted in more genetic diversity, increased virulence in both potato (Solanum tuberosum L.) and tomato (Solanum lycopersicum L.), and increased resistance to metalaxyl (Goodwin 1997), the most widely used fungicide for late blight control.
122 Rapid detection and identification of the origin of P. ramorum, and determination of the mating type are critical in preventing or controlling its spread. Currently, there are quarantines enforced on all nursery plants coming from infested areas. However, with the discovery that large infested areas contain different mating types, it is important to take this information into consideration and try to prevent contact between the two mating types. Currently, methods to differentiate the European and North American P. ramorum are based on AFLPs, restriction fragment length polymorphisms (RFLPs) and microsatellites. These methods are not amenable to high throughput diagnostic assays because they use DNA from pure cultures and therefore cannot be used directly on field samples.
Diagnostic methods have been developed for the identification of P. ramorum. Real-time polymerase chain reaction (PCR) assays using the internally transcribed spacer (ITS) region and SYBR-green were designed (Hayden et al. 2004). Single nucleotide polymorphisms (SNPs) were found in the ITS region, as well as in the β-tubulin and elicitin genes (Bilodeau et al. 2005). Also, at the intraspecific level, the cytochrome oxidase I gene comprises a single nucleotide polymorphism between the NA1 and EU1 populations (Kroon et al. 2004b; Martin and Tooley 2003). Real-time PCR using TaqMan probes have subsequently been developed based on the above sequences for the detection of P. ramorum (Bilodeau et al. 2007c; Hayden et al. 2006; Hughes et al. 2006b; Schena et al. 2006; Tomlinson et al. 2005b; Tooley et al. 2006). With the exception of the assay reported in Kroon et al. (2004) these methods using RFLP target P. ramorum but usually cannot distinguish the strains from different geographic origins.
Developing reliable diagnostic assays based on SNPs is technically challenging. However, several approaches have been developed using SNPs in association studies to genotype populations. (Kwok 2001). Genotyping assays using PCR with allele specific oligonucleotides (ASO-PCR) are amenable to diagnostic assays given that the chemistry used in such assays is robust and would be useful in large-scale surveys.
123 The objectives of this study were to (i) sequence the DNA of β-tubulin and cellulose binding elicitor lectin (CBEL) genes from a worldwide collection of P. ramorum to discover intraspecific polymorphisms, and (Dodd et al.) develop a SNP genotyping assay for the rapid identification of the geographic origins of P. ramorum isolates and (iii) see whether can be uses for field samples.
III.3 Materials and methods
III.3.1
Isolates and DNA extraction
All isolates of Phytophthora used in this study are listed in Table 1. The P. ramorum collection comprised 35 isolates of European (22) and North American (13) origins collected from different hosts. Mycelium was cultivated and DNA was extracted following the procedures described in (de Cock et al. 1992b) or (Möller et al. 1992).
To assess the possibility of genotyping samples directly from the field, samples were obtained following the 2005 survey of Canadian nurseries by the Canadian Food Inspection Agency (CFIA). There were multiple positives from each nursery (in most cases) because we found an initial positive and then did extensive surveys on each nursery to delimit the infestation. In that survey, plant tissues were screened using an enzymelinked immunosorbent assay (ELISA) (the optical density measure of the test wells on a plate reader was at 405 nm.) and samples that were positive (mean A405 reading of duplicate sample wells minus the blank A405 reading must be greater than three times that of the negative control.) were then tested for the presence of P. ramorum, as described below. DNA was extracted directly from the ELISA extracts from multiple hosts that were known to be positive or negative for P. ramorum. The DNA samples and ELISA extracts were prepared from symptomatic leaf, petiole, stem or crown tissue samples. Approximately 612 leaf pieces weighing 0.15 g were cut with a No.5 cork borer. Samples were stored frozen if not tested the same day. Samples were mixed with GEB2 (general extraction buffer 2;
124 Agdia Incorporated, Elkhart, Ind.), and 5 mm stainless steel grinding balls were added, placed into the TissueLyser 24-sample adapter sets (Qiagen, Valencia, Calif.), ground and stored at -20 °C.
DNA was extracted on ELISA extracts as indicated above, using the DNeasy Plant mini kit (Qiagen) protocol according to the manufacturer’s recommendations except for the first step. Instead to use fresh tissues and AP1 Buffer, then 400 μl of ELISA GEB2 extraction buffer (ELISA extract) was added to a 2-mL tube with 1 μL of RNase A stock solution (100 mg/mL). After incubation for 10 min at room temperature (18-25°C), 130 μL of buffer AP2 was added to the lysate. Tubes were put at -20 °C for 10 min and then the lysate was centrifuged for 1 min at full speed (13,000 rpm). The lysate was treated using all regular steps of the DNeasy Plant mini kit beginning at the AP3/E step (Procedure 13 in DNeasy Plant Handbook 07/2006; http://www1.qiagen.com/literature/handbooks/literature.aspx?id=1000064).
III.3.2
Primer design for DNA sequencing
The primers listed in Table 2 were used to amplify by PCR coding sequences of two nuclear genes from P. ramorum and related Phytophthora species. β-tubulin gene primers were designed using alignment based on sequences from P. cinnamomi (GenBank accession number U22050). The primers CBEL1U and CBEL2L were designed using a similar approach (Table 2) with the CBEL sequence of Phytophthora nicotianae Breda de Haan (= Phytophthora parasitica Dastur) (Mateos et al. 1997) ; GenBank accession number X97205) and the P. infestans sequence from the Phytophthora functional genomic database at http://www.pfgd.org/ . The software Primer Premier 5.00 (Premier Biosoft International, Palo Alto, Calif.) was used to design PCR primers.
Amplifications with the primers CBEL1U and CBEL2L on DNA from P. ramorum and other species were performed to obtain amplicons and DNA sequences of CBEL. Three
125 new primers (CBEL5U, specific to P. ramorum, CBEL5UD and CBEL6) were designed to generate a longer fragment specific to P. ramorum, using an alignment of the genomic sequence
from
the
P.
ramorum
(Available
from
http://genome.jgi-
psf.org/Phyra1_1/Phyra1_1.home.html, gene ID 40789) against Phytophthora sojae Kaufmann and Gerdemann (Ps_006_22575_Jun03) and P. infestans (Pi_004_35130_Jun03 sequences from the Phytophthora functional genomic database available from http://www.pfgd.org/).
III.3.3
DNA sequencing and ASO primer extension
DNA from European and North American isolates of P. ramorum (Table 1) was amplified using β-tubulin (Oom-Btub-up415; Oom-Btub-lo1401) and CBEL (CBEL_1U; CBEL_2L) primers (Table 2). PCR products were purified with the QIAquick PCR purification Kit (Qiagen), quantified with a Nanodrop spectrophotometer ND-1000 (Nanodrop Technologies, Wilmington, Del.), and sequenced using the PCR primers with DNA template concentration from 0.05 to 60 ng/µl. CBEL5U and CBEL6L primers were used with some P. ramorum samples when amplification and sequencing were difficult to perform. CBEL 5DU is a degenerate primer used with CBEL6L when the Phytophthora isolate is not P. ramorum. Sequencing reactions were performed using a Big Dye Terminator Sequencing kit on an ABI 310 automated sequencer (PE Applied Biosystems, Foster City, Calif.). Both strands were sequenced with the primers listed in Table 2. Sequences were aligned using MegAlign version 5.08 (DNASTAR Inc., Madison, Wis) after which a ClustalW analysis was performed. Sequences were deposited in GenBank and the accession numbers are listed in Table 1. Samples without GenBank accession numbers were not sequenced or sequences were of low quality.
Based on the nucleotide sequence alignments of β-tubulin and CBEL, two SNPs were identified for each gene. Three primers were created for each of these SNPs. Two reverse primers had identical sequences but differed for the alternate base at the 3’ position,
126 and a third forward primer was common for both allelic forms (Table 2). Primer selection criteria included a melting temperature (Themann et al.) of 58 °C, primer lengths of 19-22 bp, avoidance of secondary structure and yield of PCR products of 400-500 bp for β-tubulin and 260 bp for CBEL.
III.3.4
Allele-specific genotyping
Real-time (RT) ASO-PCR (allele-specific oligonucleotide with polymerase chain reaction) was performed with an MJ Engine Opticon® 2 Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, Calif.). For each SNP assay, separate real-time-PCR reactions were conducted, one for each allele targeted, with two PCR reactions for each SNP tested. All reactions were performed in 25 µl volumes. To determine the annealing temperature to discriminate the EU1 and NA1 lineages, an annealing temperature gradient was performed on each set of primers for each SNP reaction using the two main genotypes of P. ramorum; CBS101548 (EU) and CBS110538 (NA). The optimal temperature was selected for each primer set to differentially amplify the two genotypes (EU and NA). The PCR reactions contained 0.3 µmol/L of each primer, 1X of QuantiTect SYBR® Green PCR Kit (Qiagen), and template DNA (1 µL). PCR cycling conditions were set at 95 °C for 15 min; 36 cycles at 94 °C for 15 s, 60 or 66 °C for 30 s (depending on the primers used; see Table 1), and 72 °C for 30 s, followed by a melting curve from 55 °C to 95 °C, with a reading at every 1.0 °C and a hold for 1 s measurement. Fluorescence was measured during the extension phase at 72 °C. Negative controls consisted of reactions with all reagents except template DNA. The total DNA template concentration used in the reaction was determined with a Nanodrop spectrophotometer ND-1000 ranging from 0.05 to 60 ng/µl for P. ramorum and from 5 to 110 ng/ul for other Phytophthora spp. The software Opticon Monitor version 2.01.10 (Bio-Rad Laboratories) was used to analyze the data. Data were exported as threshold cycle (Ct; Ct threshold was set at 0.015, global minimum) values and analyzed for comparisons in Excel (Microsoft® VER.9.0.3821 SR-1, Redmond, Wash.).
127
III.3.5
Allele-specific genotyping of field samples
Thirty-five samples from the 2005 CFIA survey, representing 14 host plant genera (Table 5), were obtained and analyzed using the ASO-PCR protocol. All samples were processed by ELISA to detect the presence of Phytophthora species, followed by DNA extraction, real-time PCR detection of P. ramorum (Bilodeau et al. 2007c), and ASO genotyping. PCR amplification protocols described in the previous section were used with field samples with the exception that 2 µL of DNA was used instead of 1 µL and 45 cycles were done to compensate for the lower DNA concentration and potential PCR inhibition in field samples. Two P. ramorum isolates (CBS 101553 and CBS 110542), a Phytophthora lateralis Tucker and Milbrath isolate (CBS 168.42), and water were used as positive and negative controls.
III.4 Results
III.4.1
DNA polymorphisms
Intraspecific polymorphisms among P. ramorum samples were observed in sequences of the nuclear β-tubulin and CBEL genes. The DNA sequence chromatograms revealed two peaks (G and T) at position 279 in the β-tubulin gene (β-tub279) of North American isolates of P. ramorum while a single peak (G) was present in all European isolates sequenced (Fig. 1a). This indicates that the North American strains are heterozygous at this site but that the European strains are homozygous. At position 858 (β-tub858), a single peak was present in the North American isolates whereas two peaks (A and G) were present in the European isolates (Fig. 1b). Sequences were obtained for 13 samples from North America and 22 samples from Europe (Tables 3 and 4). In the CBEL gene, two SNPs were also found, but both were heterozygous for the European isolates at position 245 (CBEL245; G and C) and 412 (CBEL412; G and A), whereas they were homozygous (G) in the American isolates at both positions (Fig. 1c and d).
128
III.4.2
Genotyping
Primers for the ASO were designed to amplify specifically either the North American or the alternate European alleles for both SNPs in each of the two genes. The PCR product lengths were 410 or 513 bp for β-tubulin and 267 or 268 bp for CBEL. Real-time PCR assays were developed using these PCR primer pairs. All European isolates were homozygous at βtubulin for SNP β-tub279 (G/G) but heterozygous for β-tub858 (A/G). By contrast, all California isolates were heterozygous for β-tub279 (T/G) but homozygous for β-tub858 (A/A). For CBEL, all European isolates were heterozygous for CBEL245 (G/C) and CBEL412 (G/A), but California isolates were homozygous at both SNP loci for CBEL245 (G/G) and CBEL412 (G/G). However, one of the Canadian nursery isolates (MSOD105C) and two Oregon isolates (Pr-3-74-1 and Pr-3-74-2), collected in 2003, were homozygous at β-tub279 (G/G), heterozygous at β-tub858 (A/G), heterozygous for CBEL245 (G/C) and CBEL412 (G/A), showing the profile found in the European (EU1) isolates (Tables 3 and 4).
III.4.5
Genotyping of field samples
A total of 35 samples from nurseries were tested with the four SNP genotyping assays (Table 5). These samples were collected as part of the CFIA 2005 SOD survey and included samples from British Columbia nurseries where the pathogen had previously been detected. All specimens were processed by ELISA to identify the presence of Phytophthora spp., then assayed by real-time PCR for detection of P. ramorum, plate cultured, and genotyped via ASO. Two of these samples were negative for Phytophthora in both the ELISA and the genotyping assays (samples 377 and 1998). Of the 33 other samples, 26 were positive for P. ramorum with the PCR diagnosis assay and seven were negative (Bilodeau et al. 2007c) (results not shown). Twenty-three samples showed the North American profile and three had the European one (Identification Nos. 107, 196, 11077).
129 The marker CBEL412 seems to be less sensitive than the other markers with a difference of 7 Ct between the two ASO primer sets used. The European profiles detected were probably too weak to be detected with this marker. DNA concentration was probably too low in sample No. 11077 to allow reliable genotyping. Only the two β-tubulin markers yielded reliable SNP profiles. These three European-like samples gave the North American profile with marker CBEL412A. Three other samples (Identification Nos. 12321, 12322, 16203) give an ambiguous North American profile, but are mostly based on only three of the four markers for the two first identification numbers and on one of four for the last one and they are probably too low DNA concentration. Some of the samples yielded no PCR Ct values and gave ambiguous profiles (Identification Nos. 56, 110, 175, 787, 2602, 2966 and 4959). Borderline detection of the high Ct values was also observed with these samples for the P. ramorum–specific detection assay (data not shown). The other seven field samples that were PCR negative for genotyping were also negative with the P. ramorum–specific detection assay (data not shown). On average, field samples that were analysed with the ASO assay had Ct values ranging from 22 to 42 depending on the markers used. Samples yielding a Ct >40 were considered out of range or below the detection limit. As the proportion of DNA concentration is inversely proportional to Ct value, high Ct values are usually caused by a low quantity of DNA. For control purposes, two P. ramorum samples (North American and European isolates) and one sample of P. lateralis were used in this assay to verify genotype and specificity. Phytophthora lateralis appears to amplify on the four markers but only at high DNA concentrations (8 ng/uL). It should be noted that these primers were designed to target SNPs to discriminate between P. ramorum populations (EU or NA), but not to differentiate P. ramorum from other Phytophthora species. When tested against the DNA of other Phytophthora spp., some combinations of primer pairs amplified a DNA fragment of one genotype for β-tubulin, for example from P. lateralis, P. cryptogea, Phytophthora ilicis Buddenhagen and Young (not shown), when the DNA from these species were at high concentrations (15 ng/uL-0.4ng/uL). However, the results were discordant among the SNP genotypes for P. lateralis with some SNPs being homozygous for the NA type, others for the EU type, even within a single gene (Table 5). This was also seen using CBEL markers, the data detect the two alleles and could be distinct to P. ramorum using both SNPs on the two genes.
130
III.5 Discussion DNA sequences of CBEL and β-tubulin genes of P. ramorum revealed polymorphisms between the European and Californian populations which we targeted for the development of a SNP genotyping assay. This assay could be used to identify the geographic origin of the P. ramorum strain isolated in cultures. We also showed that these genotyping assays can be used directly from field tissues without culturing with little ambiguities and provide a quick assay of geographic origin.
Our findings confirm earlier results with AFLP showing a clear distinction between European and North American populations (Ivors et al. 2004). The SNP genotyping results presented here also highlights the similarity between the Canadian or Oregon P. ramorum isolates from 2003 and the European population. These results are significant in the light of the differences observed in the etiology and the epidemiology of the European and California types. If the European type is mostly a nursery disease that does not result in extensive mortality, the Canadian strain discovered in 2003 could be considered a lesser threat to the nursery and forestry industries as well as to the natural stands of oaks. All host plants found to be associated with this positive find were destroyed according to the CFIA nursery eradication protocol. Since 2004, both types have been discovered in Canadian nurseries.
The genotyping methods used here were specific to the SNPs assayed and were confirmed with sequencing of pure cultures (Genbank accession; Table 1). The primer extension with ASO-PCR was efficient in detecting the alleles of P. ramorum and was in agreement with the DNA sequence results. The DNA of P. lateralis yielded some positive results when used in the SNP genotyping assays. The concentration of DNA from pure cultures used in the assays was far higher than those expected from environmental samples (usually 10-100 pg; (Tomlinson et al. 2005a). Therefore, we used conditions that were
131 conducive to false positives on purpose.
Phytophthora lateralis shared DNA
polymorphisms with P. ramorum in the targeted gene regions. In the CBEL sequence, P. lateralis presents an insertion in the sequence, as a minisatellite, when aligned with P. ramorum (not shown). However, when all four assays were combined, there was no ambiguity between P. ramorum and P. lateralis. This result reveals one of the strengths of this approach: instead of obtaining a present/absent result, as with diagnostic assays targeting only P. ramorum, the SNP assays can reveal new combinations, including recombinants, as well as potentially closely related taxa. Combined with a previous round of P. ramorum-specific assays, as used in this report, they provide reliable assays to determine geographic origin. The assays developed here are easily amenable to upscaling and high throughput testing in plate formats, and could be incorporated into quarantine surveys.
Although this survey was conducted only on two gene targets and four SNPs, it is noteworthy that we did not find any recombinants between the European and North American types. Our SNP genotyping analyses conducted on both culture and field samples do not support the hypothesis that there is recombination within P. ramorum. The complete absence of homozygotes at two polymorphic loci indicates that there is genetic variability, but sexual reproduction either does not occur or is rare in nature. The A2 mating type was recently discovered in Europe (Werres and De Merlier 2003a). The absence of recombination could be explained by uneven distribution of the mating type alleles in the population in North America. P. ramorum is a heterothallic organism; it requires two complementary mating types to sexually reproduce. However, it can reproduce asexually by the production of zoospores and chlamydospores, which probably explains the presence in North America of large clonal populations (Erwin and Ribeiro 1996; Werres and Zielke 2003).
Mating types was completely correlated with the SNP genotypes, except in a single Belgian sample (CBS 110901) where the European DNA profile was present in a sample with the A2 mating type (Tables 3 and 4). Prior to the discovery of the second mating type in Europe (Werres and De Merlier 2003a), there was a complete linkage between
132 geographic origin (Europe and North America) and mating type. The discovery of a European profile with the A2 mating type clearly shows that the A2 mating type can occur within the European DNA profile and that caution should be taken not to equate the European DNA profile with the A1 mating type.
In the spring 2003, the discovery of an unusual strain of P. ramorum in nurseries in Washington state, in northern Oregon and in British Columbia (CFIA 2003; Hansen et al. 2003b; Werres and De Merlier 2003a) was believed to be the result of an introduction most likely from Europe. The identical DNA genotypes of these North American samples with European samples and the presence of the A1 mating type in these samples are strong indicators that these new introductions were from sources distinct from the California types and that the pathogen did not migrate from Californian to Canadian nurseries in 2003. A second interception in 2004 related the sample to the California genotype and was associated with a Monrovia, California nursery, as shown by trace-back records. In our sampling, there was a complete correlation, excluding the Canadian and Oregon samples from 2003, between the DNA haplotype and the geographic origin. The British Columbia and Oregon nursery samples from 2003 had a haplotype identical to the European (EU1) populations, strongly suggesting the recent and distinct introduction of a new strain of P. ramorum into North America from Europe. Field samples collected in Canadian nurseries in 2004-2005 showed that some of these samples could also come from California, most likely from the Monrovia nursery. But this field survey also showed that the European genotype is still present in three nurseries (Table 5). The redundancy provided by the genotyping of two genes increases the confidence in the assignment of the genotype to its geographic origin. Indeed, both genes gave concordant results relative to the geographic origin. The impact of this discovery in terms of understanding the epidemiology is important. The presence of the EU1 and NA1 types in Canadian nurseries could also have an effect on the application of the quarantine and the discovery of the introduction pathways of the pathogen.
The primer extension assays developed here provide an easy and low-cost method of genotyping P. ramorum. The assays developed in this study use ASOs and were used in
133 real-time PCR reactions. However, the assays could also be used on conventional PCR, with the detection conducted by agarose gel electrophoresis (data not shown).
The
disadvantage of conducting this assay with agarose gel electrophoresis would be that it is more time consuming than real-time PCR and it adds extra manipulation.
The assay was also used for genotyping field samples, demonstrating the potential of this assay on infected plant samples. A majority of host samples that tested positive for Phytophthora with ELISA, and more specifically for P. ramorum with real-time PCR, were genotyped using ASO and displayed either the North American or the European profile. Since these were field samples from nurseries, it appears that a majority of American genotypes seem to be present in Canadian nurseries in 2005 but a few European genotypes were also present in only two nurseries. These European genotypes were detected only on three or fewer of the four assays with a high Ct value, which means low DNA concentration for these samples. Unfortunately, culture isolations were not possible with these samples. Perhaps because of lower copy numbers in the genes targeted compared with other DNA assays used in diagnostics such as the ITS region, the ASO-PCR assays were less sensitive.
These assays could inform us about several outcomes, such as the origin of the strain and possible recombinants of P. ramorum. DNA sequences of P. ramorum from two coding regions enabled us to discover intraspecific polymorphisms and to generate an SNP genotyping assay for the rapid identification of the geographic origin of P. ramorum. This tool has potential applications for quarantine procedures to determine the origin and (or) the mating type of the pathogen.
Another assay, based on cytochrome oxidase I gene
amplification and restriction enzyme digestion, has been used to discriminate between European and North American P. ramorum (Kroon et al. 2004). When this assay was used on field samples, a success rate of 21 of 36 was obtained (discussed in Kroon (2004)). Our assay showed also the potential to use ASO to genotype infested plant material with a success rate of 20 out of 26, with 6 genotypes that were inconclusively determined, possibly because of DNA concentration. The advantage of using SNP genotyping assay on multiple nuclear biparentally inherited genes, instead of a single maternally inherited
134 mitochondrial gene, is that recombinants or novel types can also be determined. By sampling coding genes, we can also target non-synonymous polymorphisms in genes involved in evolution or adaptation. This assay is easy, rapid and cheap if used in a highthroughput format. This approach could have an impact on preventing the introduction of a new genotype not present in a geographic area.
III.6 Acknowledgements The authors thank the laboratory team of Dr François Rousseau, St-François d’Assise Hospital, Quebec City, for their help with the genotyping technique. We thank Gervais Pelletier for his help with DNA extraction of ELISA products and Nicole Desaulniers for her help with DNA sequencing and providing samples. We also thank Lucyna Kumor for providing ELISA lysate and information. Funding for this work was received from the Canadian Biotechnology Strategic Fund and the CBRN Research and Technology Initiative (CRTI grant 04-0045RD).
135
Table 1. Isolates of Phytophthora ramorum from different culture collections used in this study. GenBank accession CBS Isolate No. Origin Host No. No. β-tubulin CBEL 101553 BBA 9/95 Typea Germany EF117938 EF117956 Rhododendron catawbiense Michx. 109278 BBA 16/99 a Germany EF117940 EF117958 Viburnum ×bodnantense Rhododendron sp. EF117926 EF117950 101331 PD 98/8/2627 a Ede, the Netherlands, Rhododendron sp. EF117925 EF117948 101329 PD 98/8/6285 a Lisse, the Netherlands 101332 PD 94/844 a Roosendaal, the Rhododendron sp. EF117933 EF117951 Netherlands 101330 PD 98/8/5233 a Hazerswoude, Viburnum sp. EF117932 EF117949 the Netherlands Rhododendron sp. EF117924 EF117947 101328 PD 98/8/6743 a Bentveld, the Netherlands 101548 BBA 69082 a Germany Rhododendron, hybrid EF117934 EF117964 'Schneewolke' Germany Recycling water, --EF117952 101549 BBA 104/5 a nursery 101551 BBA 12/98 a Germany EF117936 EF117954 Rhododendron catawbiense 101552 BBA 9/3 a Germany Recycling water, EF117937 EF117955 nursery Rhododendron sp. EF117941 EF117959 109279 BBA 13/99 -1 a Germany a Germany EF117935 EF117953 101550 BBA 14/98-a Rhododendron catawbiense 101326 PD 98/8/6933 a The Rhododendron sp. EF117930 EF117945 Netherlands Germany Recycling water, EF117939 EF117957 101554 BBA 2/4 a nursery 101327 adc 98.36 a Boskoop, the Rhododendron sp. EF117931 EF117946 Netherlands, Rhododendron sp. EF117929 EF117962 110548 det.nr.2002-195 France a
110534 Pr01 a 110535 Pr03 a
Marin Co., California, California
110537 Pr52 a 110538 Pr65 a
California California
Quercus agrifolia Née EF117915 EF117965 Lithocarpus densiflora EF117916 EF117966 (Hook. & Arn.)Rehd. Rhododendron sp. EF117917 EF117967 EF117918 EF117971 Quercus parvula Greene
136 110539 Pr70 a
California
EF117914 EF117968 Vaccinium ovatum Pursh 110541 Pr86 a California EF117919 EF117972 Arbustus menziesii Pursh 110542 Pr110 a California EF117920 EF117973 Umbellularia californica (Hook. & Arn.)Nutt. 110543 Pr159 a California Lithocarpus densiflora EF117921 EF117969 a 110544 PrPRJL3.5.3 California Sequoia sempervirens EF117922 EF117970 (D. Don) Endl. 110545 Rh/2/00 a Poland Rhododendron sp. EF117923 EF117960 a 110546 Rh/6/00 Poland Rhododendron sp. EF117927 EF117961 110547 Rh/122/98 a Poland Rhododendron sp. EF117928 --a 110601 Pr84-sz California Soil ----110900 CRA 2386 a Belgium Rhododendron sp. ----Belgium Viburnum sp. ----110901 CRA 2338 a --MSOD 105 c Canada Rhododendron sp., EF117942 EF117963 nursery --Pr-3-74-1 c Oregon Pieris sp. EF117943 --c --Pr-3-74-2 Oregon Viburnum sp. EF117944 --a From Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands. b ---, No GenBank accession no. (The genotypes were determined only by ASO for these samples) c From CFIA, Pest DNA Diagnostics Laboratory, Centre for Plant Quarantine Pests, CFIA, Ottawa, Ont.
137 Table 2. Primers used for PCR assays targeting Phytophthora spp. and P. ramorum. Name
Primer sequencea
Region
Amplicon size
Annealing T°C
Primers for sequencing Oom-Btub-up415
CGCATCAACGTGTACTACAA
β-tubulin
˜986pb
62
Oom-Btub-lo1401
CGCTTGAACATCTCCTGG
β-tubulin
˜986pb
62
CBEL_1U
TGYCARCCNTGGAAYGC
CBEL
˜350pb
60
CBEL_2L
GCRCANGCRTCRCARCA
CBEL
˜350pb
60
CBEL5U
TTGCTCGCATTACTGCCGTCT
CBEL
˜512pb
60
CBEL5DU
TYGCTCGCATTACTGYCGYCT
CBEL
˜512pb
60
CBEL6L
CAGCAGAGCGACGGGAGGC
CBEL
˜512pb
60
ASO Primers ASObt279-G
GTTCCAGATCACGCACTCG
β-tubulin
˜410pb
63
ASObt279-T
GTTCCAGATCACGCACTCT
β-tubulin
˜410pb
63
Common279bt
GCGAGGTCAGCGGGGCGAAAC
β-tubulin
˜410pb
63
ASObt858-A
CACGTTCAGCATCTGCTCA
β-tubulin
˜513pb
65
ASObt858-G
CACGTTCAGCATCTGCTCG
β-tubulin
˜513pb
65
Common858bt
TGAGGAGTACCCGGACCGTATC
β-tubulin
˜513pb
65
ASOCBEL245G
ACCAGCACGCCTACACCG
CBEL
˜267pb
66
ASOCBEL245C
ACCAGCACGCCTACACCC
CBEL
˜267pb
66
CommonCBEL 245
CGCAGCAGAGCGACGGGA
CBEL
˜267pb
66
ASOCBEL412G (C)
TGGATGCACTGGTAGAAGCTG
CBEL
˜268pb
66
ASOCBEL412A (T)
TGGATGCACTGGTAGAAGCTA
CBEL
˜268pb
66
CommonCBEL 412
TTCGTGAACAGCAACCCTGGACA
CBEL
˜268pb
66
a
Polymorphism SNPs are underlined.
138 Table 3. ASO genotyping of β-tubulin gene in a selection of Phytophthora ramorum from different geographic origins and mating types. ASO genotype a Mating SNP1: SNP2: Sample No. Country ORIGIN b Type c ASO279 ASO858 CBS 101553
Germany
G/G
A/G
EU
A1
CBS 109278
Germany
G/G
A/G
EU
A1
CBS 101331
Netherlands
G/G
A/G
EU
A1
CBS 101329
Netherlands
G/G
A/G
EU
A1
CBS 101332
Netherlands
G/G
A/G
EU
A1
CBS 101330
Netherlands
G/G
A/G
EU
A1
CBS 101328
Netherlands
G/G
A/G
EU
A1
CBS 101548
Germany
G/G
A/G
EU
A1
CBS 101549
Germany
G/G
A/G
EU
A1
CBS 101551
Germany
G/G
A/G
EU
A1
CBS 101552
Germany
G/G
A/G
EU
A1
CBS 109279
Germany
G/G
A/G
EU
A1
CBS 101550
Germany
G/G
A/G
EU
A1
CBS 101326
Netherlands
G/G
A/G
EU
A1
CBS 101554
Germany
G/G
A/G
EU
A1
CBS 101327
Netherlands
G/G
A/G
EU
A1
CBS 110548
France
G/G
A/G
EU
A1
CBS 110545
Poland
G/G
A/G
EU
A1
CBS 110546
Poland
G/G
A/G
EU
A1
CBS 110547
Poland
G/G
A/G
EU
A1
MSOD105C
Canada 2003
G/G
A/G
EU
A1
139 Pr-3-74-1
Oregon 2003
G/G
A/G
EU
A1
Pr-3-74-2
Oregon 2003
G/G
A/G
EU
A1
CBS 110534
California
T/G
A/A
NA
A2
CBS 110535
California
T/G
A/A
NA
A2
CBS 110537
California
T/G
A/A
NA
A2
CBS 110538
California
T/G
A/A
NA
A2
CBS 110539
California
T/G
A/A
NA
A2
CBS 110541
California
T/G
A/A
NA
A2
CBS 110542
California
T/G
A/A
NA
A2
CBS 110543
Oregon
T/G
A/A
NA
A2
CBS 110544
California
T/G
A/A
NA
A2
CBS 110601
California
T/G
A/A
NA
A2
CBS 110900
Belgium
G/G
A/G
EU
A1
CBS 110901
Belgium
G/G
A/G
EU
A2
a
For each sample, the target nucleotide of the SNP assay is listed if a Ct < 40 was observed in the ASO assay. b
Origin determined based on the ASO genotyping. EU, European (EU1 lineage); North American (NA1 lineage). c Determined by previous papers (Ivors et al. 2006; Werres and Kaminski 2005), the others samples not used in these two references were determined by correlation.
140
Table 4. ASO genotyping of CBEL gene in a selection of Phytophthora ramorum from different geographic origins and mating types. ASO genotype a Sample No.
Country
SNP1:
SNP2:
ASO245
ASO412
ORIGIN b
Mating Type c
CBS 101553
Germany
G/C
G/A
EU
A1
CBS 109278
Germany
G/C
G/A
EU
A1
CBS 101331
Netherlands
G/C
G/A
EU
A1
CBS 101329
Netherlands
G/C
G/A
EU
A1
CBS 101332
Netherlands
G/C
G/A
EU
A1
CBS 101330
Netherlands
G/C
G/A
EU
A1
CBS 101328
Netherlands
G/C
G/A
EU
A1
CBS 101548
Germany
G/C
G/A
EU
A1
CBS 101549
Germany
G/C
G/A
EU
A1
CBS 101551
Germany
G/C
G/A
EU
A1
CBS 101552
Germany
G/C
G/A
EU
A1
CBS 109279
Germany
G/C
G/A
EU
A1
CBS 101550
Germany
G/C
G/A
EU
A1
CBS 101326
Netherlands
G/C
G/A
EU
A1
CBS 101554
Germany
G/C
G/A
EU
A1
CBS 101327
Netherlands
G/C
G/A
EU
A1
CBS 110548
France
G/C
G/A
EU
A1
CBS 110545
Poland
G/C
G/A
EU
A1
CBS 110546
Poland
G/C
G/A
EU
A1
CBS 110547
Poland
G/C
G/A
EU
A1
141 MSOD105C
Canada 2003
G/C
G/A
EU
A1
Pr-3-74-1
Oregon 2003
G/C
G/A
EU
A1
Pr-3-74-2
Oregon 2003
G/C
G/A
EU
A1
CBS 110534
California
G/G
G/G
NA
A2
CBS 110535
California
G/G
G/G
NA
A2
CBS 110537
California
G/G
G/G
NA
A2
CBS 110538
California
G/G
G/G
NA
A2
CBS 110539
California
G/G
G/G
NA
A2
CBS 110541
California
G/G
G/G
NA
A2
CBS 110542
California
G/G
G/G
NA
A2
CBS 110543
Oregon
G/G
G/G
NA
A2
CBS 110544
California
G/G
G/G
NA
A2
CBS 110601
California
G/G
G/G
NA
A2
CBS 110900
Belgium
G/C
G/A
EU
A1
CBS 110901
Belgium
G/C
G/A
EU
A2
a
For each sample, the target nucleotide of the SNP assay is listed if a Ct < 40 was observed in the ASO assay. b
Origin determined based on the ASO genotyping. EU, European (EU1 lineage); North American (NA1 lineage). c Determined by previous papers (Ivors et al. 2006; Werres and Kaminski 2005), the others samples not used in these two references were determined by correlation.
142
Table 5. Allele-specific oligonucleotide genotyping of field samples (nurseries).* Nurserya
Sample identification No. c
Hosts
β-tubulin 279Gb
279Tb
Genotype
Profile
858Ab
858Gb
Genotype
Profile
Rhododendron sp.
N/A
N/A
N/A
-
N/A
N/A
N/A
-
A
56
A
107 d
Camellia sp.
36.63
N/A
GG
EU
35.68
33.37
AG
EU
A
110 c
Rhododendron sp.
N/A
N/A
N/A
-
N/A
N/A
N/A
-
A
175 c
Viburnum sp.
N/A
N/A
N/A
-
N/A
N/A
N/A
-
A
196 d
Rhododendron sp.
34.05
N/A
GG
EU
35.13
31.73
AG
EU
A
377 c,e
Acer palmatum
N/A
N/A
N/A
-
N/A
OR
N/A
-
B
787 c
Leucothoe sp.
N/A
N/A
N/A
-
N/A
N/A
N/A
-
C
1310 d
Camellia sp.
32.92
34.52
GT
NA
31.16
N/A
AA
NA
D
1998 c,e
Rhododendron sp.
N/A
N/A
N/A
-
N/A
N/A
N/A
-
E
2602 c
Viburnum sp.
N/A
N/A
N/A
-
N/A
N/A
N/A
-
E
2865 d
Camellia x Williamsi
38.12
OR
GT ?
NA
38.18
N/A
AA
NA
E
2866 d
Pieris sp.
N/A
N/A
N/A
-
36.41
N/A
AA
NA
E
2925 d
Camellia sasanqua
39.39
N/A
GG?
EU?-
35.93
N/A
AA
NA
E
2966 c
Rhododendron sp.
N/A
OR
N/A
-
N/A
N/A
N/A
-
E
2979 d
Rhododendron sp.
N/A
OR
N/A
-
OR
N/A
N/A
-
143
E
3000 d
Rhododendron sp.
31.81
34.61
GT
NA
33.32
N/A
AA
NA
E
3043 d
Hamamelis intermedia
37.15
38.66
GT
NA
36.94
OR
AA
NA
F
4166 d
Kalmia
34.08
36.89
GT
NA
35.33
N/A
AA
NA
F
4959 c
Pyracantha
OR
N/A
N/A
-
N/A
OR
N/A
-
G
10103 d
Rhododendron sp.
OR
N/A
N/A
-
39.10
N/A
AA
NA
H
11077 d
Magnolia
40.18
N/A
GG
EU?
39.66
35.37
AG
EU?
I
11125 d
Rhododendron sp.
28.81
33.78
GT
NA
31.37
N/A
AA
NA
J
12321 d
Rhododendron sp.
36.64
OR
GT ?
NA?
OR
N/A
N/A
-
J
12322 d
Rhododendron sp.
32.29
38.88
GT ?
NA?
34.74
38.20
AG
EU?
K
12746 d
Rhododendron sp.
37.56
40.49
GT
NA
37.87
N/A
AA
NA
K
12768 d
Rhododendron sp.
36.26
39.91
GT
NA
36.98
N/A
AA
NA
L
12977 d
Rhododendron sp.
25.71
30.30
GT
NA
27.51
33.88
AG
EU?
E
14791 d
Michelia maudiae
30.52
31.78
GT
NA
31.38
N/A
AA
NA
E
14792 d
Michelia wilsinii
29.87
32.47
GT
NA
33.32
37.56
AG
EU?
E
15000 d
Parrotia persica
27.36
30.71
GT
NA
29.98
OR
AA
NA
E
15006 d
Parrotia persica
28.16
31.06
GT
NA
30.74
OR
AA
NA
E
16203 d
Prunus lusitanica
N/A
N/A
N/A
-
N/A
N/A
N/A
-
E
16207 d
Prunus lusitanica
38.05
39.87
GT
NA
36.79
N/A
AA
NA
144
E
17017 d
Camellia sp.
35.99
37.10
GT
NA
36.32
N/A
AA
NA
E
18695 d
Gaultheria shallon
34.23
34.87
GT
NA
32.62
N/A
AA
NA
CBS 101553 (P. ramorum) 5ng/uL
23.40
N/A
GG
EU
25.47
23.57
AG
EU
CBS 110542 (P. ramorum)
24.66
25.58
GT
NA
24.80
OR
AA
NA
Water (Negative control)
N/A
N/A
N/A
-
N/A
N/A
N/A
-
CBS 168.42 (P. lateralis) 8ng/uL
20.84
OR
GG
EU
22.57
OR
AA
NA
Note: Samples were collected as part of the Canadian Food Inspection Agency annual sudden oak death survey. All specimens were processed by ELISA to identify presence of Phytophthora spp., followed by real-time PCR diagnostic and ASO genotyping. A question mark indicates that the analysis yielded an ambiguous type, and a profile could not be determined for the missing data. N/A, No amplification; NA, North American; EU, European; OR, Out of range. a Samples were taken from 12 different nurseries, each designated by a different letter (A to L). b Values are calculated Ct value; the Ct thresholds manually were set at 0.015 using global minimum. c Tested negative with real-time PCR diagnosis assay of Phytophthora ramorum using TaqMan on ITS, β-tubulin and elicitin. d Tested positive with real-time PCR diagnosis assay of Phytophthora ramorum using TaqMan on ITS, β-tubulin and elicitin (Bilodeau et al. 2007c), samples with high Ct (>40) are in boldface. e ELISA negative to Phytophthora. f This European profile (EU) was determined on 75% of the four assays tested, for most of them Ct value seems to be too high, meaning low DNA concentration. g Samples from control were from CBS cultures (Centraalbureau voor Schimmelcultures).
145
Table 5. Allele-specific oligonucleotide genotyping of field samples (nurseries). (suite) * CBEL Sample identification No.
245Gb
245Cb
Genotype
Profile
412Gb
412Ab
Genotype
Profile
Cumulative profile
56 c
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
107 d
34.97
31.47
GC
EU
35.55
N/A
GG
NA?
EUf
110 c
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
175 c
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
196 d
32.15
30.98
GC
EU
34.05
N/A
GG
NA?
EUf
377 c,e
N/A
N/A
N/A
N/A
N/A
N/A
-
-
787 c
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
1310 d
29.64
N/A
GG
NA
32.21
N/A
GG
NA
NA
1998 c,e
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
2602 c
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
2865 d
33.03
N/A
GG
NA
36.09
N/A
GG
NA
NA
2866 d
34.57
N/A
GG
NA
37.16
N/A
GG
NA
NA
2925 d
34.54
N/A
GG
NA
37.10
N/A
GG
NA
NA
2966 c
N/A
N/A
N/A
-
N/A
N/A
N/A
-
-
2979 d
33.77
N/A
GG
NA
N/A
N/A
N/A
-
NA
146
3000 d
27.36
N/A
GG
NA
31.02
N/A
GG
NA
NA
3043 d
32.29
N/A
GG
NA
35.65
N/A
GG
NA
NA
4166 d
29.68
N/A
GG
NA
31.86
N/A
GG
NA
NA
4959 c
OR
OR
N/A
-
N/A
N/A
N/A
-
-
10103 d
35.73
OR
GG
NA
37.34
N/A
GG
NA
NA
11077 d
N/A
OR
N/A
-
N/A
N/A
N/A
-
EUf
11125 d
24.63
OR
GG
NA
30.10
N/A
GG
NA
NA
12321 d
32.07
N/A
GG
NA
42.88
N/A
GG
NA
NA ?
12322 d
28.07
N/A
GG
NA
36.42
N/A
GG
NA
NA?
12746 d
31.96
N/A
GG
NA
35.25
N/A
GG
NA
NA
12768 d
30.53
N/A
GG
NA
34.62
N/A
GG
NA
NA
12977 d
22.99
N/A
GG
NA
28.75
N/A
GG
NA
NA
14791 d
25.67
N/A
GG
NA
29.36
OR
GG
NA
NA
14792 d
24.91
OR
GG
NA
29.02
OR
GG
NA
NA
15000 d
22.37
N/A
GG
NA
27.90
N/A
GG
NA
NA
15006 d
24.08
N/A
GG
NA
30.34
OR
GG
NA
NA
16203 d
37.28
N/A
GG
NA
N/A
N/A
N/A
-
NA ?
16207 d
N/A
N/A
N/A
-
38.05
N/A
GG
NA
NA
147
17017 d
29.46
N/A
GG
NA
34.00
OR
GG
NA
NA
18695 d
28.75
N/A
GG
NA
33.68
N/A
GG
NA
NA
23.15 21.78 GC EU 24.20 31.26 GA EU EU CBS 101553 (P. ramorum) 5ng/uL 21.88 OR GG NA 21.49 N/A GG NA NA CBS 110542 (P. ramorum) N/A N/A N/A N/A N/A N/A Water (Negative control) 37.23 33.56 GC EU32.68 N/A GG NA -? CBS 168.42 (P. lateralis) 8ng/uL Note: Samples were collected as part of the Canadian Food Inspection Agency annual sudden oak death survey. All specimens were processed by ELISA to identify presence of Phytophthora spp., followed by real-time PCR diagnostic and ASO genotyping. A question mark indicates that the analysis yielded an ambiguous type, and a profile could not be determined for the missing data. N/A, No amplification; NA, North American; EU, European; OR, Out of range. a
Samples were taken from 12 different nurseries, each designated by a different letter (A to L). Values are calculated Ct value; the Ct thresholds manually were set at 0.015 using global minimum. c Tested negative with real-time PCR diagnosis assay of Phytophthora ramorum using TaqMan on ITS, β-tubulin and elicitin. d Tested positive with real-time PCR diagnosis assay of Phytophthora ramorum using TaqMan on ITS, β-tubulin and elicitin (Bilodeau et al. 2007c), samples with high Ct (>40) are in boldface. e ELISA negative to Phytophthora. f This European profile (EU) was determined on 75% of the four assays tested, for most of them Ct value seems to be too high, meaning low DNA concentration. g Samples from control were from CBS cultures (Centraalbureau voor Schimmelcultures). b
148
β-tubulin a
b
EU
EU
EU
NA
NA
CBEL c
d
EU
EU
EU
NA
NA
Figure 1. SNPs observed on the chromatogram sequence of β-tubulin and CBEL: a) SNP1 at position 279 of β-tubulin, b) SNP2 at position 858 of β-tubulin, c) SNP1 at position 245 of CBEL, and d) SNP2 at position 412 of CBEL. Arrows show the SNPs. EU, European type; NA, North American type; K = T and G; R = A and G; S = C and G; Y = C and T.
149
a
G/G G T
b
G/T
G
T
Figure 2. Allele-specific oligonucleotide (ASO) genotyping using real-time PCR amplification with SYBR® Green on SNP at position 279 of β-tubulin: a) European sample CBS 110545 and b) North American sample CBS 110538. The two primer sets used are ASObt279-G with Common279bt and ASObt279-T with Common279bt. Details of the primers are given in Table 2.
150
III.7 References Bilodeau, G.J., C.A. Lévesque, A.W.A.M. de Cock, C. Duchaine, S. Brière, P. Uribe, F.N. Martin, and R.C. Hamelin (2007) Molecular detection of Phytophthora ramorum by real-time polymerase chain reaction using TaqMan, SYBR Green, and molecular beacons. Phytopathology 97: 632-642. Brasier, C., (2003) Sudden Oak Death: Phytophthora ramorum exhibits transatlantic differences. Mycol. Res. 107: 257-259. Brasier, C.M., J. Rose, S.A. Kirk, and J.F. Webber (2002) Pathogenicity of Phytophthora ramorum isolates from North America and Europe to bark of European Fagaceae, American Quercus rubra and other forest trees. In Proceedings of the Sudden Oak Death Science Symposium: The State of Our Knowledge, December 15-18, 2002, Monterey, CA. [online] [updated November 12, 2002] [accessed August 08, 2007]. Available from http://danr.ucop.edu/ihrmp/sodsymp/paper/paper09.html. CFIA (2003) Hosts of Phytophthora ramorum (with notes on geographical distribution and mating types) [cited July 2003]. Available from http://www.cnr.berkeley.edu/comtf/pdf/P.ramorum.hosts.June.2003.pdf. de Cock, A.W.A.M., A. Neuvel, G. Bahnweg, J.C.J.M. deCock, and H.H. Prell (1992) A comparison of morphology, pathogenicity and restriction fragment patterns of mitochondrial DNA among isolates of Phytophthora porri Foister. Neth. J. Plant Pathol. 98: 277-289. Erwin, D.C., and O.K. Ribeiro (1996) Phytophthora Diseases Worldwide. APS Press, St. Paul, Minn. 562 pp. Garbelotto, M., D.M. Rizzo, K. Hayden, M. Meija-Chang, J.M. Davidson, and S. Tjosvold (2002) Phytophthora ramorum and Sudden Oak Death in California: III. Preliminary studies in pathogen genetics. USDA For. Serv. Gen. Tech. Rep. PSWGTR-184. pp. 765-774. Garbelotto, M., J.M. Davidson, K. Ivors, P.E. Maloney, D. Hüberli, S.T. Koike, and D.M. Rizzo (2003) Non-oak native plants are main hosts for sudden oak death pathogen in California. Calif. Agric. 57: 18-23. Goodwin, S.B., (1997) The population genetics of Phytophthora. Phytopathology, 87: 462473. Hansard, C., (2003) Introduced Species Summary Project. Sudden Oak Death (Phytophthora ramorum). Columbia University, May 10, 2007 [online] Columbia University, New York. Available from http://www.columbia.edu/itc/cerc/danoff-
151 burg/invasion_bio/inv_spp_summ/Phytophthora_ramorum.htm. [updated February 17, 2003; accessed 10 May 2007]. Hansen, E.M., P.W. Reeser, W. Sutton, and L.M. Winton (2003) First report of A1 mating type of Phytophthora ramorum in North America. Plant Dis. 87: 1267. Hayden, K.J., D. Rizzo, J. Tse, and M. Garbelotto (2004) Detection and quantification of Phytophthora ramorum from California forests using a real-time polymerase chain reaction assay. Phytopathology, 94: 1075-1083. Hayden, K., K. Ivors, C. Wilkinson, and M. Garbelotto (2006) TaqMan chemistry for Phytophthora ramorum detection and quantification, with a comparison of diagnostic methods. Phytopathology, 96: 846-854. Hughes, K.J.D., J.A.Tomlinson, R.L. Griffin, N. Boonham, A.J. Inman, and C.R. Lane (2006) Development of a one-step real-time polymerase chain reaction assay for diagnosis of Phytophthora ramorum. Phytopathology, 96: 975-981. Ivors, K.L., K.J. Hayden, P.J.M. Bonants, D.M. Rizzo, and M. Garbelotto (2004) AFLP and phylogenetic analyses of North American and European populations of Phytophthora ramorum. Mycol. Res. 108: 378-392. Ivors, K., M. Garbelotto, I.D.E. Vries, C. Ruyter-Spira, B.T. Hekkert, N. Rosenzweig, and P. Bonants (2006) Microsatellite markers identify three lineages of Phytophthora ramorum in US nurseries, yet single lineages in US forest and European nursery populations. Mol. Ecol. 15: 1493-1505. Kroon, L.P.N.M., E.C.P. Verstappen, L.F.F. Kox, W.G. Flier, and P.J.M. Bonants (2004) A rapid diagnostic test to distinguish between American and European populations of Phytophthora ramorum. Phytopathology, 94: 613-620. Kwok, P.-Y., (2001) Methods for genotyping single nucleotide polymorphisms. Annu. Rev. Genomics Hum. Genet. 2: 235-258. Martin, F.N., and P.W. Tooley (2003) Phylogenetic relationships among Phytophthora species inferred from sequence analysis of mitochondrially encoded cytochrome oxidase I and II genes. Mycologia 95: 269-284. McPherson, B.A., D.L. Wood, A.J. Storer, P. Svihra, D.M.Rizzo, N.M.Kelly, and R.B. Standiford (2000) Oak mortality syndrome: sudden death of oaks and tanoaks. Cal. Dept. For. Tree Note No. 26. Möller, E.M., G. Bahnweg, H. Sandermann, and H.H. Geiger (1992) A simple and efficient protocol for isolation of high molecular weight DNA from filamentous fungi, fruit bodies, and infected plant tissues. Nucleic Acids Res. 20: 6115-6116.
152
Pogoda, F., and S. Werres (2002) Pathogenicity of European and American P. ramorum isolates to rhododendron. [online] In Proceedings of the Sudden Oak Death Science Symposium: The State of Our Knowledge. December 15-18, 2002, Monterey, Calif. Available from http://danr.ucop.edu/ihrmp/sodsymp/poster/poster26.html. [updated November 12, 2002; accessed August 08, 2007]. Schena, L., K.J.D. Hughes, and D.E.L. Cooke (2006) Detection and quantification of Phytophthora ramorum, P. kernoviae, P. citricola and P. quercina in symptomatic leaves by multiplex real-time PCR. Mol. Plant Pathol. 7: 365-379. Themann, K., S. Werres, R. Lüttmann, and H.-A. Diener (2002) Observations of Phytophthora spp. in water recirculation systems in commercial hardy ornamental nursery stock. Eur. J. Plant Pathol. 108: 337-343. Tomlinson, J.A., N. Boonham, K.J.D. Hughes, R.L. Griffin, and I. Barker (2005) On-site DNA extraction and real-time PCR for detection of Phytophthora ramorum in the field. Appl. Environ. Microbiol. 71: 6702-6710. Tooley, P.W., F.N. Martin, M.M. Carras, and R.D. Frederick (2006) Real-time fluorescent polymerase chain reaction detection of Phytophthora ramorum and Phytophthora pseudosyringae using mitochondrial gene regions. Phytopathology, 96: 336-345. Villalba Mateos, F., M. Rickauer, and M.T. Esquerré-Tugayé (1997) Cloning and characterization of a cDNA encoding an elicitor of Phytophthora parasitica var. nicotianae, that shows cellulose-binding and lectin-like activities. Mol. PlantMicrobe Interact. 10: 1045-1053. Werres, S., and D. De Merlier (2003) First detection of Phytophthora ramorum mating type A2 in Europe. Plant Dis. 87: 1266. Werres, S., and K. Kaminski (2005) Characterisation of European and North American Phytophthora ramorum isolates due to their morphology and mating behaviour in vitro with heterothallic Phytophthora species. Mycol. Res. 109: 860-871. Werres, S., and B. Zielke (2003). First studies on the pairing of Phytophthora ramorum. J. Plant Dis. Protect. 110: 129-130. Werres, S., R. Marwitz, W.A. Man in't Veld, A.W.A.M. de Cock, P.J. Bonants, M. de Weerdt, K. Themann, E. Ilieva, and R.P. Baayen (2001) Phytophthora ramorum sp. nov., a new pathogen on Rhododendron and Viburnum. Mycol. Res. 105: 11551165.
153
Chapitre IV
SNP Discovery and Multilocus Strain Genotyping in Phytophthora ramorum.
154
SNP
Discovery
and
Multilocus
Strain
Genotyping
in
Phytophthora ramorum. G. J. Bilodeau1, C. A. Lévesque2 and R. C. Hamelin1 1
Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, P.O. Box
3800, 1055 du P.E.P.S., Québec, Quebec G1V 4C7, Canada; 2
Agriculture and Agri-Food Canada, National Program on Environmental Health -
Biodiversity, 960 Carling Avenue, Ottawa, Ontario K1A 0C6, Canada;
155
IV.1 Résumé/Abstract IV.1.1
Résumé
Les outils moléculaires pour le génotypage sont importants pour mieux comprendre la biologie des populations et des mouvements de l’agent pathogène le Phytophthora ramorum. La disponibilité des séquences du génome complet de cet organisme, depuis 2004, ajoute aux ressources avec plus de 16000 gènes prédits. Une stratégie employant la notion de volatilité des codons pour la sélection des gènes candidats pour les études de populations et la conception d’amorce a donc été développée pour des marqueurs de sites polymorphes (SNPs) sur des gènes candidats avec une importance fonctionnelle. Les SNPs représentent la catégorie la plus abondante de polymorphismes dans le génome. La volatilité des codons est une mesure qui peut permettre la détection de la sélection se basant sur les séquences d’un seul génome, selon la vraisemblance qu’une substitution nonsynonyme origine d’une mutation silencieuse ou non-synonyme. À l’aide d’analyses bioinformatiques et du séquençage de l’ADN, nous avons découvert plus de 86 SNPs, en plus de séquences d’insertions et de délétions, totalisant 13 gènes. Les gènes étant les plus polymorphiques ont été retrouvés dans la catégorie avec la plus haute volatilité à 64% et avec la plus faible volatilité avec 36%. De plus, trois des quatre loci les plus polymorphiques (plus de 9 SNPs) se retrouvent dans le groupe de la haute volatilité. Dans une collection de 100 isolats provenant d’Europe et d’Amérique du nord, nous avons identifié des profils de SNPs qui sont distincts et sont fortement en corrélation avec l’origine géographique. Les populations du P. ramorum en Amérique du nord consistaient en trois profils multilocus nord-américains avec une forte clonalité ainsi que trois isolats présentant deux génotypes de profils européens. Quinze profils européens observés en Europe sont basés sur les SNPs des différentes séquences des loci analysés. De plus, malgré qu’il y ait moins de génotypes multilocus observés en Amérique du nord qu’en Europe la diversité nucléotidique parait assez comparable, probablement parce que la dominace de deux génotypes multilocus en Europe et de la présence d’un génotype divergent en Amérique du nord. Cette étude ajoute de nouvelles informations sur la variabilité des séquences des isolats du P. ramorum des différentes collections mondiales des Phytophthora.
156
IV.1.2
Abstract
Molecular genotyping could be an important tool to better understand the population biology and movements of pathogens such as Phytophthora ramorum. The availability of the complete genome sequences of this pathogen, since 2004, provides a resource of approximately 16000 predicted genes. A strategy based on codon volatility to design single nucleotide polymorphism (SNP) markers within candidate genes of functional importance was developed. Single nucleotide polymorphisms represent the most abundant type of sequence variation in genomes. The codon volatility is a method to detect selection on the basis of a single genome sequence, by assessing the likelihood that a non-synonymous substitution originated from a silent or non-synonymous mutation. Using bioinformatics analyses and DNA sequencing, we uncovered more than 86 SNPs, in addition to insertions and deletions in 13 genes. More polymorphisms were found in the group of genes with high volatility (64%) than in the group with low volatility (36%). In addition 3 of the 4 most polymorphic genes (comprising more than 9 SNPs) were in high volatility groups. Nonsynonymous mutations were observed equally in the low and high volatility groups. In a collection of 100 P. ramorum samples from Europe and North-America, we identified SNP profiles that were distinct and mostly correlated with geographic origin. Populations of P. ramorum in North-America consisted of three North-American multilocus genotype profiles and appeared to be clonally-derived. Fifteen multilocus DNA profiles were observed in samples from Europe. Two multilocus DNA profiles originating from North America had DNA profiles that matched European profiles. Although far less multilocus genotypes were observed in North America than in Europe, overall nucleotide diversity were comparable, probably because of the dominance of two multilocus genotypes in Europe, and the presence of a divergent multilocus genotype in North America. This study adds new information and new tools for the study of molecular epidemiology of P. ramorum.
157
IV.2 Introduction Phytophthora ramorum (Werres et al. 2001) causes the disease known as “sudden oak death” (SOD) and is responsible of the death of thousands of red oaks in California and Oregon (Rizzo et al. 2002). This pathogen is also present in nurseries in many countries in Europe, in British Columbia, Canada and several states in the U.S. More than one hundred host plant species are affected by this disease but various symptoms can be produced, which can make diagnostic difficult (Rioux et al. 2006). Quarantine measures are in effect in several countries to prevent the spread of the pathogen. In addition, eradication of this pathogen has been attempted in several U.S. states and in B. C. One of the reasons for the success of Phytophthora ramorum is that it can be propagated by spores in water, soil and plant products. But it appears that exchange of infected nursery material is the most efficient means of dispersal. There are still many unanswered questions about the epidemiology of SOD. Population genetics can provide important data to improve our understanding of SOD epidemiology and provide tools for regulation.
Multiple morphological variants were found in the American population of P. ramorum instead of Europe (Pogoda and Werres 2002). The data reported suggests that the American and European populations are adaptively different (Brasier 2003). European and North American isolates were found to be genetically distinct, as indicated by different DNA fingerprints (AFLP) (Ivors et al. 2004), microsatellite profiles (Garbelotto et al. 2002; Ivors et al. 2006; Prospero et al. 2004b; Prospero et al. 2007) and Inter-Simple Sequence Repeats (ISSR) (Wiejacha et al. 2007). Although the results so far did not allow the investigators to determine the source of the epidemics more DNA variation was found in the European than in the North American populations. This suggests that a population bottleneck occurred in North America probably due to more exchanges and variations in nurseries than in forest (Ivors et al. 2006). Allele specific oligonucleotides (ASOs) were developed recently using SNPs in β-tubulin and CBEL coding regions. This assay was used on pure cultures and field samples and allowed to differentiate European and North American genotypes (Bilodeau et al. 2007). However, these SNP assays did not allow the
158 study of intra-group variability. The availability of the complete genome of P. ramorum (Tyler et al. 2006) (http://genome.jgi-psf.org/Phyra1_1/Phyra1_1.home.html), made it possible to develop large numbers of microsatellites. Using an array of microsattellites, a distinct new genotype was discovered in the U.S. (US4, clade 3) (Ivors et al. 2006). This finding indicated that the variability in North America was higher than previously thought, but that there are a still large portions of the population that is clonal.
Single nucleotide polymorphism (SNP) is the most abundant type of sequence variation in genomes and are valuable genetic markers for revealing the evolutionary history of populations. Microsatellites have a high mutation rate (10-4), mutation rates are lower by several orders of magnitude in SNPs (10-8-10-9). Mutation rates in microsatellites are difficult to estimate and vary across loci and alleles on the same locus (Brumfield et al. 2003). While microsattelites can have multiple alleles at a single locus, SNPs are restricted to four characters (A, T, C and G) and mostly to 2 alleles per locus. When used for parentage and fine scale geographic structure analysis they also require more loci than microsatellites. However, for population analyses SNPs can be complementary to microsatellites and reveal patterns about selection and neutrality at candidate loci and about allele phylogeny.
Approximately 16,000 genes were predicted from the P. ramorum genome for 66.6 Mbp. The possibility to develop SNPs at multiple loci is an interesting prospect for use in population studies. However, with so many genes available, an important question remains which genes should be sampled for SNP analysis? One criterion for population studies should be to sample loci that have the most likelihood of displaying polymorphisms. One possible strategy is to use codon volatility (Plotkin et al. 2004) to select genes presenting more potential to display variability. Codon volatility is a method to detect selection on the basis of a single genome sequence based on non-synonymous substitutions. If a nucleotide sequence for a region coding for a protein implicates an excessive numbers of amino-acid substitutions, the region will on average contain an excess of volatility codons compared to the whole genome. For each 61 codons, the volatility is defined by a proportion of neighbouring point mutations that encode different amino-acids. The codon volatility is
159 used to quantify the probability than the most recent mutation accepted to a site caused an amino-acid change. With this concept, the entire genome can be scanned to find genes that show significantly more, or less, pressure for amino-acid substitution than the whole genome. Genes can then be ranked with regard to their volatility. Volatile genes can be hypothesized to have a higher probability of containing polymorphisms and indicate strong diversifying selection.
The objectives of our research were 1- to scan the genome of P. ramorum using codon volatility and rank genes according to their volatility; 2- to select the genes (loci) at the extremes of the distributrion of codon volatility, and to compare polymorphisms (SNPs); 3- sequence this panel of genes on different P. ramorum isolates from a worldwide collection to assess their variability and to measure and compare nucleotidic diversity and divergence in P. ramorum in North America and Europe.
IV.3. Materials and methods
IV.3.1
Isolates and DNA extraction
All isolates of Phytophthora used in this study are listed in Table 1 with information on number of isolates, country of origin, plant host, collection source and the genotypes as determined in this work. The P. ramorum panels comprised 100 isolates of European (72) and North American (28) origin isolated from different Phytophthora collections from around the world. Mycelium was grown and DNA was extracted following the procedures described in (Decock et al. 1992) or (Moller et al. 1992). Samples from Slovenia and CBS collection were extracted from dried mycelium using DNeasy Plant mini kit (Qiagen) protocol according to the manufacturer’s recommendations.
160
IV.3.2
Codon volatility genome determination
A file of all transcripts (gene models) of from the full genome of P. ramorum was downloaded (http://genome.jgi-psf.org/Phyra1_1/Phyra1_1.download.ftp.html), from P. ramorum annotation release 1.0. The software used on web base (Codon Volatility Computation Server v1.0) at: http://volatility.cgr.harvard.edu/cgi-bin/volatility.pl could only analyse genome with less of 10 Mbp. For the P. ramorum genome with 66.6 Mbp, we used an attached source code and compiled it and ran it on your own computer. This attached source code (Codon Volatility Analysis Software ver 1.0beta) was provided by Joshua B. Plotkin, University of Pennsylvania Departments of Biology & Computer Science Lynch Laboratories, PA, USA. The P-value can be calculated using a randomization procedure or by direct formulae based upon a normal approximation. This software uses the normal approximation, which produces P-values that are virtually identical
to
those
obtained
by
the
randomization
procedure
(http://volatility.cgr.harvard.edu/moreinfo.html). This program was run with UNIX interface with the 16 066 transcript sequences. This program (using a Perl script) first cleans sequences in multiple of 3 bp for codons. Sequences were separated with .fasta.cd for fine sequences and .fasta.bcd for “bad” coding sequences. Sequences with ambiguous nucleotides (different of ATCG) or that did not end in frame were considered bad sequences and were excluded from the volatility computation program. After cleaning, 10252 sequences were tested in the volatility program using a Kappa of 1.0. The output of this analysis was a text file (.txt) with the gene name, gene length, observed volatility per residue, expected volatility per residue, the variance in volatility per residue and finally the volatility P-value for each gene analysed.
IV.3.3
Selection of genes with high and low codon volatility
All the sequences analysed with the volatility program were ranked according to their P-value. The genes with higher volatility had P-values close to 0 and the genes with the lowest volatility had values close to 1 (Figure 1). Fifteen genes were selected from the
161 lower and higher extremes of the codon volatility scale to design primers for PCR amplification and sequencing.
IV.3.4
Primer design
The primers were designed using the software Primer Premier 5.00 (Premier Biosoft International, Palo Alto, CA). The selection criteria were the following: Tm (melting temperature) 55-65°C, primer length 17-23 bp, and absence of secondary structure whenever possible (Table 2). A total of 36 primer pairs were designed to amplify 24 of these 30 genes selected, 14 with high volatility and 10 with lowers (Table 1). Six genes did not permitted to designed primers with right amplicon size and and no primer dimmers. The β-tubulin and CBEL primers (Bilodeau et al. 2007) were also used to obtained DNA sequences becaused they also showed SNPs and had information at the genomic population study. The list of final primers used for the 11 loci (22 primers) are annotated in the Table 2 in boldface. Primer length and annealing temperature used are presented in Table 2.
IV.3.5
PCR amplification
PCR was performed with a DNA Engine (PTC-200) Peltier Thermal Cycler (BioRad Laboratories, Inc., Hercules, CA). The PCR reactions were prepared in 20 μL volume reactions containing 0.5 μM of each primer (Table 2), 1X of PCR reaction buffer without MgCl2 (Invitrogen Corporation, Carlsbad, CA), 1.25 units Platinum® Taq DNA Polymerase (Invitrogen), 1.5 mM MgCl2 (Invitrogen), 0.15 mM dNTP Set (Invitrogen), sterile mQ-water and template DNA. PCR cycling conditions were set at 95°C for 3 min, 36 cycles at 94°C for 45 s, annealing at temperature from 55 to 65°C for 30 s, depending of the set of primers used (Table 2), and 72°C for 30 s, and a final extension at 72°C for 8 min. The DNA template concentration used in the reaction was determined with a Nanodrop spectrophotometer ND-1000 (Nanodrop Technologies, Wilmington, DE) and ranged from 0.05 to 60 ng/μL for P. ramorum. Eight samples were tested for this
162 amplification, four from U.S. and four from Europe. Amplicons were run on 1.5 % agarose gel, TAE buffer, 110 volts with 100bp DNA ladder and checked for the number of bands and quality of amplification.
IV.3.6
PCR-Single-strand conformation polymorphism (SSCP)
Single-Strand Conformational Polymorphism (SSCP) gels were made to screen genes for polymorphisms between North American (U.S.) and European strain without sequencing all amplicons. The amplicons were analysed by SSCP (Orita et al. 1989) following a protocol adapted from (Bagley et al. 1997), and described elsewhere (Hamelin et al. 2005; Joly 2005). The parameters used were 0.5 × MDE gel solution (Cambrex Bio Science Rockland Inc., Rockland, ME), and 0.6 × tris-borate- EDTA (TBE) buffer. Specific running conditions, including temperature, electrical power, and migration time, were 10°C, 2 Watts (9 mA and 210 V) an ran 18 hours. After migration, DNA was visualized by silver staining using the Bio-Rad silver stain kit (Bio-Rad Laboratories, Hercules, CA), according to manufacturer’s protocol with the following modifications: all fixative steps were done without acetic acid, the duration of the two last developing steps was 2.5 min instead of 5 min, and that of the stopping step was 30 min instead of 5 min.
Distinct SSCP mobility profiles (consisting of two bands) (Figure 2), indicated different alleles which were confirmed by direct sequencing using an ABI 3730xl DNA Analyzer (Applied Biosystems, Foster City, CA) for all alleles discovered.
IV.3.7
Sequence analysis, SNP, and haplotype determination
Sequences were edited and analyzed using software SeqMan Pro version 7.2.0 (DNASTAR Inc., Madison, WI). Sequences were analysed on each chromatogram aligned to find polymorphisms. Phytophthora is a diploid organism and SNPs were detected by the
163 presence of heterozygous polymorphisms (2 peaks on sequence chromatogram). SNPs were identified and multiple alleles were cloned for haplotype determination. Moreover, after sequencing confirmation of the polymorphism detected by SSCP, all isolates tested (100) were sequenced for SNP detection for each locus. For cloning, the QIAGEN PCR Cloning kit (Qiagen, Valencia, CA) was used. A ratio of five times more PCR product DNA than pDrive Cloning Vector DNA was used for ligation. The ligation protocol was use according to manufacturer’s protocol with the modifications: 1 hour at 16°C on thermal cycling block (PTC-200) and ligation-reaction mixtures were stored at -20°C until used. The transformation protocol was used according to manufacturer’s protocol with the following modifications: a volume of 0.76 μL of ligation-reaction mixture was added with 19 μL of QIAGEN EZ Competent Cells. We added this to 100 μL of SOC medium at room T°C and plated 50 μL of this mixture onto LB agar plates. Plates were incubated overnight at 37°C and put at 4°C for 2h minimum. Four white colonies were taken with a 5 ul tips and amplified by PCR using the same PCR conditions than PCR amplifications (see above) using one colony instead of total DNA template. All amplicons were run on 1.5 % agarose gel, TAE buffer, 110 volts with 100bp DNA ladder. Amplicons for each colony were sequenced to obtain haplotypes sequences. All haplotypes were confirmed with the heterozygous sequences. All sequences of haplotypes were also aligned with the software MegAlign version 7.2.0 (DNASTAR Inc., Madison, WI) and BioEdit sequence alignment editor version 7.0.2 (Hall, T.A., Ibis Therapeutics, a division of Isis Pharmaceuticals, Carlsbad, CA) to display the SNPs and insertions/deletions.
IV.3.8
Sequence and population analyses
All genes presenting polymorphisms were analysed with the program DnaSP version 04.50.0.2 (University of Barcelona, Barcelona, Spain). Isolates representing NorthAmerica (NA) and Europe (EU) were analysed and compared for SNP frequency, DNA polymorphisms, heterozygosity, and the nucleotide diversity (π).. All analyses were
164 conducted using each haplotype for each genotype isolate. However, a clone correction was applied to avoid redundant sampling due to the high clonality in populations, mostly in North America (Ivors et al., 2006). The level of differentiation between NA and EU was calculated by comparing polymorphic sites, fixed and shared mutations, and the ratio of synonymous and non-synonymous mutations. Recombination and clonality were assessed by determining the number of haplotypes and by calculating the recombination coefficient. For the gene set studied, mutations were classified as synonymous and non-synonymous. To test whether or not there were differences in the lowest and higherst volatility groups, student t-tests were used. Genotype frequencies and heterozygosities were also calculated into all loci and each individual selected using sample sequences and allele genotypes using Excel (version 9.0.3821 SR-1; Microsoft, Redmond, Wash.).
IV.4 Results
IV.4.1
Phytophthora ramorum codon volatility
After the 16066 gene transcripts from the P. ramorum genome were screened for ambiguous nucleotides and absence of stop codons, 10252 genes were retained for analysis. The 10252 sequences analysed with the volatility program yielded a frequency distribution that is skewed toward the extreme values, indicating an over-representation of high volatility and low volativility genes. A total of 5107 sequences had a P-value < 0.5 and 5145 with a P-value > 0.5. As described previously by Plotkin et al. (2004), with the genome sequence of Mycobacterium tuberculosis and Plasmodium falciparum, we observed sequences with high volatility with P-value ranging from 0.0-0.1 and low volatility with a P-value ranging from 0.9-1.0 (Figure 1).
165
IV.4.2
Primers and PCR amplification
A total of 36 primer pairs were designed to amplify 14 genes with high volatility and 10 with lower volatility (Table 2). All primer sets were tested on a panel comprising four strains from each of Europe and North America. Samples that yielded multiple bands were presumed to have paralogues or be part of gene families or introns and were excluded from further sequence analysis (Figure 3). A total of 25 primer sets presenting only one band were analysed by SSCP. Genes with polymorphism (11) were identified on SSCP and were further analyzed by DNA sequencing. All isolates (100) were then sequenced for determination of polymorphisms.
IV.4.3
Sequences
Sequences obtained for each locus were deposited in GenBank (Supplement material 1). We uncovered 86 single nucleotide polymorphisms (SNP) in 13 genes (total of 6.3 kb) (Table 3). Of the 86 SNPs based on the annotations of the JGI, 48 (57%) were synonymous, and 37 (43%) were non-synonymous. Introns were present in two genes but polymorphisms were found only in one of the introns. Moreover, six indels were found in the different isolates. The size of these indels ranged from 3 bp to 102 bp and they were tandemly repeated (as in SSR). In the 6 genes with the highest nucleotide diversity, 5 were from the highest volatility group (Table 3). However, locus 79140 (low volatility) had the highest nucleotide diversity and also the largest number of mutations (15) and nonsynonymous mutations (9). The ratio of nucleotide diversity of non-synonymous mutation πa on nucleotide diversity of synonymous mutation πs and the ratio of the number of nonsynonymous/synonymous mutations (dN/dS) were calculated. One locus (81813, a high volatility gene) had a dN/dS ratio>1. All other genes had ratios lower than 1.00.
Polymorphisms were found in seven of the 11 genes selected in the highly volatile genes (64%) and in four of low volatility genes (36%). In addition three of the four most
166 polymorphic genes (more than 9 SNPs) were located in high volatility group. The number of polymorphisms was also higher with the group with the highest volatility. Student-T test performed on nucleotide diversity, number of mutation in different population and on the heterozygosity did not reveal differences for any of these variables between high and low volatility loci (data not shown). The only exception was for the number of mutations that was significantly higher in loci in the group with high volatility than with low volatility genes.
Eleven of the genes sequenced in the populations revealed polymorphisms that differed between European and North American populations as well as polymorphisms within populations. The comparison of polymorphims between the two populations revealed a similar level of diversity. The European population possessed 64 SNPs compared to 57 for the North American population. The average nucleotide diversity was similar in the two populations (Table 4). No fixed difference was detected between the North American and European populations at the 13 loci studied. A total of 35 mutations were shared by both populations, 29 were polymorphic only in the European population and 22 only in the North-American. The number of haplotypes (i.e. alleles) ranged from 2 to 6 and averaged 4.6. Only two loci presented potential of recombination into the locus sequences (Table 5).
In a collection of 72 samples from Europe and 28 from North-America genotyped at all 13 loci, we identified SNP profiles that were distinct and mostly correlated with geographic origin (Table 6, Figure 4). The multilocus genotypes show that populations of P. ramorum in North-America (excepted for two Oregon 2003 and one Canadian isolates that are believe to be distinct recent introductions and associated to the EU4 and EU1 genotype) are consisted of three unique SNP profiles (Table 6, Figure 4). In Europe fifteen different genotypes were identified (Table 6). The multilocus genotypes with the highest frequency were EU1 (55/72) in Europe and NA1 (20/28) in North America. The most divergent genotype was NA2 which possessed unique alleles at 10 loci. At several loci, this multilocus genotype possesses two unique alleles in an heterozygous state. However, the North American isolates comprised only three multilocus genotypes but present more
167 alllele changes betwen NA and also EU isolates. More than 18 changes were related between the NA1 and NA2 and EU genotypes.
Some loci, e.g. 83989, and 79140 display three different haplotypes (alleles) for some genotypes. Since this pathogen is supposed to be diploid, this represents an anomaly. It is possible that these loci have paralogues in the genome, or that some strains are polyploids.
The heterozygosity per individual (Hi), (data not shown here) (Avise 2004), and average heterozygosity over all individuals (H), were calculated The H was 0.684 in Europe, 0.878 in North-America and 0.796 overall. When two Oregon and one Canadian isolates with European profiles were excluded, the heterozygosity was 0.909. The number of genotypes found per locus was higher in North-America (35) than in Europe (31) but was lower when the genotype of the three isolates in North-America presenting a European SNP profile was excluded from this group (25). A total of 52 genotypes were counted when all loci were considered. Three loci presented only one genotype in Europe; the locus 79140 showed more allele in Europe than in North-America. The locus 83000 had less polymorphisms and heterozygosity between the different isolates tested.
By homology searches in GenBank, several of the loci studied were assigned to functions related to cell membranes, such as cellulobio-hydrolase cell wall protein, GPI anchored cell wall protein, and putative glycine-rich protein (GRPs). A weak homology was denoted for the locus 79072 for the MAEBL, a minor type 1 membrane protein. Two loci showed weak similarity with dentin sialophosphoprotein and three loci presented homology for membrane glycoprotein, such as fibronectin. The pinin protein homology was also denoted. Finally, one locus showed homology with membrane transporter as ABC-type transport proteins. No distinction was observed between low and high volatility gene selected and the association with a cell wall protein.
168
IV.5 Discussion Codon volatility has been reported to provide a measure of the potential of selection based on a single genome (Plotkin et al. 2004). Our objective was to test the hypothesis that genes from the high volatility group have a higher probability to be polymorphic and could provide a quick in silico screen to generate a panel of markers for population studies. Overall, our results show that this hypothesis is partly supported. The hypothesis that polymorphism should be more present in most volatile gene was supported for the total of polymorphisms found with all genotypes. However, sequence variations in the different populations and results on the nucleotide diversity and heterozygosity did not support this hypothesis. Therefore, although more polymorphisms were identified in the higher volatility group, the actual diversity did not differ between the two groups. Codon volatility has been criticised since publication of the concept. Questions as to how this concept can be applied to measure selection using have been raised (Dagan and Graur 2005; Friedman and Hughes 2005; Nielsen and Hubisz 2005; Pillai et al. 2005; Sharp 2005; Stoletzki et al. 2005; Zhang 2005). To our knowledge, this is the first study in which this concept was tested with biological material by taking the study from the in silico analysis to the development of molecular markers for population studies. This method is a first screen to identify genes presenting polymorphisms for subsequent use in our genotyping.
The majority of loci selected were homologous to proteins implicated in cell wall protein surface or interaction. This could be significant as these are believed to be involved in host-pathogen recognition and therefore could be subjected to positive selection. These GRP proteins are known to be components of the cell walls of many higher plants, the MAEBL, a minor type 1 membrane protein (is a single transmembrane-like protein) identified in Plasmodium yoelii and P. falciparum where erythrocyte binding activity expressed in the apical organelles and on the surface of invasive merozoites (Preiser P 2004). The dentin sialophosphoprotein is known to be an extracellular matrix protein (Chen et al. 2004) and the glycoprotein, such as fibronectin or PE-PGRS glycine-rich proteins (Komatsuzawa et al. 2000) are extracellular adhesion molecules. The pinin protein is a
169 major component of the epithelial intracellular gap junctional complex, in animal cell (Ouyang and Sugrue S.P. 1996) or fmtB gene encoding a cell wall-associated protein. Finally the ABC-type transport proteins, known to be membrane permeases. The β-tubulin and CBEL genes were also implicated in cell wall functions.
That is interesting because cell wall protein (as GPI anchored cell wall protein) is often implicated in host-pathogen interactions. The polymorphisms in those genes could represent variation allowing the pathogen to overcome defences of the plant host and could have an implication in pathogenicity or aggressiveness of the Phytophthora pathogen agent (Jang et al. 2006). This is supported by the fact a large number of mutations (37 of the 86 SNP) were non-synonymous and therefore should result in difference in phenotypes. These genes could be implicated in selection and evolution of the organism. By comparing the rate of non-synonymous substitutions, dN, on synonymous substitutions, dS, respectively, it is possible to obtain evidence of adaptative evolution on a protein coding gene. Loci with ratios under one suggest a purifying selection. Locus 81813 (with homology to putative glycine-rich protein) had a ratio above 1, which could indicate a recurrence of diversifying selection (Anisimova 2007). In this study the locus 81813 could suggest a neutral evolution with ratio of 1 and no effect on fitness. Many indels were detected and often seemed to represent minisatellites of portion of the coding region. These indels could be implicated in cell wall proteins, with replicates of some amino-acid sequences that permit to modify the extracellular cell wall protein.
The sample US4 (Ivors et al. 2006) known to be NA2 genotype often presented more indel polymorphisms compared to all other genotypes. This genotype also revealed polymorphisms shared by North American NA1 at locus 79310 (homologous to the pinin). It was also associated to the European EU1 or other genotypes in locus 79140 (homologous to a membrane glycoprotein). In general NA2 possessed more alleles than all other genotype which resulted in its placement on a third clade in previous works using microsatellites (Ivors et al. 2006). The indels observed in this strain were frequently in a heterozygous state. Because each population of P. ramorum was believed to be clonal (a
170 genetic copy), further population analyses used a clone correction with both haplotypes of each genotype.
The partial sequences of the target genes were obtained for all of the 100 isolates from NA and EU to reveal their genotype. Two main genotypes (EU1 and NA1) representing two clone groups were identified. Fifteen EU genotypes were found in our analyses compared to three NA genotypes. This is consistent with previous work with microsatellites where more numerous multilocus genotypes were observed in Europe than in North America (Ivors et al. 2006; Ivors et al. 2004; Prospero et al. 2004a). However, not all isolates tested in this assay were similar to previous isolates used in the studies of Ivors et al. 2006. Our data set shared only the US1, US2, US4 and EU1, EU3, EU5 with the study of Ivors et al. 2006. Other isolates were used by that group and therefore the results cannot be directly compared. However our study uncovered more genotypes in Europe than in the previous microsatellite work (fifteen vs seven). The number of loci studied was comparable with 13 coding region sequences compared to 12 microsatellite loci. In general all EU genotypes were presenting a maximum of 3 different genotypes. EU1 formed a node of frequent genotypes from which originated most of the rare EU genotypes. Those differed from EU1 by only a few alleles. NA1 and NA3, by contrast, were differentiated from EU1 by different alleles at 12 loci. But NA1 and NA3 differed at only one locus. The NA2 genotype differed from most other genotypes at 10 loci. It possessed 12 unique alleles that distinguished it from the other genotypes.
Comparison of divergence in Europe and North America revealed more polymorphism and more genotypes in Europe than in North America. The dominance of distinct clones of P. ramorum is in agreement with the hypothesis that distinct introductions occurred on the two continents and suggests that the NA and EU populations underwent strong founder effects (Brasier 2003). The origin of P. ramorum is still unknown. But this study supports the hypothesis that it could be from sources other than Europe and North America. Although the discovery of larger number of multilocus genotypes in Europe than in North America appears to indicate a more diverse population, the overall nucleotidic diversity was comparable in these populations, reflecting a similar evolutionary potential.
171 Indeed, the haplotype network clearly shows that although more multiplocus genotypes are present in Europe, most differ only at a few loci. However, it is not clear whether or not these genetically distinct populations are also adaptively different. The possibility of an Asian origin has been proposed (Kluza et al. 2007) but so far there is no evidence to support this hypothesis as the pathogen has not been discovered there. Three isolates were recovered from nurseries in Oregon and British Columbia in 2003. They possessed multilocus genotypes matching those of the most frequent European genotypes (Bilodeau et al. 2007; Ivors et al. 2006). However, those interceptions were isolated genotypes and did not apparently participate in the larger epidemic of P. ramorum. For the sake of comparing diversity in Europe and North America, we excluded them from our analysis on the premise that they represent recent accidental introductions and were immediately eradicated before becoming established. Interestingly the sample isolated from a Canadian nursery had a genotype matching EU1, the most frequent European genotype but the two Oregon nursery isolates had genotypes matching EU4 a less frequent genotype found in France and Slovenia. These genotypes differed from EU1 only for the locus 81804 (homologous to fibronectin). Previous work using SNP genotyping at two loci (CBEL and β-tubulin) (Bilodeau et al. 2007) showed that the majority of samples from the Canadian survey possessed a NA multilocus genotype. That could indicate that these interceptions could originate from different location.
Some loci (e.g. 79140 and 83989) yielded more than two alleles for some EU isolates and the NA2 sample. Since P. ramorum is generally believed to be diploid, this could be explained by presence of trisomy (polyploidy), gene duplication or introgression from homologous regions. These possibilities were also observed for P. ramorum at microsattelite loci (Ivors et al 2006). This phenomenon is known and documented in Phytophthora species. The presence of numerous copies in close proximity along with remnant sequences indicates multiple gene duplication events, occurring at different times. This was seen with endoglucanase that possess multiple gene copies (Costanzo et al. 2007) and Glucose-6-phosphate isomerase (Gordon et al. 2007; Ospina-Giraldo & Jones 2003). Multiploidy was also reported Phytophthora infestans (Goodwin et al. 1992; Goodwin et
172 al. 1998; van der Lee et al. 2004). Interestingly, recombination events were detected at the two loci that contained more than two alleles (83989 and 79140). One possible explanation for this observation is that multiple gene copies are present or introgression occurred from other Phytophthora. Phytophthora ramorum is a heterothallic organism and needs two mating types to produce oospores. Sexual reproduction was never found to produce viable progeny in P. ramorum, suggesting that P. ramorum is reproducing basically via nonsexual reproduction (zoospores and chlamydospores) in Europe and North America.
The SNP discovery and genotyping study reported here confirms results from previous population studies using AFLP (Ivors et al. 2004) and microsatellites (Ivors et al. 2006). A study using SNP with ASO on the β-tubulin and CBEL loci on P. ramorum from a worldwide collection as well as on epidemic strains from British Columbia revealed SNPs that could distinguish isolates from North America and Europe (Bilodeau et al. 2007). When these SNPs were genotyped on the isolate panel used in the current study, SNPs at the CBEL locus also yielded a distinct genotype for NA2 and could therefore be used in diagnostic assays to differentiate among the three major clades described here and in previous work (Ivors et al. 2006). In NA2, the SNP 245, was homozygous (G/G) and SNP 412, was heterozygous (G/A). In NA1, both SNPs were homozygous while in EU1, both SNPs were heterozygous (Bilodeau et al. 2007). ASO primers were also designed and tested (not shown) using some of the loci described here and could be used to identify the additional multilocus genotypes observed. This technique (ASO) could be valuable tool to quickly determine genotypes using SNPs without sequencing. They were used successfully on environmental samples without prior culturing. Additionally, during the development of the primers, some displayed the presence of multiple bands on agarose gels. This could also yield a simple and cheap assay that could be used as complement to differentiate between isolates with EU or NA genotypes without DNA sequencing.
One main advantage here of using SNP polymorphism in genotyping the different P. ramorum populations is the establishment of a molecular barcode (fingerprint) using all SNPs for each locus to obtain a multilocus genotype. This can be used to associate each individual to a genotype and a geographic origin. Such information could contribute
173 important epidemiological information on the strains studied. The codon volatility provided an interesting tool to determine which genes to select in finding polymorphism and permitting to differentiate P. ramorum populations. The codon volatility concept did not show clear evidence of more diversity in highly volatile genes, but allowed to select genes presenting SNPs revealing the different genotypes tested in our collection. The genotypes already characterized with microsatellites and AFLP correlated with our work. This project adds potential valuable information of new DNA coding sequences which can be used in molecular epidemiology studies. This should enhance our capacity to characterize P. ramorum and further refine diagnosis and quarantine efforts to stop the spread of this major plant pathogen.
174
IV.6 Acknowledgements We thank Josée Grondin, Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, for the her help on PCR amplification and DNA sequencing and Marie-Josée Bergeron for is help with SSCP and cloning techniques. Special thanks to people that provide DNA for these research: Arthur .W.A.M. de Cock, from Centraalbureau voor Schimmelcultures, The Netherlands; Metka Žerjav and Alenka Munda Agricultural Institute of Slovenia, Slovenia; Sabine Werres, from Federal Biological Research Centre for Agriculture and Forestry, Germany; Anne Chandelier, Agricultural Research Centre, Department Biological Control and Plant Genetic Resources, Belgium; Claude Husson, INRA Nancy, France; Micheal D. Coffey from University of California, Riverside; Stéphan C. Brière from CFIA, Ottawa, Canada. Funding for this work was received from Canadian Biotechnology Strategic Fund and the CBRNE Research and Technology Initiative (CRTI grant 04-0045RD).
175
Table 1. Isolates of Phytophthora ramorum from different culture collections used in this study with genotypes bases on SNP profiles. Other isolate Isolate number Location Plant host Genotype o reference no. h CBS101326 a PD 98/8/6933 The Netherlands Rhododendron sp. EU1 CBS 101327 a i adc 98.36 The Netherlands, Rhododendron sp. EU1 Boskoop CBS 101328 a PD 98/8/6743 The Netherlands, Rhododendron sp. EU1 Bentveld CBS 101329 a i PD 98/8/6285 The Netherland, Rhododendron sp. EU1 Lisse CBS 101330 a i PD 98/8/5233 The Netherlands, Viburnum sp. EU1 Hazerswoude CBS 101331 a PD 98/8/2627 The Netherlands, Rhododendron sp. EU1 Ede CBS 101332 a PD 94/844 The Netherlands, Rhododendron sp. EU1 Roosendaal CBS 101548 a BBA 69082 Germany Rhododendron, hybrid EU1 'Schneewolke' CBS 101549 a BBA 104/5 Germany Recycling water, nursery EU1 CBS 101550 a i BBA 14/98-a Germany Rhododendron catawbiense EU2 Michx. CBS 101551 a BBA 12/98 Germany Rhododendron catawbiense EU1 CBS 101552 a BBA 9/3 Germany Recycling water, nursery EU1 CBS 101553 a i BBA 9/95 (ex type Germany Rhododendron catawbiense EU1 strain), CBS 101554 a BBA 2/4 Germany Recycling water, nursery EU1 CBS 109278 a i BBA 16/99 Germany Viburnum ×bodnantense EU1 CBS 109279 a i BBA 13/99-1 Germany Rhododendron sp. EU1 CBS 110534 a l Pr01, BBA-PR01 USA, California, Quercus agrifolia Née NA1 Marin Co. CBS 110535 a l Pr03 USA, California Lithocarpus densiflora NA1 (Hook. & Arn.)Rehd. CBS 110536 a Pr04 USA, California Quercus kelloggii Newberry NA1 CBS 110537 a l Pr52, BBAPR52, USA, California Rhododendron sp. NA1 USA217 CBS 110538 a l Pr65 USA, California Quercus parvula Greene NA1 CBS 110539 a l Pr70 USA, California Vaccinium ovatum Pursh NA1 CBS 110540 a l Pr71 USA, California, Quercus agrifolia NA1 Sonoma Co. CBS 110541 a l Pr86 USA, California Arbustus menziesii Pursh NA1 CBS 110542 a l Pr110 USA, California Umbellularia californica NA1 (Hook. & Arn.)Nutt. CBS 110543 a Pr159 USA, California Lithocarpus densiflora NA1 CBS 110544 a PRJL3.5.3, PR 342 USA, California Sequoia sempervirens (D. NA1 Don) Endl. CBS110545 a Rh/2/00 Poland Rhododendron sp. EU1 CBS 110546 a Rh/6/00 Poland Rhododendron sp. EU1 CBS 110547 a Rh/122/98 Poland Rhododendron sp. EU1 CBS 110548 a det.nr.2002-195 France Rhododendron sp. EU1 CBS 110601 a Pr84-sz USA, California Soil EU1 CBS 110900 a CRA2386, Belgium Rhododendron sp. EU1 BBA27/02 CBS 110901 a i CRA2338, Belgium Viburnum sp. EU1
176
CBS 110953 a
BBA26/02 Pr36
CBS 110954 a
Pr72
CBS 110955 a m
Pr97
CBS 110956 a l
Pr106
CBS 110957 a CBS 111762 a F001 b F002 b F004 b F005 b F006 b 2N386 b k 2N389 b j 3N5-1 b 3N91 b 3N208-3 b 4N200-1 b 4N203-6 b 4N372-2 b 4N461-2 b 4N501-3 b 4N657-2 b 4N662-1 b 4N717-4 b 4N773-3 b 4N780-1 b 4N793-3 b 4N992-1 b POL1 b POL2 b A326 c n A1379 c n
Pr158 8895 BBA 10/02
A1382 c 05-119 d 05-053-1 d 05-054-2 d 05-078 d 05-079-3 d 05-422-9 d 06-080-1 d 06-046 d 06-043-3 d 06-027 d 06-114 d 06-037-1 d 06-038-3 d 06-024-1 d 06-318-1 d 06-351-1 d
PD20031690 PD20031691 BBA3N0005-A
PR-79, P10090 c P11047, WSDA 3403, PR-04-014 c P11048 , WSDA 3765, PR-04-015 c
USA, California, Sonoma Co. USA, California, Alameda Co. USA, California, Napa Co. USA, California, Sonoma Co. USA, Oregon Norway France France France France France France France France France France France France France France France France France France France France France France Pologne Pologne USA, California USA, Washington
Quercus agrifolia
NA1
Rhododendron macrophyllum
NA1
Quercus agrifolia
NA1
Umbellularia californica
NA1
Lithocarpus densiflora Rhododendron catawbiense Rhododendron sp. Rhododendron sp. Rhododendron sp. Rhododendron sp. Rhododendron sp. Viburnum x bodnantense Viburnum tinus L. Viburnum tinus Rhododendron sp. Rhododendron sp. Rhododendron sp. Rhododendron sp. Rhododendron sp. Viburnum tinus Rhododendron sp. Viburnum opulus Rhododendron sp. Viburnum tinus Rhododendron sp. Rhododendron sp. Rhododendron sp. Viburnum x bodnantense Calluna vulgaris (L.)Hull. Photinia sp. Vaccinium ovatum Rhododendron sp.
NA1 EU1 EU3 EU1 EU1 EU1 EU1 EU1 EU15 EU1 EU4 EU5 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 EU1 NA3 NA2
USA, Washington
Rhododendron sp.
NA2
Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia Slovenia
Viburnum x bodnantense Rhododendron sp. Rhododendron catawbiense Rhododendron sp. Rhododendron sp. Viburnum x bodnantense Rhododendron sp. Rhododendron sp. Rhododendron sp. Rhododendron repens Rhododendron sp. Rhododendron sp. Rhododendron repens Viburnum tinus Kalmia angustifolia L. Rhododendron sp.
EU15 EU6 EU7 EU8 EU9 EU10 EU1 EU11 EU1 EU1 EU1 EU1 EU4 EU1 EU12 EU13
177 06-268-2 d BBA 104 e BBA 24/02 e BBA 22/01-5 e i BBA 4/02-4 e i BBA PR 05m e BBA PR 88 e BBA PR 84 e l Strain 3206 f Strain 3211 f 1270626-1 g 1341211-4 g CSL 1715 g CSL 1620 g
PRI 658 Rh/6/01, PD200110055 P2512, Phy ram3, PD20023443/3 Pr05 PD20018933PD21008933 PRI 487
Slovenia Germany Italy
EU14 EU1 EU1
Poland
Rhododendron catawbiense Recycling water in nursery Rhododendron x yakushimanum Rhododendron sp.
Spain (Mallorca)
Rhododendron sp.
EU1
USA, USA,
Lithocarpus densiflora Umbellularia califonica
NA1 NA1
USA, California Belgium (southern) Belgium (southern) USA, San Luis Obispo, California USA, Sacramento, California UK UK
Soil Rhododendron sp. Viburnum bodnatense Pieris japonica (Thunb.)D.Don. Rhododendron sp.
NA1 EU1 EU15 NA2
Rhododendron sp. Viburnum sp.
EU1 EU1
EU1
NA2
MSOD105 g i MSOD03105 Canada Rhododendron sp. EU1 USA, Oregon Pieris sp. EU4 Pr03-74-1 g i USA, Oregon Viburnum sp. EU4 Pr03-74-2 g i a From Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands; b From INRA Nancy, France; c From the World Phytophthora Collection, University of California, Riverside, USA; d From Agricultural Institute of Slovenia, Slovenia; e From Federal Biological Research Centre for Agriculture and Forestry, Germany; f From Agricultural Research Centre, Department Biological Control and Plant Genetic Resources, Belgium; g From Canadian Food Inspection Agency, Pest DNA Diagnostics Laboratory, Centre for Plant Quarantine Pests, Ottawa, Ont. h The different reference numbers were identified in: Information on Phytophthora isolates (Kaminski and Werres 2007). i Previously genotyped EU1, (Ivors 2006) with microsatellites j Previously genotyped EU3. k Previously genotyped EU5. l Previously genotyped US1. m Previously genotyped US2. n Previously genotyped US4. o Based on SNPs and allele profiles (Table 6).
178
H
81813
H
82924
H
79133
H
83000 79140
79586
80426
H L
L
H
5' CGCCTTGAACGACACCA 3' 5' TGTAGAACTTGCTACGGTTGG 3'
547
56
X
X*
83987_87U
5' CGCCTTGAACGACACCA 3'
508
56
X
X*
83987_594L
5' GGCGGCATTTCCAACAA 3'
83987_610U
5' ACGCCAACCGTAGCAAG 3'
404
59
X
X
X
83987_1013L
5' TAGTAGGTCGGGAGGCACT 3' 429
56
X
X*
X
501
56
X
X
522
56
X
424
59
X
405
56
296
59
X
333
58
M
443
d
514
d
579
56
X
463
54
M
420
60
X
651
60
M
274
60
X
385
60
d
333
60
X
79072_1500U
5' GGAGCGAATGAGGCGTAT 3'
79072_1928L
5' GATGGAGGGCGGAACAC 3'
79072_106U
5' GCTGAAAGCAGCGTGAGG 3'
79072_606L
5' CTTCATTCCAGGCGAACT 3'
78537_465U
5' CTCGGTCTCCATCATCAGC 3'
78537_465U
5' CTCGGTCTCCATCATCAGC 3'
83989-82U
5' AGCCTCCGTACCGAGCAA 3'
83989-505L
5' CGGCATCGTCCGCACTT 3'
83989-422U
5' ACACCGACACGGAAGGC 3'
83989-826L
5' CCAAGAAACGCTCACCTTC 3'
81813-232U
5' ATGCAAGGCGGCTCTGT 3'
81813-527L
5' TCGCTGCTTGGGATGTTT 3'
82924-20U
5' TTGTATGTTCGACCACCG 3'
82924-352L
5' CCGTCTCCACTCCTGCTA 3'
79133-3U
5' GCGTCCGAGTTACGTTGT 3'
79133-445L
5' GCGTCCGAGTTACGTTGT 3'
79133-430U
5' ACACCGGCCTCAGGATC 3'
79133-943L
5' CTTCGCTTGACCCTGACGCA 3'
79133-927U
5' GTCAGGGTCAAGCGAAGA 3'
79133-1505L
5' AAGGCGTCGTTCCACCA 3'
83000-620U
5' GGCAACCTGACTACCGC 3'
83000-1083L
5' CAACGCAGACTTTGGACT 3'
79140_194U
5' TGGCTGCGGAGAAGAAA 3'
79140_613L
5' CGCTCGTGGATGAACCTG
79140_581U
5' AGACGCAGTCTACCGCAGGTT 3'
79140_1231L
5' CGTTAAGCGTGTTGAAAGTCA
79586_78U
5' GACGGAACCAGGGAACG 3'
79586_351L
5' GTGCGGAACTGGAGGAG 3'
79586_622U
5' AAGGAGGCAGGAATCACA 3'
79586_1006L
5' CGTATTCGCTCGGGTGC 3'
80426_53U
5' TCACCACCTGCGACACTG 3'
SNPs polymorphisms
83987_87U 83987_633L
Only one band on gel
Tested on SSCP
83989
L
More than on band on gel
78537
L
Annealing T°C
79072
Amplicon size bp
H
Primer sequence
Codon volatility a
83987 b
Primer names
Locus name
Table 2. Primers used for amplification and DNA sequencing
X
X
X*
X
X*
X
X
X X
X
X
179 80426_385L
81496
L
81804
L
82129
L
82939
H
82917 83046 83053 79803 82919 78393
H H H H H H
79310
L
81879
L
78478
L
5' TGGCGCTAGACTCCACAT 3'
80426_387U
5' TTCTGACGGTCGTCGTTT 3'
80426_731L
5' CTGTTGCTACTCGCATTACTTT 3'
81496_259U
5' ATCACTGGGTCCACTTGGG 3'
81496_485L
5' TCCGTGAGCGTGGTCGTCTG 3'
81496_283U
5' GTGCTGCCGTCCGTAGGTT 3'
81496_837L
5' TGTAGTTGACGGAACTGAGGTTT 3'
81804_197U
5' AGCCGTCGCTCTTCTCG 3'
81804_640L
5' GGTCTTCCACGCATCCAA 3'
82129_2336U
5' AGTATCCTGGCGGTATGTGC 3'
82129_2751L
5' CGGATTCAGAGCCGTAAGA 3'
82939_381U
5' TGTTACTCCCGATGAAGGC 3'
82939_718L
5' ATTCGATGCGGTAGATGC 3'
82939_76U
5' CTGCGAGGCGTCTGCTA 3'
82939_401L
5' ATGCCTTCATCGGGAGTA 3'
82917_215U
5' AAGAAAGCAGCAGCACG 3'
82917_622L
5' TCTCGCCAGTGTCATCC 3'
83046_21U
5' TGCTCTTCTTATCGCTGTCG 3'
83046_672L
5' CCCTGCTCCACCATTTCC 3'
83053_103U
5' ACGAGTGGTGGTCTCCCTG 3'
83053_586L
5' TTCCTGGCCCGTAACCC 3'
79803_66U
5' GCTGGGCACGAACGAAA 3'
79803_432L
5' AGCACCCGCTCCGACAA 3'
82919_7U
5' ATCTCCCACATCATCGCTCTT 3'
82919_439L
5' CCTTCTTCTCGCCTTCCTG 3'
78393_909U
5' CGACATCAGCGAAGAGTGG 3'
78393_1297L
5' TTCCCTGCCGTTTATTGC 3'
78393_3705U
5' ATTGCTGCGGAACTATCCTA 3
78393_4076L
5' TCCAAGCTCGTAAATCGTCT 3'
79310_1020U
5' GTTTGTCAATTTGGTGCCT 3'
79310_1518L
5' TCGTTCGGATACATTCTGG 3'
81879_78U
5' CCTGTTGTCGGTGTTGATG 3'
81879_421L
5' GATGTACCTGTCTTGGCTGA 3'
78478_1744U
5' GCGAAGGTGATTCCTGCTA 3'
78478_2161L
5' TCATGTGAGAATTATTTGGGTG 3'
345
60
X
X
227
60
X
X*
554
60
X
X
444
60
X
X
416
60
X
X*
338
60
X
X
326
60
X
X*
408
60
X
652
60
X
X*
484
55
X
X*
367
60
433
60
389
60
X
X
372
60
X
X
499
60
X
X
347
55
X
X*
417
60
X
X
X
X
X
X
X* X
X
* = worse on SSCP; d = do not amplified; M = multiples band; X = tested H=High Volatility (Value P < 0.5); L=Low Volatility (Value P > 0.5) b Locus and primers in boldface were the locus with polymorphisms. They are the primer sets used for the population genomic analyses. a
180
Table 3. Characteristics of Phytopthora ramorum genes sequenced Locus name
βtubulin
CBEL
83987
79133
81813
83000
83989
83046
82939
79072
79140
81804
79310
Volatility a Homology
βtubulin
Cellulose Binding elicitor lectin
H Cellulobiohydrolase, cell wall protein
H GPI anchored cell wall protein
H Putative glycinerich protein
H Dentin sialophosphoprotein
L Fibronectin
L Pinin or fmtB protein
618
369
535
261
414
H ABCtype transport protein s 302
L membrane glycoprotein
866
H PEPGRS glycinerich protein 614
L MAEBL membrane protein
Fragment size bp Nb Mut. (SNPs)
H Malate dehydrogenase and dentin sialophosphopreotein 572
394
456
409
461
6271
2
3
10
6
4
2
9
8
13
3
15
4
7
86
Syn. Mut. Non-Syn. Mut
2 0
3 0
7 3
4 2
1 3
1 0
6 3
6 2
5 8
2 1
6 9
2 2
3 4
48 37
Indels
-
-
1 (30bp) b
2 (45, 12bp)
-
1 (3bp)
1 (102bp)
1 (6 bp)
-
1
-
c
Introns 1.15
2.37
10.74
5.58
8.62
2 (one mutation) 1.17
πa NonSyn. πs*Syn.
0
0
2.96
2.01
8.85
0.98
2.68
1.32
9.70
1.44
13.23
2.29
2.84
4.71
9.39
34.60
17.25
8.36
0
16.30
14.27
14.42
5.28
31.57
3.82
8.84
πa/ πs
0.00
0.00
0.09
0.11
1.06
-
0.18
0.09
0.67
0.27
0.42
0.60
0.32
π* *
1 5.36
4.60
10.62
2.36
14.82
2.65
4.43
0.00 0.07 0.15 1.15 0.14 0.10 0.62 0.31 0.43 0.60 0.28 ω =dN/dS 0.00 (ka/ks) * X 10-3, π = nucleotide diversity ; a H=High Volatility (Value P < 0.5); L=Low Volatility (Value P > 0.5); - not determined with codon volatility (CBEL situated in H and β-tubulin not determined). b Number of bp concerned. c Based on genome annotation and tblasx on GenBank.
Total
−
181
Table 4. Comparison of divergence in European and North American populations of P. ramorum Locus name βCBEL 83987 79133 81813 83000 83989 83046 82939 tubulin Volatility a H H H H H H H Number of 4 seq. analysed (Total) Population 1: Europe Number of 2 seq. analysed (haplotypes) Number of 1 mutations (pol. sites): π*: 1.15 Population 2: America Number of 2 seq. analysed (haplotypes) Number of 1 mutations (pol. sites): π* : 1.15
79072
79140
81804
79310
L
L
L
L
Mean
6
8
10
8
6
11
8
10
8
14
12
10
2
4
6
4
2
7
4
6
4
9
6
8
3
5
6
4
0
8
5
7
1
15
2
7
64
4.85
7.40
6.28
7.66
0
6.26
4.88
7.88
1.27
14.86
1.68
4.42
5.33
4
4
4
4
4
4
4
4
4
5
6
2
2
7
4
3
2
3
5
7
2
15
3
3
57
1.89
10.67
4.53
5.75
1.75
4.06
4.88
13.51
2.96
15.79
3.02
6.51
5.88
* X 10-3, π = nucleotide diversity ; a H=High Volatility (Value P < 0.5); L=Low Volatility (Value P > 0.5); - not determined with codon volatility.
182
Table 5. Polymorphism comparison between North American and European populations of P. ramorum Locus name βCBEL 83987 79133 81813 83000 83989 83046 82939 79072 tubulin Volatility a H H H H H H H L
79140
81804
L
L
L
Fixed differences
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Polymorphic in EU, monomorphic in NA b Polymorphic in NA, monomorphic in EU Shared Mutations
1
1
3
2
1
0
6
3
6
1
0
1
4
29
1
0
5
0
0
2
1
3
6
2
0
2
0
22
0
2
2
4
3
0
2
2
1
0
15
1
3
35
3
4
5
5
3
2
5
6
5
4
6
6
6
0
0
0
0
0
0
1
0
0
0
3
0
0
Number of haplotypes Recombination Recombination location a b
140247
H=High Volatility (Value P < 0.5); L=Low Volatility (Value P > 0.5); - not determined with codon volatility. EU, Europe; NA, North-America
72-79; 119-165; 168-318
79310
Total
183
Table 6. Multilocus genotypes of P. ramorum from Europe and North America at 13 loci Genotypes N a CBS number From β-tubulin CBEL 83987 79133 81813 83000 EU1 55(56) 101553 Germany AB AB AB AB AA AA EU2 1 101550 Germany AB AB AA BB AA AA F001 France AB AB AB FB AA AA EU3 1 3N91 France AB AB AB AB AA AA EU4 2(4) 3N208-3 France -b AB AB AB AA AA EU5 1 05-053-1 Slovenia AB AB AB AB AA AA EU6 1 05-054-2 Slovenia AB AB AB AB AA EU7 1 05-078 Slovenia AB AB AB AB AA EU8 1 05-079-3 Slovenia AB AB AA EU9 1 05-422-9 Slovenia AB AA AA EU10 1 06-046 Slovenia AB AB AB AB AA AA EU11 1 06-318-1 Slovenia AB AB AA AA EU12 1 06-351-1 Slovenia AB AA AA EU13 1 06-268-2 Slovenia AB AB AB AA AA EU14 1 2N389 France AB AB AB AB AA AA EU15 3 110534 USA AC c AC CC CC BB AA NA1 20 NA3 1 A326 USA AC AC CC CC BB AA A1379 USA AC AD d DE DA BC AB NA2 4 MSOD105 Canada AB AB AB AB AA AA EU1 1 Pr03-74-1 USA AB AB AB AA AA EU4 2 a Number in parenthesis were associated at the three North-American samples with European profile. b – No sequences or low quality sequences were obtained for these samples. c Italic genotype showed the polymorphism in genotype with the EU1 genotype. d Underlined genotype shown the polymorphism in genotype with the NA1 genotype. e Possibility of three alleles with some genotypes with this locus. European isolates were over the hashed line and North-American isolates under.
83989 e ABE ABE ABE ABE ABE AC AC ABE AA AC ABE AC AA AC ABE AC AC AD ABE ABE
83046 BF BF BF BF BF BF BF BF BF BF BF AA BF BF DE DE AC BF BF
82939 AA AA AA AA AB AA AA AA AA AA AA AA AA AA AC AA AA DE AA AA
79072 AB AB AB AB AB AB AA AB AB AB AA AB AB AB CD CD AC AB AB
79140e ABC CC ABC ABC AB ABC ABC ABC ABC AB BB ABC ABC ABC ABC DE DE ABC ABC ABC
81804 AB AB AB AA DD AB AB AB AB AB AB AC AB AB AB AD AA EF AB AA
79310 AB AB AB AB EF AB AB AB AB AB AB AB AA AD AB AC AC AC AB AB
184
Table 7. Genotype frequencies and heterozygosis at 13 different loci in and between European and North-American population. 83987
79133
81813
83000
83989
83046
82939
79072
79140 c
81804
79310
H
H
H
H
H
H
H
L
L
L
L
64 AB 0.969 BB 0.016 DA 0.000 FB 0.016 CC 0.000 0.984
71 AA BB AB BC
0.944 0.000 0.028 0.014 0.014
71 AB 0.944 AD 0.000 AA 0.028 AC 0.014 DD 0.014 0.958
72 AB AC AD AA EF 0.986
N 28 26 28 Genotype AB 0.107 AB 0.038 AB 0.107 frequency AC 0.893 AC 0.808 AA 0.000 AD 0.154 CC 0.750 DE 0.143
28 AB BB CC DA FB
28 AA BB BC AB
0.250 0.750 0.000 0.000 0.000
28 AB AC EF AD AA
Hjd
0.250
0.143
28 AB 0.036 AD 0.714 EF 0.143 AA 0.107 AC 0.000 DD 0.000 0.893
N 95 92 100 Genotype AB 0.737 AB 0.728 AB 0.740 frequency AC 0.263 AC 0.228 AA 0.010 AD 0.043 CC 0.210 DE 0.040
92 AB BB CC DA FB
99 AA BB BC AB
Hj d
0.761
Loci
βCBEL tubulin Volatility a Population 1: Europe (Nmax : 72 b) N 67 66 Genotype AB 1.000 AB 1.000 frequency AC 0.000 AC 0.000 AD 0.000 Hj d
1.000
1.000
72 AB AA DE CC
0.986 0.014 0.000 0.000
0.986
69 72 0.972 AA 1.000 ABE 0.000 AB 0.000 AC 0.028 AA 0.000 AD
0.028
0.000
0.903 0.069 0.028 0.000
0.972
70 BF DE AA AC
0.986 0.000 0.014 0.000
72 AA DE AB AC
0.944 0.000 0.014 0.042
71 AB AC CD AA
0.972 0.000 0.000 0.000
0.986
0.056
0.972
27 BF DE AC AA
28 AA DE AC AB
27 AB AC AA CD
72 ABC DE AB BB CC 0.972
Mean
69.9 0.958 0.000 0.014 0.014 0.014
0.763e
Population 2: North America (Nmax : 28 )
1.000
1.000
0.250
0.107 0.000 0.750 0.143 0.000
28 28 0.107 AA 0.857 ABE 0.750 AB 0.143 AC 0.143 AD 0.000 AA 0.143
0.107 0.750 0.143 0.000
1.000
0.111 0.741 0.148 0.000
0.857 0.143 0.000 0.000
0.111 0.148 0.000 0.741
28 ABC DE CC AB BB
1.000
0.143
1.000
1.000
97 BF DE AC AA
100 AA DE AC AB
98 AB AC AA CD
100 ABC DE CC AB BB
27.7 0.107 0.893 0.000 0.000 0.000
1.000
0.679e
All isolates (both populations) (Nmax : 100)
a
1.000
1.000
0.780
0.707 0.011 0.228 0.043 0.011
97 100 0.727 AA 0.959 ABE 0.212 AB 0.041 AC 0.040 AD 0.020 AA
0.061
0.041
0.980
0.680 0.260 0.040 0.020
0.742 0.206 0.041 0.010
0.990
0.920 0.040 0.030 0.010
0.080
0.735 0.041 0.020 0.204
0.738 e
H=High Volatility (Value P < 0.5); L=Low Volatility (Value P > 0.5); - not determined with codon volatility. Nmax is the number of isolated tested in this project. N could be different cause by some isolate with no or awful sequences for few locus. c Triploid locus. d Heterozygosity per loci Hj (Lievens et al.) e Heterozygosity of the locus, H b
0.980
0.750 0.210 0.010 0.020 0.010
99 AB AD EF AA AC DD 0.949
0.687 0.202 0.040 0.051 0.010 0.010
AB AC EF AD AA
100 97.6 0.720 0.250 0.010 0.010 0.010
0.990
0.738e
185
1600
Number of genes
1400 1200 1000 800 600 400 200 0 0.0-0.1
0.1-0.2
0.2-0.3
0.3-0.4
0.4-0.5
0.5-0.6
0.6-0.7
0.7-0.8
0.8-0.9
0.9-1.0
Volatility P-value
Figure 1. Distribution of volatility of 10252 genes of P. ramorum. At left side, high volatility with P-value close to 0, at right side, low volatility with P-value close to 1.
186
a
b
Figure 2. SSCP profiles of eight P. ramorum isolates, four from Europe and four from North America, with different migration profiles. Two alleles were detected on; (a) the gene 83987 and (b) the gene 81804.
187
L
a
b
Figure 3. Agarose gel electrophoresis of the PCR amplicon of the gene 83987 with primers 83987-87U and 83987-594L on 1.5% agarose gel, TAE buffer 1X, stained with ethidium bromide. The two bands were obtained for all European samples (a); one single band was obtained from the North American samples (b). L;100 bp DNA ladder (Invitrogen, Carlsbad, CA).
188
Clade 3
Clade 2 Clade 1
Figure 4. Unweighted pair group dendrogram presenting the different genotypes. Each change in different allele found was represented by the line length and the frequence of each genotypes is represented by the circle size. European Isolate is white circle and North American by grey circle. The EU1 and EU4 present some proportion of NA isolates. Three distinct clades are represented.
189
IV.7 References Anisimova, M., and D.A. Liberles (2007) The quest for natural selection in the age of comparative genomics. Heredity, 99: 567-579. Avise, J.C., (2004) Molecular Markers, Natural History, and Evolution, Second Edition. Sinauer Associates, Inc., Sunderland, MA. Bagley, M.J., J.F. Medrano, and G.A.E. Gall (1997) Polymorphic molecular markers from anonymous nuclear DNA for genetic analysis of populations. Mol. Ecol., 6(4): 309320. Bilodeau, G.J., C.A. Lévesque, A.W.A.M. de Cock, S.C. Brière, and R.C. Hamelin (2007) Differentiation of European and North American genotypes of Phytophthora ramorum by real-time polymerase chain reaction primer extension. Can. J. Plant Pathol., 29(4): 408-420. Brasier, C., (2003) Sudden oak death: Phytophthora ramorum exhibits transatlantic differences. Mycol. Res., 107: 258-259. Brumfield, R.T., P. Beerli, D.A. Nickerson, and S.V. Edwards (2003) The utility of single nucleotide polymorphisms in inferences of population history. Trends Ecol. Evol., 18(5): 249-256. Chen, S., A. Unterbrink, S. Kadapakkam, J. Dong, T.T. Gu, J. Dickson, H.-H. Chuang, and M. MacDougall (2004) Regulation of the Cell Type-specific Dentin Sialophosphoprotein Gene Expression in Mouse Odontoblasts by a Novel Transcription Repressor and an Activator CCAAT-binding Factor. J. Biol. Chem., 279(40): 42182-42191. Costanzo, S., M. Ospina-Giraldo, K. Deahl, C. Baker, and R. Jones (2007) Alternate intron processing of family 5 endoglucanase transcripts from the genus Phytophthora. Curr. Genet., 52 (2-3) 115-123. Dagan, T., and D. Graur (2005) The comparative method rules! Codon volatility cannot detect positive Darwinian selection using a single genome sequence Mol. Biol. Evol., 22(4): 1158-1158. deCock, A.W.A.M., A. Neuvel, G.Bahnweg, J.C.J.M. deCock, and H.H. Prell (1992) A Comparison of Morphology, Pathogenicity and Restriction Fragment Patterns of Mitochondrial-DNA among Isolates of Phytophthora-Porri Foister. Neth. J. Plant Pathol., 98(5): 277-289.
190 Friedman, R., and A.L. Hughes (2005) Codon volatility as an indicator of positive selection: Data from eukaryotic genome comparisons Mol. Biol. Evol., 22(4): 11581158. Garbelotto, M., D.M. Rizzo, K. Hayden, M. Meija-Chang, J.M. Davidson, and S. Tjosvold (2002) Phytophthora ramorum and sudden oak death in California: III. Preliminary studies in pathogen genetics. USDA For. Serv. Gen. Tech. Rep., PSW-GTR-184: 765-774. Goodwin, S.B., L.J. Spielman, J.M. Matuszak, S.N. Bergeron, and W.E. Fry (1992) Clonal Diversity and Genetic Differentiation of Phytophthora-Infestans Populations in Northern and Central Mexico. Phytopathology, 82(9): 955-961. Goodwin, S.B., C.D. Smart, R.W. Sandrock, K.L. Deahl, Z.K. Punja, and W.E. Fry (1998) Genetic change within populations of Phytophthora infestans in the United States and Canada during 1994 to 1996: Role of migration and recombination. Phytopathology, 88(9): 939-949. Gordon, S.G., K. Kowitwanich, W. Pipatpongpinyo, S.K.S. Martin, and A.E. Dorrance (2007) Molecular marker analysis of soybean plant introductions with resistance to Phytophthora sojae. Phytopathology, 97(1): 113-118. Hamelin, R.C., M. Allaire, M.-J. Bergeron, M.-C. Nicole, and N. Lecours (2005) Molecular epidemiology of white pine blister rust: Recombination and spatial distribution. Phytopathology, 95(7): 793-799. Ivors, K., M. Garbelotto, I.D.E. Vries, C. Ruyter-Spira, B.T. Hekkert, N. Rosenzweig, and P. Bonants (2006) Microsatellite markers identify three lineages of Phytophthora ramorum in US nurseries, yet single lineages in US forest and European nursery populations. Mol. Ecol., 15(6): 1493. Ivors, K.L., K.J. Hayden, P.J.M. Bonants, D.M. Rizzo, and M. Garbelotto (2004) AFLP and phylogenetic analyses of North American and European populations of Phytophthora ramorum. Mycol. Res., 108: 378-392. Jiang, R.H.Y., B.M. Tyler, and F. Govers (2006) Comparative analysis of Phytophthora genes encoding secreted proteins reveals conserved synteny and lineage-specific gene duplications and deletions. Mol. Plant Microbe In., 19(12): 1311-1321. Jiang, R.H.Y., B.M. Tyler, and F. Govers (2006) Comparative analysis of Phytophthora genes encoding secreted proteins reveals conserved synteny and lineage-specific gene duplications and deletions. Mol. Plant Microbe In., 19(12): 1311-1321. Joly, D., (2005) Analyse de la diversité génomique des populations de Cronartium ribicola, agent responsable de la rouille vésiculeuse du pin blanc. Thesis, Université Laval, Québec.
191 Kaminski, K., and S. Werres (2007) Information on Phytophthora ramorum isolates, Federal Biological Research Centre for Agriculture and Forestry (BBA), Braunschweig,Germany. Kluza, D.A., D.A. Vieglais, J.K. Andreasen, and A.T. Peterson (2007) Sudden oak death: geographic risk estimates and predictions of origins. Plant Pathol., 56(4): 580-587. Komatsuzawa, H., K. Ohta, M. Sugai, T. Fujiwara, P. Glanzmann, B. Berger-Bachi, and H. Suginaka (2000) Tn551-mediated insertional inactivation of the fmtB gene encoding a cell wall-associated protein abolishes methicillin resistance in Staphylococcus aureus. J. Antimicrob. Chemother., 45(4): 421-431. Lievens, B., L. Claes, D.J. Vakalounakis, A. Vanachter, and B. Thomma (2007) A robust identification and detection assay to discriminate the cucumber pathogens Fusarium oxysporum f. sp cucumerinum and f. sp radicis-cucumerinum. Environ. Microbiol. 9(9): 2145-2161. Moller, E.M., G. Bahnweg, H.Sandermann, and H.H. Geiger (1992) A Simple and Efficient Protocol for Isolation of High-Molecular-Weight DNA from Filamentous Fungi, Fruit Bodies, and Infected-Plant Tissues. Nucleic Acids Res., 20(22): 6115-6116. Mousavi, A., and Y. Hotta (2005) Glycine-rich proteins, A class of novel proteins. Appl. Biochem. Biotech., 120(3): 169-174. Nei, M., and W.-H. Li (1979) Mathematical Model for Studying Genetic Variation in Terms of Restriction Endonucleases. Proc. Natl. Acad. Sci. USA, 76(10): 52695273. Nielsen, R., and M.J. Hubisz (2005) Evolutionary genomics: Detecting selection needs comparative data. Nature, 433(7023): E6. Orita, M., H. Iwahana, H. Kanazawa, K. Hayashi, and T. Sekiya (1989) Detection of Polymorphisms of Human DNA by Gel-Electrophoresis as Single-Strand Conformation Polymorphisms. Proc. Natl. Acad. Sci. USA, 86(8): 2766-2770. Ospina-Giraldo, M.D., R.W. Jones, (2003) Characterization of the glucose-6-phosphate isomerase gene in Phytophthora infestans reveals the presence of multiple alleles. Fungal Genet Biol., 40(3): 197-206. Ouyang, P., and S.P. Sugrue (1996) Characterization of Pinin a novel protein associated with the desmosome-intermediate filament complex. J. Cell. Biol., 135: 1027-1042. Pillai, S.K., S.L.K. Pond, C.H. Woelk, D.D. Richman, and D.M. Smith (2005) Codon volatility does not reflect selective pressure on the HIV-1 genome. Virology, 336(2): 137-143.
192 Plotkin, J.B., J. Dushoff, and H.B. Fraser (2004) Detecting selection using a single genome sequence of M. tuberculosis and P. falciparum. Nature, 428(6986): 942-945. Pogoda, F., and S. Werres (2002) Pathogenicity of European and American P. ramorum isolates to rhododendron. In Sudden Oak Death Science Symposium. Monterey, California. December 15 - 18, 2002. Preiser P, R.L., N. Singh, B. Balu, W. Jarra, T. Voza, O. Kaneko, P. Blair, M. Torii, I. Landau, and J.H. Adams (2004) Antibodies against MAEBL ligand domains M1 and M2 inhibit sporozoite development in vitro. Infect Immun., 72(6): 3604-3608. Prospero, S., J.A. Black, and L.M. Winton (2004a) Isolation and characterization of microsatellite markers in Phytophthora ramorum, the causal agent of sudden oak death. Mol. Ecol. Notes, 4(4): 672-674. Prospero, S., J.A. Black, and L.M. Winton (2004b) Isolation and characterization of microsatellite markers in Phytophthora ramorum, the causal agent of sudden oak death. Mol. Ecol. Notes, 4(4): 672. Prospero, S., E.M. Hansen, N.J. Grunwald, and L.M. Winton (2007) Population dynamics of the sudden oak death pathogen Phytophthora ramorum in Oregon from 2001 to 2004. Mol. Ecol., 16(14): 2958-2973. Rioux, D., B. Callan, and D. McKenney (2006) Phytophthora ramorum: causal agent of sudden oak death, ramorum blight, ramorum bleeding canker, ramorum (shoot) dieback. , CFIA & CFS. Rizzo, D.M., M. Garbelotto, J.M. Davidson, G.W. Slaughter, and S.T. Koike (2002). Phytophthora ramorum as the cause of extensive mortality of Quercus spp. and Lithocarpus densiflorus in California. Plant Dis., 86(3): 205-214. Sharp, P.M., (2005) Gene "volatility" is most unlikely to reveal adaptation. Mol. Biol. Evol., 22(4): 807-809. Stoletzki, N., J. Welch, J. Hermisson, and A. Eyre-Walker (2005) A dissection of volatility in yeast. Mol. Biol. Evol., 22(10): 2022-2026.
193
Tyler, B.M., S. Tripathy, X.Zhang, P. Dehal, R.H.Y. Jiang, A. Aerts, F.D. Arredondo, L. Baxter, D. Bensasson, J.L. Beynon, J. Chapman, C.M.B. Damasceno, A.E. Dorrance, D. Dou, A.W. Dickerman, I.L. Dubchak, M. Garbelotto, M. Gijzen, S.G. Gordon, F. Govers, N.J. Grunwald, W. Huang, K.L. Ivors, R.W. Jones, S. Kamoun, K. Krampis, K.H. Lamour, M.-K. Lee, W.H. McDonald, M. Medina, H.J.G. Meijer, E.K. Nordberg, D.J. Maclean, M.D. Ospina-Giraldo, P.F. Morris, V. Phuntumart, N.H. Putnam, S. Rash, J.K.C. Rose, Y. Sakihama, A.A. Salamov, A. Savidor, C.F. Scheuring, B.M. Smith, B.W.S. Sobral, A. Terry, T.A. Torto-Alalibo, J. Win, Z. Xu, H. Zhang, I.V. Grigoriev, D.S. Rokhsar, and J.L. Boore, (2006) Phytophthora genome sequences uncover evolutionary origins and mechanisms of pathogenesis. Science 313, 1261-1266. van der Lee, T., A. Testa, A. Robold, J. van 't Klooster, and F. Govers (2004) High-Density Genetic Linkage Maps of Phytophthora infestans Reveal Trisomic Progeny and Chromosomal Rearrangements. Genetics, 167(4): 1643-1661. Werres, S., R. Marwitz, W.A. Man-in'-t-Veld, A.W.A.M. de Cock, and P.J. Bonants (2001) Phytophthora ramorum sp. nov., a new pathogen on Rhododendron and Viburnum. Mycol. Res., 105(10): 1155-1165. Wiejacha, K., A. Trzewik, L.B. Orlikowski, G. Szkuta, and T. Orlikowska (2007) Genomic polymorphism of isolates of Phytophthora ramorum from Polish ornamental nurseries compared with other European and North American isolates. J. Appl. Genetics, 48(4): 413-419. Zhang, J.Z., (2005) On the evolution of codon volatility. Genetics, 169(1): 495-501.
194
IV.8 Supplement material
195
Supplement material 1. Isolates of Phytophthora ramorum used in this study with the Genbank accession number. Genbank accession no. Isolate no. CBEL β83987 79133 81813 83000 83989 83046 82939 79072 a tubulin CBS101326 CBS101327 CBS101328 CBS101329 CBS101330 CBS101331 CBS101332 CBS101548 CBS101549
EF117945 EF117946 EF117947 EF117948 EF117949 EF117950 EF117951 EF117964 EF117952
CBS101550
b
c
EF117930 EF117931 EF117924 EF117925 EF117932 EF117926 EF117933 EF117934 EU688880
EU688096 EU688097 EU688098 EU688099 EU688100 EU688101 EU688102 EU688103 EU688104
EF117953
EF117935
EU688105
EU688674 EU688675 EU688676 EU688677
CBS101551 CBS101552 CBS101553
EF117954 EF117955 EF117956
EF117936 EF117937 EF117938
EU688106 EU688107 EU688108
CBS101554 CBS109278 CBS109279 CBS110534
EF117957 EF117958 EF117959 EF117965
EF117939 EF117940 EF117941 EF117915
CBS110535 CBS110536 CBS110537 CBS110538 CBS110539 CBS110540 CBS110541 CBS110542
EF117966 EU688908 EF117967 EF117971 EF117968 EU688909 EF117972 EF117973
CBS110543 CBS110544
EF117969 EF117970
d
EU688666 EU688667 EU688668 EU688669 EU688670 EU688671 EU688672 EU688673
79140
81804
79310 EU687904 EU687905 EU687906 EU687907 EU687908 EU687909 EU687910 EU687911
EU688491 EU688492 EU688493 EU688487 EU688494 EU688495 EU688496 EU688497
EU688193 EU688194 EU688195 EU688196 EU688197 EU688198 EU688199 EU688200 EU688201
EU688290 EU688291 EU688292 EU688293 EU688294 EU688295 EU688296 EU688297 EU688298
EU688584 EU688585 EU688586 EU688587 EU688579 EU688588 EU688589 EU688590 EU688591
EU687999 EU688000 EU688001 EU687994 EU688002 EU688003 EU688004 EU688005
EU687809 EU687810 EU687811 EU687812 EU687813 EU687814 EU687815 EU687816 EU687817
EU688387 EU688388 EU688389 EU688390 EU688391 EU688392 EU688393 EU688394 EU688395
EU688776 EU688777 EU688778 EU688779 EU688780 EU688781 EU688782 EU688783 EU688784
EU688498
EU688202
EU688299
EU688592
EU688006
EU687818
EU688396
e
EU687912
EU688678 EU688679 EU688680
EU688499 EU688500 EU688501
EU688203 EU688204 EU688205
EU688300 EU688301 EU688302
EU688593 EU688594 EU688595
EU688007 EU688008 EU688009
EU687819 EU687820 EU687821
EU688397 EU688398 EU688399
EU688785 EU688786 EU688787 EU688788 EU688789
e
EU688109 EU688110 EU688111 EU688087
EU688681 EU688682 EU688683 EU688684
EU688502 EU688503 EU688504 EU688505
EU688206 EU688207 EU688208 EU688209
EU688303 EU688304 EU688305 EU688306
EU688596 EU688597 EU688598 EU688599
EU688010 EU688011 EU688012 EU688013
EU687822 EU687823 EU687824 EU687805
EU688400 EU688401 EU688402 EU688403
EU688790 EU688791 EU688792 EU688793 EU688794 EU688795
EU687913 EU687914 EU687915
e
EF117916 EF117917 EF117918 EF117914 EU688881 EF117919 EF117920
EU688112 EU688113 EU688094 EU688114 EU688115 EU688116 EU688117 EU688118
EU688685 EU688686 EU688687 EU688688 EU688689 EU688690 EU688691 EU688692
EU688506 EU688507 EU688508 EU688509 EU688510 EU688511 EU688512 EU688513
EU688210 EU688211 EU688212 EU688213 EU688214 EU688215 EU688216 EU688217
EU688307 EU688308 EU688309 EU688310 EU688311 EU688312 EU688313 EU688314
EU688600 EU688601 EU688581 EU688602 EU688603 EU688604 EU688605 EU688606
EU688014 EU688015 EU687996 EU688016 EU688017 EU688018 EU688019 EU688016
EU687825 EU687826 EU687827 EU687828 EU687829 EU687830 EU687831 EU687832
EU688404 EU688405 EU688406 EU688407 EU688408 EU688409 EU688410 EU688411
EU688796 EU688797 EU688798 EU688799 EU688800 EU688801 EU688802 EU688803 EU688804 EU688805
EU687916 EU687917 EU687918 EU687919
e
EF117921 EF117922
EU688119 EU688120
EU688693 EU688694
EU688514 EU688515
EU688218 EU688219
EU688315 EU688316
EU688607 EU688608
EU688020
EU687833
EU688412 EU688413
EU688806 EU688807 EU688808 EU688809 EU688810
EU687920 EU687921 EU687922 EU687923 EU687924 EU687925 EU687926 EU687927
e
EU687928 EU687929
196
Supplement material 1. Isolates of Phytophthora ramorum used in this study with the Genbank accession number. Genbank accession no. Isolate no. CBEL β83987 79133 81813 83000 83989 83046 82939 79072 a tubulin
79140
81804 EU688811 EU688812 EU688813 EU688814 EU688815 EU688816 EU688817 EU688818 EU688819 EU688820 EU688821 EU688822 EU688823 EU688824 EU688825 EU688828 EU688829 EU688830 EU688831 EU688832 EU688747 EU688748 EU688750 EU688751 EU688749 EU688752 EU688753 EU688754 EU688755 EU688756 EU688757 EU688758 EU688759 EU688760 EU688761 EU688762
CBS110545
EF117960
EF117923
EU688121
EU688695
EU688516
EU688220
EU688317
EU688609
EU688021
EU687834
EU688414
CBS110546 CBS110547 CBS110548 CBS110601 CBS110900 CBS110901 CBS110953 CBS110954 CBS110955 CBS110956 CBS110957 CBS111762 F001 F002 F004 F005 F006 2N386 2N389 3N5-1 3N91 3N208-3 4N200-1 4N203-6 4N372-2 4N461-2 4N501-3 4N657-2 4N662-1 4N717-4 4N773-3 4N780-1 4N793-3
EF117961 EU688932 EF117962 EU688933 EU688934 EU688935 EU688951 EU688949 EU688936 EU688910 EU688911 EU688912 EU688950 EU688916 EU688917 EU688918 EU688919 EU688922 EU688923 EU688948 EU688953 EU688954 EU688924 EU688925 EU688926 EU688955 EU688956 EU688952 EU688957 EU688903 EU688904 EU688905 EU688906
EF117927 EF117928 EF117929 EU688882 EU688883 EU688875 EU688884 EU688885 EU688886 EU688887 EU688888 EU688889 EU688892 EU688893 EU688894 EU688895 EU688896 EU688853 EU688854 EU688855 EU688856 EU688857 EU688858 EU688859 EU688860 EU688861 EU688862 EU688863 EU688864 EU688865 EU688866 EU688867
EU688122 EU688123 EU688124 EU688125 EU688126 EU688127 EU688128 EU688129 EU688130 EU688131 EU688132 EU688133 EU688136 EU688137 EU688138 EU688139 EU688140 EU688064 EU688065 EU688066 EU688067 EU688063 EU688068 EU688069 EU688070 EU688071 EU688072 EU688073 EU688074 EU688075 EU688076 EU688077 EU688078
EU688696 EU688697 EU688698 EU688699 EU688700 EU688701 EU688702 EU688721 EU688703 EU688704 EU688705 EU688706 EU688709 EU688710 EU688711 EU688712 EU688713 EU688640 EU688726 EU688641 EU688642 EU688643 EU688644 EU688645 EU688646 EU688647 EU688648 EU688649 EU688650 EU688651 EU688652 EU688653
EU688517 EU688518 EU688519 EU688520 EU688485 EU688484 EU688521 EU688522 EU688523 EU688524 EU688525 EU688526 EU688480 EU688529 EU688530 EU688531 EU688532 EU688460 EU688461 EU688463 EU688464 EU688462 EU688465 EU688466 EU688467 EU688468 EU688469 EU688470 EU688471 EU688472 EU688473 EU688474 EU688475
EU688221 EU688222 EU688223 EU688224 EU688225 EU688226 EU688227 EU688228 EU688229 EU688230 EU688231 EU688232 EU688183 EU688235 EU688236 EU688163 EU688164 EU688166 EU688167 EU688165 EU688168 EU688169 EU688170 EU688171 EU688172 EU688173 EU688174 EU688175 EU688176 EU688177 EU688178
EU688318 EU688319 EU688320 EU688321 EU688322 EU688323 EU688324 EU688325 EU688326 EU688327 EU688328 EU688329 EU688282 EU688332 EU688333 EU688334 EU688335 EU688262 EU688263 EU688265 EU688266 EU688264 EU688267 EU688268 EU688269 EU688270 EU688271 EU688272 EU688273 EU688274 EU688275 EU688276 EU688277
EU688610 EU688611 EU688602 EU688603 EU688604 EU688577 EU688612 EU688613 EU688614 EU688615 EU688616 EU688617 EU688612 EU688620 EU688621 EU688622 EU688623 EU688554 EU688555 EU688557 EU688558 EU688556 EU688559 EU688560 EU688561 EU688562 EU688563 EU688564 EU688565 EU688566 EU688567 EU688568 EU688569
EU688022 EU688023 EU688024 EU688025 EU688026 EU688027 EU688028 EU688029 EU688030 EU688031 EU688034 EU688035 EU688036 EU688037 EU688038 EU687969 EU687970 EU687972 EU687973 EU687971 EU687974 EU687975 EU687976 EU687977 EU687978 EU687979 EU687980 EU687981 EU687982 EU687983 EU687984
EU687835 EU687836 EU687837 EU687838 EU687839 EU687840 EU687841 EU687842 EU687843 EU687844 EU687845 EU687846 EU687849 EU687850 EU687851 EU687852 EU687853 EU687782 EU687783 EU687784 EU687785 EU687786 EU687787 EU687788 EU687789 EU687790 EU687791 EU687792 EU687793 EU687794 EU687795 EU687796
EU688415 EU688416 EU688417 EU688418 EU688419 EU688420 EU688421 EU688422 EU688423 EU688424 EU688425 EU688426 EU688429 EU688430 EU688431 EU688432 EU688433 EU688360 EU688361 EU688363 EU688364 EU688362 EU688365 EU688366 EU688367 EU688368 EU688369 EU688370 EU688371 EU688372 EU688373 EU688374 EU688375
79310 e
EU687930
EU687931 EU687932 EU687933 EU687934 EU687935 EU687936 EU687937 EU687938 EU687939 EU687940 EU687943 EU687944 EU687945 EU687946 EU687947 EU687877 EU687879 EU687880 EU687878 EU687881 EU687882 EU687883 EU687884 EU687885 EU687886 EU687887 EU687888 EU687889 EU687890 EU687891
197
Supplement material 1. Isolates of Phytophthora ramorum used in this study with the Genbank accession number. Genbank accession no. Isolate no. CBEL β83987 79133 81813 83000 83989 83046 82939 79072 a tubulin 4N992-1 POL1 POL2 A326 A1379
EU688907 EU688920 EU688921 EU688913 EU688914
EU688868 EU688898 EU688899 EU688871 EU688869
EU688079 EU688142 EU688143 EU688086 1EU688082 2EU688083
e
EU688655 EU688656
A1382
EU688915
EU688870
1EU688084 EU688085
e
05-119 05-053-1 05-054-2 05-078 05-079-3 05-422-9 06-080-1 06-046 06-043-3 06-027 06-114 06-037-1 06-038-3 06-024-1 06-318-1 06-351-1 06-268-2 BBA104 BBA22/015 BBA24/02 BBA4/02-4 BBAPR 05m BBAPR 84 BBAPR 88
EU688927 EU688928 EU688931 EU688929 EU688930 EU688940 -
EU688849 EU688840 EU688841 EU688842 EU688847 EU688846 EU688845 EU688843 EU688848 EU688844 EU688850 EU688872 EU688873
EU688050 EU688046 EU688047 EU688048 EU688049 EU688051 EU688058 EU688057 EU688056 EU688053 EU688059 EU688054 EU688055 EU688052 EU688061 EU688062 EU688060 EU688090 EU688091
EU688657 EU688658 EU688659 EU688635 EU688632 EU688633 EU688727 EU688639 EU688638 EU688636 EU688634 EU688637 EU688722 EU688723
EU688939 EU688941
EU688874 EU688876 EU688877
EU688092 EU688093 EU688088
EU688942 EU688943
EU688878 EU688879
EU688089 EU688095
EU688654 EU688715 EU688728 EU688720
79140
81804
79310
EU688763 EU688834 EU688835 EU688768
EU687892 EU687949 EU687950 EU687895 EU687893
e
EU688476 EU688534 EU688535 EU688479 EU688477
EU688179 EU688238 EU688239 EU688182 EU688180
EU688278 EU688337 EU688338 EU688281 EU688279
EU688570 EU688626 EU688627 EU688573 EU688571
EU687985 EU688040 EU688041 EU687989 EU687986
EU687797 EU687855 EU687856 EU687800 EU687798
EU688376 EU688435 EU688436 EU688379 EU688377
e
EU688478
EU688181
EU688280
EU688572
EU687987
EU687799
EU688378
EU688764 EU688765 EU688766 EU688767
EU688445 EU688441 EU688442 EU688443 EU688444 EU688446 EU688453 EU688452 EU688451 EU688448 EU688454 EU688449 EU688450 EU688447 EU688456 EU688457 EU688455 EU688481 EU688482
EU688151 EU688148 EU688149 EU688150 EU688152 EU688158 EU688157 EU688156 EU688153 EU688159 EU688154 EU688155 EU688160 EU688184 EU688185
EU688247 EU688243 EU688244 EU688245 EU688246 EU688248 EU688255 EU688254 EU688253 EU688250 EU688256 EU688251 EU688252 EU688249 EU688258 EU688259 EU688257 EU688283 EU688284
EU688542 EU688540 EU688541 EU688543 EU688549 EU688548 EU688547 EU688544 EU688550 EU688545 EU688546 EU688551 EU688574 EU688575
EU687956 EU687955 EU687957 EU687963 EU687962 EU687961 EU687964 EU687959 EU687960 EU687958 EU687966 EU687965 EU687990 EU687991
EU687767 EU687763 EU687764 EU687765 EU687766 EU687767 EU687775 EU687774 EU687773 EU687770 EU687776 EU687771 EU687772 EU687769 EU687778 EU687779 EU687777 EU687801 EU687802
EU688345 EU688343 EU688344 EU688346 EU688353 EU688352 EU688351 EU688348 EU688354 EU688349 EU688350 EU688347 EU688356 EU688357 EU688355 EU688380 EU688381
EU688733 EU688729 EU688730 EU688731 EU688732 EU688734 EU688740 EU688739 EU688736 EU688741 EU688737 EU688738 EU688735 EU688743 EU688744 EU688742 EU688769 EU688770
EU687865 EU687861 EU687862 EU687863 EU687864 EU687866 EU687871 EU687870 EU687867 EU687872 EU687868 EU687869 EU687874 EU687875 EU687873 EU687896 EU687897
EU688724 EU688725 EU688661
EU688483 EU688486 EU688488
EU688186 EU688189 EU688190
EU688285 EU688286 EU688287
EU688576 EU688578 EU688580
EU687992 EU687993 EU687995
EU687803 EU687804 EU687806
EU688382 EU688383 EU688384
EU688771 EU688772 EU688773
EU687898 EU687899 EU687900
EU688664 EU688665
EU688489 EU688490
EU688191 EU688192
EU688288 EU688289
EU688582 EU688583
EU687997 EU687998
EU687807 EU687808
EU688385 EU688386
EU688774 EU688775
EU687901 EU687902
e
EU687894
198
Supplement material 1. Isolates of Phytophthora ramorum used in this study with the Genbank accession number. Genbank accession no. Isolate no. CBEL β83987 79133 81813 83000 83989 83046 82939 79072 a tubulin Strain 3206 Strain 3211 1270626-1 1341211-4 CSL 1620 CSL 1715 MSOD105
EU688946 EU688947 EU688937 EU688938 EU688944 EU688945 EF117963
Pr03-74-1
-
Pr03-74-2
-
a
EU688901 EU688902 EU688851 EU688852 EU688890 EU688891 EU688897 EF117942 EU688900 EF117943 EF117944
79140
81804
79310
EU688146 EU688147 EU688080 EU688081 EU688134 EU688135 EU688141
EU688718 EU688719 EU688707 EU688708 EU688714
EU688538 EU688539 EU688458 EU688459 EU688527 EU688528 EU688533
EU688242 EU688161 EU688162 EU688233 EU688234 EU688237
EU688341 EU688342 EU688260 EU688261 EU688330 EU688331 EU688336
EU688630 EU688631 EU688552 EU688553 EU688618 EU688619 EU688625
EU688044 EU688045 EU687967 EU687968 EU688032 EU688033 EU688039
EU687859 EU687860 EU687780 EU687781 EU687847 EU687848 EU687854
EU688439 EU688440 EU688358 EU688359 EU688427 EU688428 EU688434
EU688838 EU688839 EU688745 EU688746 EU688826 EU688827 EU688833
EU687953 EU687954 EU687876 EU687941 EU687942 EU687948
EU688144
EU688716
EU688536
EU688240
EU688339
EU688628
EU688042
EU687857
EU688437
EU688836
EU687951
EU688145
EU688717
EU688537
EU688241
EU688340
EU688629
EU688043
EU687858
EU688438
EU688837
EU687952
More isolates details shown on Table 1. b Accession no. beginning by EF were sequenced and used for CBEL and β-tubulin in (Bilodeau et al. 2007) c Accession no. beginning by EU were sequenced and alignments were deposed for these study d “-” Not sequenced or not deposited in GenBank. e Multiple haploid sequences designing haplotypes obtained by clone isolation.
199
Discussion générale et conclusion
200
Discussion générale et conclusion
La première hypothèse de cette thèse se présentait comme étant que le P. ramorum présente des différences interspécifiques des autres espèces du genre Phytophthora. Les alignements de séquences des régions de l’ITS, de la β-tubuline et de l’élicitine pour plusieurs espèces de Phytophthora ont permit de confirmer cette hypothèse. Dans le premier chapitre de ce document, l’objectif était de développer des outils moléculaires pour identifier le P. ramorum et le différencier des autres Phytophthora. Même si des outils de détection étaient déjà disponibles, ceux-ci n’étaient pas satisfaisants étant donné qu’ils présentaient de fortes proportions de faux-positifs et de faux-négatifs. Notre hypothèse était qu’une meilleure mise au point et que des sondes ciblant plusieurs gènes résulteraient en des outils moléculaires plus sensibles et spécifiques. Un autre objectif consistait à comparer divers types de sondes rapporteures pour effectuer la détection en PCR en temps réel. La réalisation d’outils de diagnostic permettant de détecter rapidement et efficacement le P. ramorum a donc été effectuée avec succès en utilisant la PCR en temps réel et la sonde TaqMan comme molécule reportrice de l’agent pathogène. La PCR en temps réel est la technique possédant les meilleures caractéristiques pour la détection moléculaire d’agents pathogènes (Martin et al. 2000; McCartney et al. 2003; Schaad and Frederick 2002). Plus sensible que la PCR traditionnelle, elle permet de détecter et de quantifier la présence du microorganisme cible et cela en moins de trois heures (Dorak 2006). Ce chapitre apporte plusieurs informations nouvelles par rapport aux diverses techniques de détection en PCR en temps réel des organismes phytopathogènes comme les Phytophthora. Les différentes techniques essayées consistaient à l’emploi de trois technologies reportrices ; le SYBRGreen, le phare moléculaire (« molecular beacon ») et la sonde TaqMan. Parmi ces trois technologies, la sonde TaqMan démontrait des avantages pour la détection, la sensibilité, la double spécificité par les SNPs sur les amorces et la sonde, et par la facilité de conception par rapport aux deux autres techniques. Par exemple, pour la région de l’ITS où il y a seulement 1,4% de divergence entre le P. ramorum et le Phytophthora lateralis, une autre espèce très apparentée, des amorces avec une seule base polymorphe et trois
201 bases pour la sonde ont donc été conçues. Par conséquent, c’est cette méthode qui fut conservée pour la suite des travaux. D’autre part, l’utilisation de trois régions géniques du P. ramorum pour la création du test diagnostic a apporté de bonnes informations quant à l’avantage de la redondance dans l’utilisation de trois marqueurs au lieu d’un seul pour l’établissement d’un diagnostic. De plus, la limite de détection de ces méthodes utilisant la PCR en temps réel, permet de déterminer qu’elle est très efficace et comparable aux limites retrouvées par les méthodes de PCR nichées (« Nested PCR ») préalablement décrites (Garbelotto 2003). Dans ce chapitre, des échantillons environnementaux infectés et non infectés ont été testés pour la présence du P. ramorum avec le milieu de culture PARP-V8, nos trois sondes TaqMan et le TaqMan cytochrome oxydase (cox) (Tooley et al. 2006). Un total de 24% des échantillons se sont avérés être positifs pour les tests de PCR en temps réel mais pas avec les cultures en milieu pour le P. ramorum. Les séquences des régions des espaces de la cox I et II ont pu confirmer la présence du pathogène dans la plupart des échantillons. Les limites de détection sont d’ailleurs augmentées ici par l’emploi de la PCR en temps réel comparativement aux cultures qui parfois ne permettent pas de détecter de façon aussi sensibles l’agent pathogène en cause. La raison de ce manque de sensitivité des méthodes de culture provient de la contamination des milieux soit par des contaminants communs de laboratoire ou par des endophytes ou microorganismes opportunistes aggressifs qui supplantent la croissance du P. ramorum en culture. Dépendamment de la région utilisée, la limite de sensibilité peut varier. Les régions de l’ITS et le l’élicitine qui sont en copies multiples permettent une plus grande sensibilité en comparaison au gène de la β-tubuline qui n’en possède qu’une ou quelques-unes. Les limites de détection que nous avons obtenues avec nos sondes permettent de détecter des concentrations d’ADN qui correspondent au niveau de (10-100pg) généralement retrouvé dans la littérature (Tomlinson et al. 2005) dans des échantillons environnementaux.
Le protocole de ce chapitre fît l’objet de validation et est actuellement utilisé par l’ACIA pour ses enquêtes annuelles. Par exemple, en 2007, du 1er avril jusqu’au 14 décembre 2007 seulement, plus de 46000 échantillons ont été testés utilisant ce protocole, 15620 testés par PCR avec 3 à 8 puits pour chacun avec diverses dilutions. Au total, 164 échantillons ont été détectés infectés par le P. ramorum (communication personnelle
202 Stéphan Brière, ACIA). De plus, une importante validation internationale où plus de trois cents échantillons ont été envoyés à l’aveugle à plus de huit laboratoires testant ainsi 12 marqueurs différents. Plusieurs espèces de Phytophthora et de Pythium, un autre oomycète, ont été testées en plus d’échantillons environnementaux, une soixantaine, provenant de différents hôtes infectés par le P. ramorum ou d’autres espèces de Phytophthora. Parmi les différents laboratoires mis à l’essai, notre test fut l’un des plus performants où l’utilisation de trois sondes spécifiques au lieu d’une seule, nous a permis de diagnostiquer avec confiance les différents isolats testés. Cette étude fut d’ailleurs soumise tout dernièrement (Martin 2008) à la revue « Phytopathology » et fut l’objet d’affiche lors des congrès 2006 et 2007 (Martin et al. 2006; Zeller 2007) de la société américaine de phytopathologie (APS). De plus, une note d’application (Annexe 3) a également été publiée pour l’utilisation de cette méthode, utilisant les trois technologies, en lecture de fin de réaction (endpoint) afin de permettre l’utilisation de cet outil avec un lecteur de fluorescence sur plaque, tel le Fluoroskan Ascent (Thermo), quand un thermocycleur en temps réel n’est pas disponible.
Le chapitre 2 est également basé sur la première hypothèse qui concerne le polymorphisme interspécifique du P. ramorum et le même objectif de développer des outils moléculaires pour le diagnostic de l’agent pathogène. On a vu au premier chapitre, les avantages à l’utilisation de plusieurs sondes et plusieurs gènes pour la détection du P. ramorum. Ce chapitre-ci a permis de rendre ce test utilisable en une seule réaction PCR ce qui rend la méthode plus rapide et moins coûteuse. Deux agencements de réactions multiplexes ont été développés dans cet article. Premièrement, les trois sondes et les trois paires d’amorces du chapitre premier, ciblant les régions de l’ITS, de la β-tubuline et de l’élicitine, ont été mises en commun tout en changeant la chimie des molécules fluorescentes. En plus, une sonde et une paire d’amorces spécifiques au genre Phytophthora, dans la région de la β-tubuline, ont été ajoutées. Cette réaction en multiplexe permet donc dans une seule réaction de décerner la présence du P. ramorum et de détecter si d’autres Phytophthora sont présents dans l’échantillon testé. La seconde réaction multiplexe ciblait le P. ramorum, le Phytophthora, les oomycètes et la plante. Ces réactions ont donc été vérifiées avec fiabilité dans les différentes espèces de Phytophthora ainsi que
203 par les ADN d’échantillons environnementaux d’hôtes infectés et d’extrait de lysat d’ELISA. Ces réactions de PCR en temps réel, en multiplexe, ont démontré une sensibilité légèrement inférieure aux réactions en simplex. Cette réduction est probablement causée par l’agencement de quatre fluorochromes et absorbeurs (quenchers) qui se font compétition dans le mélange réactif et peut réduire la quantité de lumière émise par les sondes. De plus, les amorces étant conçues initialement pour des réactions en simplex sont maintenant mises en commun avec le multiplexe et peuvent interagir l’une avec les autres, comme probablement le cas pour les amorces de la β-tubuline et de l’élicitine et ainsi réduire l’efficacité de l’amplification. Le gène de la β-tubuline semble présenter quelques réactions croisées permettant la présence de quelques faux positifs. Par contre, on retrouve ici l’avantage de joindre plusieurs sondes dans la même réaction et permet alors de diagnostiquer les échantillons présentant le P. ramorum. Cela réduit énormément les coûts du test, en reduisant la concentration des sondes, les éléments jetables (embouts et tubes de plastique), la polymérase et le temps d’utilisation. Cette plus grande efficacité permet de performer des tests à plus haut débit, ce qui est important dans un contexte de vastes inventaires comme ceux initiés par l’ACIA. La sonde ciblant le P. ramorum la plus efficace étant la sonde ITS a aussi été utilisée avec d’autres sondes présentant un niveau hiérarchique permettant un témoin et de déterminer la présence d’autres organismes pouvant être apparentés et recherchés dans la même réaction. De plus, il s’agit d’un excellent témoin de l’extraction de l’ADN qui dans le cas d’échantillons environnementaux contient bien souvent plus d’ADN végétal que d’ADN cible. L’utilisation de la RuBisCO en copies multiples permet donc une bonne détection de ce matériel. Ces marqueurs nouvellement développés pourraient permettre également de développer des sondes de type TaqMan spécifiques à d’autres Phytophthora ou oomycètes problématiques en conjonction avec ces trois nouvelles sondes. Donc en concevant des sondes Taqman ciblant un Phytophthora ou un Pythium et en combinant cette sonde avec les sondes Phytophthora, Oomycètes, ou de plantes, selon le besoin. Ce nouvel outil pourrait donc s’avérer également polyvalent et utile pour les agences de réglementations comme l’ACIA qui doivent tester des dizaines de milliers d’échantillons
204 par années et ainsi réduire les coûts et le temps pour diagnostiquer la maladie sur les plants récoltés dans les différentes pépinières du pays.
Les deux chapitres suivants se reportaient quant à eux à l’hypothèse qu’il existe du polymorphisme intra spécifique à l’intérieur même de l’espèce P. ramorum et que ces polymorphismes peuvent être utilisés pour répondre à des questions d’épidémiologie moléculaire, par exemple pour déterminer l’origine des infestations. Ils ont donc comme objectif premier, de découvrir des loci différenciant le polymorphisme intraspécifique du P. ramorum. Comme deuxième objectif, il s’agissait de réaliser des études de populations sur les populations européennes et nord-américaines connues et finalement de développer des outils permettant de détecter ces polymorphismes.
Au troisième chapitre, deux régions géniques ont été ciblées utilisant les différentes séquences disponibles sur GenBank à cette époque afin d’amplifier des séquences du P. ramorum. Les séquences du gène de la β-tubuline et de la CBEL (cellulose binding elicitor lectin) ont démontré du polymorphisme intraspécifique au P. ramorum entre les populations nord-américaines et européennes. Par la suite, des outils appelés ASO (oligonucléotide spécifique d’allèle) ont permis de génotyper les différents échantillons avec deux SNPs pour chaque gène créant ainsi une empreinte génétique de chaque souche. Des amorces spécifiques pour chaque SNP ont été conçues afin d’amplifier les allèles alternatifs à chaque locus et de générer les génotypes à l’aide de la PCR en temps réel avec le SYBRGreen. Cet outil permet donc de séparer et identifier les différents échantillons testés par leur origine géographique. Plusieurs individus européens et nord-américains ont ainsi été génotypés et séquencés pour confirmation. Fait intéressant, les échantillons Canadiens et de l’Oregon retrouvés en pépinières en 2003, présentaient des profils très différents des profils nord-américains, mais identifiques à certains profils Européens. Ces résultats indiquent que l’origine de ces souches est fort probablement européenne et non pas californienne comme la chronologie de l’infestation le suggère.
205 Étant donné que les souches d’Amérique du Nord sont du type sexuel A2 alors que ceux d’Europe sont de type A1, on a d’abord pensé que ces marqueurs étaient associés au type sexuel. Cependant, nous avons démontré qu’ils étaient plutôt associés à l’origine géographique et non au type d’appariement. En efet, une souche retrouvée en Belgique et présentant un type sexuel A2 permet de confirmer que cette souche possède bel et bien un profil génétique européen, mais un type sexuel A2. Donc, nos marqueurs sont associés à l’origine géographique et non aux types d’appariement sexuel. Cette méthode simple et peu couteuse permet donc génotyper rapidement des échantillons ayant la présence du P. ramorum.
L’Agence canadienne d’inspection des aliments utilise d’ailleurs cet outil de génotypage pour tester les différents isolats récoltés au cours de leur enquête annuelle. De plus, des échantillons extraient de lysat d’ELISA de l’enquête annuelle de 2005, ont pu être génotypés suite à la confirmation de la présence du P. ramorum dans les échantillons. Cette étude présente une plus grande proportion d’isolats infectés par le type nord-américain retrouvé que du type européen. Les échantillons environnementaux peuvent donc être génotypés directement à partir de ces lysats sans effectuer de cultures ou sans refaire d’extraction des échantillons. Ceci est important étant donné que l’utilisation de tissus différents pour l’ELISA et la PCR peut résulter en des résultats non-concordants en raison de la distribution inégale du microorganisme. Par contre, cette approche peut être limitée selon la concentration d’ADN de l’agent pathogène et peut parfois réagir avec d’autres espèces de Phytophthora. Elle représente néanmoins une méthode intéressante pour connaître rapidement le génotype et l’origine potentielle de l’organisme.
L’isolat US4 (NA2) présentait un profil différent avec les microsatellites (Ivors 2006) mais n’avait pas été testée avec nos ASO au moment de la publication du chapitre 3 (Bilodeau et al. 2007). Après utilisation de cet outil sur cet échantillon, il a été démontré qu’il présentait un nouveau profil pour le gène CBEL. Au lieu de présenter un profil de doubles homozygotes comme pour les autres nord-américains ou de doubles hétérozygotes pour les deux SNPs des isolats européens, il présentait un profil homozygote pour le premier SNP et hétérozygote au deuxième SNP ce qui permet de le différencier des deux
206 autres génotypes retrouvés. Ce polymorphisme sera d’ailleurs discuté plus rigoureusement dans le quatrième chapitre.
Le dernier chapitre, qui vise également à effectuer un cribblage plus large du génome de P. ramorum pour comparer les polymorphismes chez les populations a quant à lui bénéficié de la sortie du génome complet de l’organisme en juin 2004. La disponibilité de ce génome a alors permis d’avoir accès à près de 16000 gènes pour la découverte de nouveaux polymorphismes intraspécifiques. La notion de volatilité des codons décrite par Plotkin et al., 2004 (Plotkin et al. 2004) a permis d’utiliser ce concept et de l’appliquer pour la sélection des gènes à analyser et à utiliser pour la découverte des nouveaux SNPs. Nous avons effectivement retrouvé un nombre de polymorphismes (SNPs) dans le compartiment des loci possédant une plus haute volatilité comparé à ceux possédant une plus faible volatilité. Toutefois, les mesures de diversité nucléotidique (π) et l’hétérozygosité n’étaient pas différentes dans ces deux compartiments, ce qui nous laisse croire que la volatilité peut être utile pour la découverte de SNP, mais pas nécessairement pour la découverte de loci polymorphes dans le génome de P. ramorum.
Parmi les gènes où des polymorphismes ont été retrouvés (13), plusieurs semblaient présenter des polymorphismes non synonymes (37), qui sont impliqués dans la séquence protéique et qui peuvent générer un caractère évolutif dans l’adaptation. Parmi ces mutations non synonymes, on retrouve des gènes possiblement retrouvés dans des protéines de la paroi cellulaire (comme la Cellulobiohydrolase; la Glycosylphosphatidylinositol (GPI); la fibronectine; la pinin et la protéine de transport de type ABC) souvent impliquées dans la relation hôte-agent pathogène dans certains organismes. De plus, la combinaison de séquences multilocus permet d’obtenir un profil génétique permettant non seulement de différencier entre les souches européennes (EU1) et nord-américaines (NA1-NA2) mais également de retrouver d’autres génotypes liés à la diversité de certaines populations. Nos analyses des génotypes multilocus montrent que le génotype NA2 semble appartenir à un clade distinct du NA1 ou d’EU. Les populations du P. ramorum sont très différentes en Amérique du Nord et en Europe. Le nombre de génotypes multilocus en Europe est très élevé, mais ces génotypes sont génétiquement très rapprochés et la plupart sont rares. La
207 situation en Amérique du Nord est à l’opposé, avec seulement 3 génotypes multilocus, mais une forte divergence entre ces génotypes entre eux, et avec les génotypes européens. L’analyse révèle donc que la diversité nucléotidique est assez semblable dans les deux populations. Cependant, ces génotypes présentent moins de différences allèliques que dans la population nord-américaine. La population NA présente que trois génotypes, mais ces génotypes représentent deux clades différents et bien disctincts, représentant donc une différence allèlique plus élevée.
Ces résultats appuient donc fortement l’idée d’introductions indépendantes du P. ramorum en Europe et en Amérique du Nord, plutôt qu’une introduction commune ou continue. Par contre, un certain flux génique semble avoir eu lieu entre l’Europe et l’Amérique du Nord, puisque les génotypes multilocus les plus fréquents en Europe ont été retrouvés en Amérique du Nord. Mais ce flux génique semble être à sens unique : les trois profils génétiques Nord-Américains n’ont toujours pas été retrouvés en Europe. Une surveillance serrée serait à conseiller pour suivre cette épidémie au niveau moléculaire et identifier les migrations potentielles du pathogène. Les outils que nous avons développés aideront grandement à ces études. Nous avons en effet produit plus de 1000 séquences d’ADN sur 13 gènes, qui penvent être développés en outils de diagnostic des Phytophthora.
Ce projet de doctorat, je crois, apporte de nouveaux outils, plusieurs réponses pour la détection moléculaire des Phytophthora mais également en phytopathologie. Plusieurs techniques, sondes et nouvelles données de séquences pour le P. ramorum ont été apportées. De plus, de nouveaux génotypes ont été identifiés et d’autres ont confirmé les études précédentes sur différentes populations d’Europe et d’Amérique du Nord. De bonnes collaborations ont été développées comme avec l’ACIA, le USDA, le CBS et plusieurs laboratoires de diagnostics des phytopathogènes en Europe. Comme mentionné plus haut, plusieurs des outils développés dans cette thèse ont trouvé des applications opérationnelles à l’ACIA et potentiellement aux États-Unis (comme nos amorces et notre sonde élicitine, nouvellement ajouté au programme de diagnostic du USDA, depuis mai 2008). La production du génome du P. ramorum pendant ce doctorat a représenté un outil fantastique pour la découverte de polymorphismes. Ces SNPs apportent des renseignements et de
208 nouvelles voies pour le développement d’outils de diagnostic ciblant le P. ramorum. D’autres études non mentionnées dans cette thèse ont également été réalisées et emploient également la volatilité comme principe de sélection. Dans les perspectives futures de ce projet, on peut retrouver entre autres des analyses comparatives des génomes de Phytophthora disponibles permettant ainsi de réaliser des outils pour analyses de généalogies de gènes avec les différents marqueurs développés. Dans cette ère de génomique, pour une première fois le génome de plusieurs Phytophthora et aussi de Pythium est disponible. Cela génère un apport important pour le développement d’outils moléculaires pour la détection, l’étude et la compréhension de cet important groupe de phytopathogènes. L’utilisation de la bio-informatique, dont certains outils développés dans cette thèse, permet donc de poursuivre les analyses de ces génomes et d’identifier de nouveaux marqueurs et génotypes, de vérifier le potentiel d’implication de ces génotypes dans la pathogénicité ou la virulence et d’avoir des outils de diagnostic non seulement pour l’identification taxonomique, mais aussi pour l’évaluation du risque épidémiologique.
209
Références (Discussion générale et conclusion) Bilodeau, G.J., C.A. Lévesque, A.W.A.M. de Cock, S.C. Brière, and R.C. Hamelin (2007) Differentiation of European and North American genotypes of Phytophthora ramorum by real-time polymerase chain reaction primer extension. . Can. J. Plant Pathol., 29(4): 408-420. Dorak, M.T., (2006) Real-time PCR. Taylor & Francis, Oxford. Garbelotto, M. (2003) Molecular diagnostics of Phytophthora ramorum, causal agent of sudden oak death. In Sudden oak death: how concerned should you be? An international symposium, 21 April-12 May, 2003. American Phytopathological Society Online symposium, Online Symposium. Ivors, K., M. Garbelotto, I.D.E. Vries, C. Ruyter-Spira, B. T. Hekkert, N. Rosenzweig and P. Bonants (2006) Microsatellite markers identify three lineages of Phytophthora ramorum in US nurseries, yet single lineages in US forest and European nursery populations. Mol. Ecol., 15(6): 1493. Martin, F.N., M. Coffey, P. Berger, R. Hamelin, P. Tooley, M. Garbelotto, K. Hughes, and T. Kubisiak (2006) Validation of molecular markers for Phytophthora ramorum detection and identification using a standardized library of isolates. Phytopathology, 96: S74. Martin, F.N., M.D. Coffey, K. Zeller, R.C. Hamelin, P. Tooley, M. Garbelotto, K.J.D. Hughes, T. Kubisiak, G.J. Bilodeau, L. Levy, C. Blomquist, and P. Berger (2008) Evaluation of molecular markers for Phytophthora ramorum detection and identification; testing for specificity using a standardized library of isolates. (Unpublished). Martin, R., D. James, and C. Levesque (2000) Impacts of molecular diagnostic technologies on plant disease management. Ann. Rev. Phytopathol., 38: 207-239. McCartney, H., S. Foster, B. Fraaije, and E. Ward (2003). Molecular diagnostics for fungal plant pathogens. Pest Manag. Sci., 59(2): 129-142. Plotkin, J.B., J. Dushoff, and H.B. Fraser (2004) Detecting selection using a single genome sequence of M. tuberculosis and P. falciparum. Nature, 428(6986): 942-945. Schaad, N., and R. Frederick (2002) Real-time PCR and its application for rapid plant disease diagnostics. Can. J. Plant Pathol., 24(3): 250-258.
210 Tomlinson, J.A., N. Boonham, K.J.D. Hughes, R.L. Griffin, and I. Barker (2005) On-Site DNA Extraction and Real-Time PCR for Detection of Phytophthora ramorum in the Field. Appl. Environ. Microbiol., 71(11): 6702-6710. Tooley, P.W., F.N. Martin, M.M.Carras, and R.D. Frederick (2006) Real-time fluorescent polymerase chain reaction detection of Phytophthora ramorum and Phytophthora pseudosyringae using mitochondrial gene regions. Phytopathology, 96(4): 336. Zeller, K.A., R. M. DeVries, and L.Levy (2007) Validation of confirmatory real-time PCR diagnostic assays for detecting Phytophthora ramorum. Phytopathology, 97: S129.
211
Annexes
212
Annexe 1
Tableaux des hôtes naturels confirmés et possibles du Phytophthora ramorum. Tableau annexe 1a Tiré du tableau 1. « Hôtes naturels confirmés du Phytophthora ramorum», Pages 8 à 16 de l’évaluation des risques phytosanitaires, Phytophthora ramorum agent de l’encre des chênes rouges. 1er mai 2006, (Rioux et al.2006) 6.
6
Rioux D, B. Callan, and D. McKenney (2006) Évaluation des risques phytosanitaires, Phytophthora ramorum agent de l’encre des chênes rouges. In: ERP no 00-39, p. 147. CFIA & CFS.
213
214
215
216
217
218
219
220
221
Tableau annexe 1b Tiré du tableau 2. « Hôtes possibles du Phytophthora ramorum », Pages 16 à 36 de l’évaluation des risques phytosanitaires, Phytophthora ramorum agent de l’encre des chênes rouges. 1er mai 2006, (Rioux 2006).
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
Annexe 2 Cycle de vie probable de Phytophthora ramorum Sporange
Cycle de vie probable de Phytophthora Germination ?
Sporange
Germination directe Oospore Caryogamie Oogone
Zoospores
Germination de la chlamydospore Sporange Germination directe, tube germinatif
Anthéridie Sporangiophore
Zoospores infectant les feuilles et l’écorce
Méiose Oogone Chlamydospore dans le tissue foliaire
Mycélium dans les tissues
Anthéridie Reproduction asexuée
Feuille infectée
Reproduction sexuée (Hétérothallique)
Arbre infecté
Figure 1 (Annexe II). Cycle de vie probable de Phytophthora ramorum Tiré et adapté de RAPRA, http://rapra.csl.gov.uk/background/lifecycle.cfm également adapté de (Agrios 1988) 7
7
Agrios G.-N., (1988) Plant Pathology. 3rd ed. Academic Press, Inc., San Diego.
242
Annexe 3
Note d’application du test diagnostic en lecture de fin de réaction (endpoint) :
Bilodeau, Guillaume, Lévesque, André and Richard Hamelin (2003) Detection of Microorganisms with the Fluoroskan Ascent® using Molecular Beacons, Taqman®, and SYBR® Green assays. Thermo-Electron Corporation - Microplate Instrumentation, SOLUTION NOTE. http://www.thermo.com/eThermo/CMA/PDFs/Articles/articlesFile_21017.pdf
243
244
