conference review report on the forest trees workshop at...

Comparative and Functional GenomicsComp Funct Genom 2003; 4: 229–238.Published online 1 April 2003 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/cfg.262

Conference Review

Report on the Forest Trees Workshop atthe Plant and Animal Genome Conference

Christophe Plomion1*, Janice Cooke2, Tom Richardson 3, John Mackay2 and Gerald Tuskan4

1UMR BioGeco, INRA 69 Route d’Arcachon, 33612 Cestas Cedex, France2Centre de Recherche en Biologie Forestiere, Universite Laval, Quebec, Quebec, Canada G1K 7P43New Zealand Forest Research, Private Bag 3020, Rotorua, New Zealand4Oak Ridge National Laboratory, Environmental Sciences Division, Oak Ridge, TN 37831-6422, USA

*Correspondence to:Christophe Plomion, INRA, 69Route d’Arcachon, 33612,Cestas Cedex, France.E-mail: [email protected]

Received: 25 January 2003Accepted: 3 February 2003

Introduction

Each week, progress is made in sequencing thegenomes of species ranging from bacteria tohumans. The genetic sequences of plant speciessuch as Arabidopsis and rice have already beencompleted, and the first tree genome should besequenced by 2003 (http://www.ornl.gov/ipgc/).Analysing this new information provides an unpr-ecedented opportunity to understand the regula-tion of genes, the function of gene products and,eventually, how organisms are assembled. Justas genomics tools are acknowledged as a cru-cial component of strategies aimed at improv-ing human health and the quality of life, webelieve that genomic tools will be critical to ensurethe long-term sustainability of forest health andproductivity.

Since January 1993 — the date of the first PlantGenome Conference — the Forest Trees GenomicsCommunity has held an annual workshop in SanDiego (CA, USA). The Workshop has alwaysprovided an excellent forum for reviewing thecurrent state of knowledge and discussing newdirections of research. It has also provided oppor-tunities for establishing contacts between scien-tists with different backgrounds (physiologists,

molecular biologists, geneticists, bioinformaticians,biometricians), and new collaborations betweenparticipants. Since January 2002, the Workshophas also become the annual meeting of the ‘For-est Genomics’ working party of the Interna-tional Union of Forestry Research Organizations(http://iufro.boku.ac.at/iufro/iufronet/d2/hp20410.htm). In 2003, this forum for presentation anddiscussion was divided into four sessions. The firstsession was devoted to genetic mapping and QTLdetection. The second session focused on a quitenew research field in the forest genomics com-munity: SNP discovery and linkage disequilibriummapping. Session three was devoted to functionalgenomics in conifer (mainly pines) and broadleaves(eucalypts and poplars) and session four was con-cerned by the International Populus Genome Con-sortium (IPGC). We will briefly summarize theworkshop contributions.

Session report

Genetic mapping and QTL detection

This session emphasized the use of genetic linkagemaps as essential tools to (a) provide information

Copyright 2003 John Wiley & Sons, Ltd.

http://www.ornl.gov/ipgc/

http://iufro.boku.ac.at/iufro/iufronet/d2/hp20410.htm

http://iufro.boku.ac.at/iufro/iufronet/d2/hp20410.htm

230 C. Plomion et al.

regarding the structure (I. Scotti) and evolution(E. Hayahi, K. Krutovskii) of conifer genomes,and (b) allow the detection of the loci (QTLs) thatcontrol quantitative variation, not only of agro-nomic traits (e.g. biomass, S. Hanley; wood qual-ity, D. Pot; dormancy-related, K. Jermstad) butalso of molecular traits (e.g. transcript accumula-tion, M. Kirst).

Genome structure

As a step towards the understanding of the structureof the genome of conifers, I. Scotti (Udine Univer-sity, Italy) presented an ‘Analysis of the Distribu-tion of Marker Classes in a Genetic Map of NorwaySpruce’. He described a large-scale ‘geography’ ofthe genome of Picea abies through the applicationof statistical methods borrowed from geographicalsciences (autocorrelation) that allowed the statis-tical testing of the clustering of molecular mark-ers (AFLPs, ESTPs, S-SAPs, inter-LTRs, SSRs) ofthe same or different types (i.e. expressed or non-expressed regions, repetitive or low copy regions).This approach, applied for the first time to the anal-ysis of the distribution of markers in a linkagemap, produced an overview of the architecture ofthe genome of this species, suggesting that there isspatial autocorrelation among markers, i.e. similarmarkers appear to be clustered across the genome.

Comparative mapping

Comparative mapping relies on the developmentand mapping of orthologous loci in related species,and comparing their location along linkage groups.The transfer of genetic information (e.g. mark-ers, candidate genes, QTLs) between phylogenet-ically related species, and the study of genomeevolution, are the most promising perspectives indeveloping comparative mapping studies. Geneticlinkage maps of forest tree species have beenwidely developed in the last decade (reviewed byCervera et al., 2000), providing the framework forcomparative mapping studies to expand over thepast 3 years. The Conifer Comparative GenomicsProject (CCGP, http://dendrome.ucdavis.edu/Sy-nteny/) was the first project formed as an interna-tional collaboration to develop orthologous geneticmarkers. Expressed sequence tags (ESTs) werechosen for this study, given the low rate of transfer-ability of microsatellite markers in conifers. ESTs

cannot only be studied across different species,but they also represent functional genes. Theycan be developed relatively easily from exist-ing cDNA/EST libraries and large gene discoveryprojects.

While the first comparative mapping studies onlyfocused on Pinus species, K. Krutovskii (USDADavis, USA) ‘Comparative Mapping in ConifersUsing EST Markers’ demonstrated how EST poly-morphisms (ESTPs) have helped to line up geneticmaps between conifer genera (Pinus, Picea, Pseu-dotsuga). Overall, gene content (synteny) and order(co-linearity) were maintained, which agrees withother pairwise genome comparisons within thegenus Pinus (reviewed in Chagne et al., 2003)and supports the cytogenetic evidence that chro-mosome evolution in the family Pinaceae is con-servative. He also reported on the progress ofEST sequencing projects for Norway spruce (Piceaabies) and Douglas-fir (Pseudotsuga menziesii ), forwhich 4000 ESTs have been produced and com-pared to loblolly pine (Pinus taeda) ESTs. Homolo-gies between the three EST sets were high.

As part of a breeding program aimed at devel-oping elite trees resistant to pine wood nematodein Pinus thunbergii, E. Hayahi (Forest Tree Breed-ing Center, Japan) presented the ‘Development ofCleaved Amplified Polymorphic Sequence Mark-ers in Japanese Black Pine’. They used ESTPs tointegrate multiple maps of Japanese black pine con-structed with AFLP and RAPD markers. Out of117 EST primer pairs reported in loblolly pine,72 pairs amplified a single fragment for Japaneseblack pine. Single nucleotide polymorphisms werefound in 15 EST fragments by sequence analysis.Restriction enzymes were then selected to detectpolymorphism. Finally, these markers were placedon the linkage map of Japanese black pine.

QTL detection

Most characters important to forestry, such asbiomass production, wood quality, biotic and abi-otic stress response are complex quantitative traits,resulting from a number of different genes inter-acting with each other and with the environment.These genes act together to provide a quantitativedifference, and are referred to as quantitative traitloci, or QTLs. The availability of genetic maps inmany forest tree species has made it possible to dis-sect quantitative traits into their Mendelian inher-ited components (the QTLs) and to improve our

Copyright 2003 John Wiley & Sons, Ltd. Comp Funct Genom 2003; 4: 229–238.

http://dendrome.ucdavis.edu/Synteny/

http://dendrome.ucdavis.edu/Synteny/

Forest Trees Workshop 231

basic understanding regarding the genetic architec-ture of economically and ecologically importanttraits. Sewel and Neale (2000) recently reviewedthe science of QTL mapping in forestry. As alreadydescribed for other organisms (Mauricio, 2001), itappears from their meta-analysis that, while thereare many genes underlying a quantitative traitin segregating populations, only a few of themaccount for a major proportion of the phenotypicvariance, providing opportunities to increase breed-ing efficiency through their direct manipulation. Intheir review, Sewel and Neale (2000) concludedwith the need for QTL verification and validationbefore a commitment to marker-aided selection canbe made in tree breeding.

While QTL mapping experiments have oftenconcerned a single pedigree in a given species,the alignment of genetic maps across species canbe considered as a starting point for QTL val-idation. In ‘Wood Quality Breeding in MaritimePine: From Wood Properties Dissection to Marker-aided Selection’, D. Pot (INRA, France) showedthe usefulness of comparative mapping for verify-ing quantitative trait loci (QTLs) between species.The alignment of genetic maps between Pinustaeda and Pinus pinaster served to validate wooddensity and cell wall chemistry QTLs and toco-localize positional candidate genes controllingthese traits. Coincidence of map positions sup-ports the hypothesis that loci underlying naturalquantitative variation have been conserved dur-ing long periods of evolutionary divergence. Threepresentations showed how the accuracy of map-ping QTLs can be achieved by increasing samplesize (S. Hanley), analysing QTLs in multiple envi-ronments (K. Jermstad) or different genetic back-grounds (D. Pot).

S. Hanley (University of Bristol, UK), in ‘Gene-tic Mapping of Important Agronomic Traits inBiomass Willows’, described a program of researchto provide markers linked to key agronomic traitsrelated to biomass production for marker-assistedselections in European Willow (Salix viminalis).To improve the ability to locate QTL, two large-scale mapping populations of 950 full-sib indi-viduals each were established, one being plantedin two sites. Drawing information from the exist-ing willow map, they genotyped the two progenysets using 100 microsatellite markers. Preliminarytrait mapping identified genomic regions putativelyinvolved in all initial target traits (stem height and

diameter, fresh and dry weights, rust resistance,beetle damage).

K. Jermstad (Institute of Forest Genetics, USA),in ‘Mapping of Quantitative Trait Loci Con-trolling Adaptive Traits in Coastal Douglas Fir.III QTL by Environment Interactions and Veri-fication’, described an experiment revealing theunderlying genetic factors influencing the adaptivecapacity of Douglas fir. Using clonal replicates (20ramets from 400 progeny) from a three-generationoutbred pedigree, she estimated QTLs respondingto environmental signals (winter chilling, springtemperature, moisture availability, photoperiod) ordormancy-related traits (bud set date, bud flushdate, growth rhythm). Thirty-four QTLs weredetected among all traits evaluated. The propor-tion of phenotypic variation explained by any indi-vidual QTL was 0.7–9.5%. Co-location of QTLsamongst growth cessation and rhythm traits wasfrequently observed. Few QTL × treatment inter-actions were found for growth initiation, whereasseveral were detected among the traits measuredfor growth cessation, many of which were repeat-edly detected among traits. The repeated detec-tion of QTLs for terminal bud flush was eval-uated within and between mapping populations.QTLs for bud flush were estimated at two testsites and in one greenhouse trial, and comparedwith QTLs for bud flush that were previously iden-tified in full-sib progeny from an earlier matingof the same parents. Repeated detection of theseQTLs was observed on six linkage groups provid-ing a strong case for the validation of these QTLs.In this pedigree, K. Krutovskii also reported onthe co-location between candidate genes and thedormancy-related QTLs.

D. Pot presented a three-step strategy to pro-vide the basic background to initiate a breed-ing program for wood and end-use properties inmaritime pine. In a first step, classical quantita-tive genetics provided estimates of variability andheritability for a wide range of wood quality-related traits. In a second step, these traits weredissected into their QTLs, resulting in the iden-tification of several genomic regions harbouringloci involved in trait variation. Comparative map-ping with loblolly pine allowed for comparisonof QTL locations between the two pine species.The position of two QTLs controlling wood den-sity and cell wall components were found tobe conserved between the two species. He also



showed how sequenced cDNAs from a tissue-specific library (i.e. tissues that express the desiredtrait, in this case wood forming tissue) can beidentified and mapped. Potential positional can-didate genes can then be identified in conjunc-tion with QTL mapping. These approaches, how-ever, did not reveal the actual genes underly-ing the QTLs. In a third step, a candidate geneapproach was taken, starting with the identificationof nucleotide polymorphisms in 20 genes thoughtto be important in determining wood properties.An association study in a diallel was conductedto evaluate the potential use of these markers in awider genetic background to provide reliable earlyselection criteria for wood properties (see nextsession).

QTL detection of molecular traits (TQL andPQL for quantitative accumulation of transcriptand protein, respectively) and the comparison withQTLs for agronomic traits may also be a powerfultool for identification of candidate genes (reviewedin Shields and O’Halloran, 2002). M. Kirst (NCState University, USA) spoke about ‘GeneticalGenomics of Eucalyptus: Combining ExpressionProfiling and Genetic Segregation Analysis’. Hedescribed an experiment that uses the genetic vari-ation between related individuals in a segregatingpopulation and adds the analytical tools availablefor molecular markers to the analysis of genome-wide expression profile data. He obtained the tran-script level profiles of 2658 genes for 91 full-sibsin a highly replicated micro-array experiment, andreported on the co-location between a major QTLfor wood density and a TQL (transcript quantitylocus) for one gene whose product is 89% iden-tical to an Arabidopsis protein involved in regu-lation of cellular osmotic pressure. This gene wasfurther mapped within the TQL peak region, indi-cating that it was probably involved in the controlof its expression. Variation at this locus accountsfor 25% of the variation in wood density in thisfamily.

U. Reyes-Valdes (University A Narro, Mexico)spoke about ‘Founder-origin informativeness inoutbred pedigrees’. He used the Shannon entropyto measure informativeness of genetic maps andmapping populations for QTL analysis. Usinga simulation, he showed that this method wascomputationally faster than the classical leastsquares approach and produced the same conclu-sions.

SNP discovery, linkage disequilibrium andassociation mapping

Since its inception more than a decade ago, foresttree molecular genetic research has traversed thetwin tracks of (a) genetic linkage analysis drivenby DNA-based polymorphisms and (b) the huntfor major effect candidate genes controlling targettraits, without much opportunity to pull the trackstogether. Applications arising from the first trackhave, until recently, relied on detecting neutralmarker-trait associations and implementing MAS(marker-assisted selection) to enhance genetic gain.In forest trees these marker–trait associations areusually pedigree-specific, owing to species-level,or pedigree-level, linkage equilibrium between theneutral marker and the causative genetic variant(s)underpinning trait variation. Such pedigree speci-ficity is a major disadvantage in most forest treeimprovement programs because, in practice, manyfamilies are deployed in production forests (e.g.due to multiplication bottlenecks and plantation-wide genetic diversity targets). The second track,leveraging major effect genes to alter traits of inter-est, has also not led to paradigm shifts in geneticgain because few genes with major effects on com-mercially important, quantitatively inherited traitshave been discovered. This is true despite exper-imental populations and approaches capable ofdoing so. Moreover, where major effect genes havebeen discovered, there is not always a straightfor-ward response from gene modification where, forexample, pathway redundancy frequently obviatesthe alteration of one gene in biosynthetic pathways.However, as is often the case, there is ‘innovationat the interface’! Recent technological advancesin candidate gene discovery via transcript profil-ing, and latterly SNP discovery, have created theopportunity to bring these two tracks together byidentifying and subsequently using polymorphismsthat are in strong linkage disequilibrium (LD) withtraits (Flint and Mott, 2001). This is the increas-ingly familiar ‘association mapping’ common inhuman genetics. If validated, these approaches holdtremendous potential in forest tree selection pro-grams. Last year at this workshop we heard thepreliminary plans to begin SNP validation experi-ments and this year three leading groups presentedtheir results to date.

In ‘Gene-assisted Selection (GAS) — A NewParadigm for Selection in Forest Tree Species?′



P. Wilcox (Forest Research, New Zealand) repor-ted on his team’s efforts to validate SNP–woodcharacteristic linkage disequilibrium (LD) in Pinusradiata. They used CAGE transcript profiling toidentify more than 200 putatively differential ESTsand real-time PCR to validate the differential tran-script levels. To date, 48/60 ESTs thus tested havebeen RT–PCR-validated as differential, and eightof these 48 have been genetically mapped usinglarge full-sib pedigrees and pre-existing frame-work maps. A further four genes showing homol-ogy to genes involved in xylogenesis in mar-itime pine (based on collaborative work involv-ing INRA scientists) have also been mapped inthe same pedigree. Promisingly, five of these(8 + 4 =) 12 mapped ESTs have co-localized toQTLs for juvenile wood density, an important tar-get trait. The New Zealand team is now evalu-ating two genes by genotyping all of the SNPswithin their full-length cDNAs on an associationpopulation of 2000 unrelated genotypes. This isan onerous and potentially limiting step, but pre-vious work has shown that each SNP must beevaluated for LD with the trait, since intra-genicLD is low for these genes. Nonetheless, if theseresults are positive, then the team may well reachtheir vision of ‘gene-assisted-selection’ — wheresuperior trees are selected based on their DNAsequence and costly phenotypic testing is reduced(or eliminated).

D. Pot’s report (see the first session) alsoupdated the meeting on his progress towards val-idating SNPs within wood property-related genesin Pinus pinaster. The candidate genes for thisproject arose from three sources: functional can-didate genes selected because they are involved inlignin or cellulose biosynthesis; expressional can-didate genes identified by their different expressionpatterns in developing xylem; and ESTs previouslymapped as QTL in a P. pinaster QTL mappingpedigree. Within these candidate genes, the teamdetected SNPs at rates ranging from 12.5/exonkb to 40/ intron kb. Association tests are welladvanced for 12 candidate genes representing 50SNPs, using a half-diallel population comprised of12 parents (therefore 60 families). In contrast tothe unrelated population approach described above,this approach tested LD between SNPs and traits bycorrelating the expected SNP genotype frequencyin each family with the family’s mean phenotypicdata. To date, 11 of the 12 candidate genes tested

thus far showed significant LD between SNPs andtraits. Amongst these were associations betweenlignification genes and wood density or growthwith SNPs accounting for up to 10% to trait vari-ability. Clearly, validated effects of this magnitudewill change the game in forestry genetics.

In ‘Nucleotide Diversity in Loblolly Pine’,G. Brown (University of California, Davis, CA,USA) updated the session on SNP diversity, LDand association tests in P. taeda. He reported thatSNP frequency in P. taeda ranged from 1/196bp for non-synonymous substitutions in exons to1/37 bp in non-coding regions. Estimates of aver-age nucleotide diversity were higher than found inhumans but lower than in Drosophila and maize.Nonetheless, there is plenty of sequence variationto work with. Results for 18 genes confirmed find-ings in other pines showing that intragenic LDclearly exists and is highly variable. Intergenic LDevaluated across two approximately 20 cM inter-vals, each harbouring three genes, was negligible.If this is characteristic of pines, this suggests thatwhole-genome SNP scans to detect QTN will beinefficient. The team is now performing associationtests. They are using a replicated population of 435unrelated individuals, with two copies/genotype.Early results from 15 SNPs drawn from four candi-date genes and evaluated against eight wood prop-erties have revealed ‘tantalizing suggestions’ thatthe group plans to validate on more powerful asso-ciation populations.

In the session’s final talk, G. Gill (Universityof California, Davis, USA), in ‘Detection of theSequence Mutation Responsible for a Null Alleleof Cinnamyl Alcohol Dehydrogenase in LoblollyPine’, reported on progress towards elucidatingthe nucleotide variants responsible for the mostfamous and best-characterized major gene in pine:the CAD null mutant in P. taeda (well describedby MacKay et al., 1997). The homozygous CADnull mutant has CAD activity less than 1% ofwild-type. A result of this CAD downregulationis a significant increase in stem growth, possiblyowing to reallocation of resources from monolig-nol production (this relationship has been suggestedin transgenic poplar). Sequencing the entire cod-ing region from the wild-type and mutant allelerevealed a tandem adenosine insertion in the nullallele that creates a frameshift and leads to a pre-mature stop codon. This sequence mutation is con-sistent with the phenotype. For routine genotyping



of the sequence mutation, Gill and his team usedthe Perkin-Elmer ‘acycloprime fluorescence polar-ization’ methodology to develop an efficient SNP-based allele-specific assay. Trial testing of the assayon 160 P. taeda clones from nine different crossesthat harboured the null allele accurately determinedall of the genotypes; 400 plants were genotyped toestimate the frequency of the null allele and onlytwo offspring were identified from the heterozgy-gous tree from which the original null allele wasdiscovered. This result further supports the notionthat the CAD null mutation is rare.

Functional genomics

The presentations given in this session clearlydemonstrated that functional genomics toolkits arenow established in many forest tree research labo-ratories. The speakers highlighted the breadth of thetoolkits in the forest tree functional genomic com-munity, and emphasized that multiple approachesare often needed to dissect gene function. It wasevident from these presentations that many for-est tree functional genomics projects are usingthese tools to profile gene expression govern-ing developmental or physiological processes thatare more prominent in trees — such as woodformation — or that are a consequence of sea-sonal growth cycles associated with the perenniallifestyle of trees.

Gene discovery and large-scale projects

ESTs and genomic sequences are the keystone tofunctional genomics. With these sequences in hand,it is possible to use a cornucopia of techniques toreconstruct the programs of gene expression thatare involved in complex physiological and devel-opmental phenomena. In a similar vein, it is alsopossible to determine the function of a previouslyuncharacterized gene shown to be associated with aspecific process. The power of functional genomicscomes from recently available large-scale and high-throughput technologies (Colebatch et al. 2002).These technologies enable the simultaneous analy-sis of hundreds or thousands of transcripts, proteinsor metabolites; they also facilitate analysis of tensor hundreds of samples at the same time.

Two speakers (J. Karlsson, S. Ralph) presentedlarge-scale projects that contain a significant ESTsequencing component as part of their multi-faceted programs. Three other presentations

(U. Egertsdotter, S.-K. Yang, M. Stromvik) high-lighted projects associated with the large ESTsequencing effort in loblolly pine headed byNorth Carolina State University (Allona et al.,1998; Whetten et al., 2001). Mention was alsogiven to the Populus genome-sequencing project(G. Tuskan).

In ‘The Swedish Poplar Genome Project’,J. Karlsson (Plant Science Centre, Sweden) pro-vided an update on the program now well underway at the Swedish Centre for Tree FunctionalGenomics (Sterky et al. 1998; Hertzberg et al.,2001). This ambitious project brings together morethan 20 PIs in an integrative, multidisciplinary pro-gram that includes EST sequencing, microarraytranscriptome analysis, 2D PAGE/MALDI/QTOFproteome analysis, various kinds of mass spectro-metry-based metabolite analyses, wood and fibreanalysis, chemometrics, microscopy, bioinformat-ics, as well as a recently installed transgenic facil-ity. As part of this latter capability, this groupwill undertake high-throughput analysis of geneknockouts created by RNAi technology, using bothArabidopsis and poplar. The group is investigatingmany aspects of poplar growth and development,as well as responses to the environment.

S. Ralph (University of British Columbia,Canada) presented ‘Structural and FunctionalApproaches in Deciphering the Spruce andPoplar Genomes’, a comparatively new large-scaleproject funded by Genome Canada. The overallstructure and goals of the project on comparativegenomics in poplar/aspen, spruce, and Arabidopsiswere outlined, which include transcriptomics,proteomics and mapping. The development of full-length and normalized libraries for spruce ESTsequencing was highlighted. Sequences arisingfrom the spruce EST sequencing effort have nowbeen used to construct first-generation microarrays.An overview of BAC end sequencing in poplar,which will be used to support the Populus genomesequencing project, was also described. In addition,progress on the development of Arabidopsis arrayswith spotted long oligomers (70 nucleotides) wasreported.

Comparative genomics

The use of genomic sequence as a tool for com-parative genomics was the focus of G. Tuskan’s



(Oak Ridge National Laboratory, USA) presenta-tion, ‘The Populus Chloroplast Genome: A Com-parison of Genome Structure and Organization’.This project has arisen from the Populus genomesequencing project. Chloroplast contamination inthe DNA template used for sequencing has resultedin up to 450× coverage of the Populus chloro-plast genome. Tuskan’s group has used this finishedgenome to carry out comparative genomics stud-ies, including investigations of phylogenetic rela-tionships. They found that the gene order, genenumber and genome size in the Populus chloro-plast genome is highly conserved when comparedto other plant genomes. The number of mutationsfixed in non-synonymous positions appeared to behigh relative to corn, lotus, rice and Arabidopsis,but is likely an artifact of the hypercorrect nature ofthe Populus sequence. Finally, this unprecedentedsequencing depth has revealed intraclonal variationin chloroplast sequence that is the result of eitherheteroplasmy or chimerism.

Bioinformatics

Processing, archiving and analysing data relatingto thousands of genes at the same time demandsapproaches to data manipulation, annotation andothers that have not been previously encountered inmolecular biology (Tseng et al., 2001). Althoughmany presentations in this session incorporatedvarious aspects of bioinformatics, ‘Bioinformaticsof Loblolly Pine (Pinus taeda): An Overview ofEST Data Resources’ by M. Stromvik (Univer-sity of Minnesota, USA) was dedicated to thisvital component of functional genomics. This pre-sentation covered the approaches that this grouphas used to archive, compile and annotate theloblolly pine EST project headed by North Car-olina State University. Recent developments thatwill help to enable functional analysis include theimplementation of metabolic reconstruction tools(ERGO Plants) and a copy of the Stanford Microar-ray Database. Both are implemented at the Centerfor Computational Genomics and Bioinformatics(http://pine.ccgb.umn.edu/).

Expression profiling

The majority of talks in this session on functionalgenomics were devoted to expression profiling.Three presentations focused on gene expression

associated with wood formation and wood proper-ties. Many wood properties of interest to the forestproducts industry, including wood specific gravity,chemical composition and others, vary consider-ably between different parts of the tree, amongtrees of different species and, to a lesser degree,within species. It is generally assumed that vari-ability in wood structure and reactivity is the resultof adaptation and variability in the underlying pro-cess of wood formation. Transcript profiling studiesare now being conducted in different groups to testdifferent hypotheses linked to this assumption andto attempt to link gene expression, wood formationand wood properties. Several studies have impli-cated cell wall structural proteins or enzymes asdifferentially expressed in correlation with varia-tion or changes in wood properties (reviewed inPlomion et al., 2001). Gene expression profiling ofmuch broader arrays of genes are now becomingthe focus of these studies, aimed at developing amore comprehensive understanding of the molec-ular events that unfold during secondary xylemdifferentiation in trees.

U. Egertsdotter (Institute of Paper Science andTechnology, Atlanta, USA) presented ‘TranscriptProfiling in Wood Formation: Seasonal Variationand Induction of Compression Wood’, linkingmicroarray analyses of gene expression profileswith morphological analyses during wood forma-tion over the course of a growing season and duringthe formation of compression wood induced nat-urally or by bending trees. This study has useda mixed linear model for statistical analysis ofmicroarray data and revealed significant differentialexpression, even with differences as small as 1.4-fold, given appropriate replication of hybridizedcDNAs and slides. This analysis showed that sig-nificant differential gene expression exists betweenthe earlywood and latewood formation phases andthat expression appears coordinated among genesencoding enzymes of related function or belongingto the same pathway, e.g. many genes belongingto the lignin biosynthesis pathway had higher tran-script levels during the latewood formation.

G. Le Provost (INRA, France) presented ‘WoodFormation in Maritime Pine by Transcriptomeand Proteome Analysis’, a study that is usingcDNA–AFLP and 2D gel electrophoresis to com-pare earlywood vs. latewood, and reaction wood vs.opposite wood. This study explored changes that


http://pine.ccgb.umn.edu/


can occur within a single tree as a result of the sea-sonal growth cycle and mechanical stress on woodcells which occurs when a tree is bent (compressionwood). It identified changes in xylem transcriptprofiles during the induction of compression wood,and showed that the response varies significantlyat different times during the active growth season.Both transcriptome and proteome analysis on thesame samples showed greater differences betweenearlywood vs. latewood compared to compressionwood vs. opposite wood.

S.-K. Yang et al. (Texas A&M, USA) pre-sented ‘Gene Expression in Earlywood and Late-wood of Loblolly Pine from Different GeographicalRegions’, reporting on the use of microarray analy-ses and quantitative PCR to compare gene expres-sion profiles during early and latewood formation,as well as wood formation of trees from two prove-nances in the USA that display markedly differentwood density. Their results to date reveal relativelyfew differences in gene expression between bothprovenances, although comparatively more differ-ences were found during the latewood rather thanearlywood growth phase.

Three other presentations focused on forest treeresponses to the environment. Studies such as theseprovide important insights into how trees respondto environmental cues at the molecular level, andhow these responses impact economically impor-tant traits, such as growth rates, wood quality andcold hardiness. These studies also have the poten-tial to address ecological issues. As integral com-ponents of natural forest ecosystems, forest treespresent what is perhaps a unique gateway intothe domain of environmental genomics. Each ofthe speakers presented projects that are combiningphysiological analyses with gene expression profil-ing to examine topics relevant to climate change(more generally referred to as global environmen-tal change): adaptation to temperature, elevatedCO2 and terrestrial nitrogen availability (Sala et al.,2000; Cooke et al., 2003).

In ‘Coldtree: Unraveling Molecular Events Un-derlying Dormancy and Cold Hardiness in ForestTree Seedlings’, M. Van Wordragen (ATO, TheNetherlands) described a project using microarraysto find changes of gene expression associatedwith specific physiological stages of dormancyand cold hardiness in beech and Scots pine. Thisinitial analysis revealed that dehydrins appear to beimportant in these responses.

G. Taylor (University of Southampton, UK) pre-sented ‘Global Gene Expression of a Poplar ForestFollowing Long-term Adaptation to Future CO2Concentrations’, a project that is a part of thePOPFACE experiments in Italy. Taylor’s grouphas used microarrays to define groups of genesthat are upregulated or downregulated in poplarsin response to elevated levels of carbon dioxide.Analysis of these data suggests processes that maybe involved in this response, including ethylenebiosynthesis.

And in the presentation ‘Using a FunctionalGenomics Approach to Delineate Carbon–NitrogenDynamics in Populus’, J. Cooke (Universite Laval,Canada) outlined a project combining differentialdisplay, macroarray analysis, manipulative physi-ology, biochemical analyses and transgenic plantsto examine resource allocation and partitioning inpoplar. Nitrogen-responsive gene expression wascorrelated with concomitant changes in growth andarchitecture, including wood properties.

The International Populus Genome Consortium

The first official meeting of the InternationalPopulus Genome Consortium (IPGC), chaired byG. Tuskan, was held during the PAG XI meet-ing. As a result of the forest genetics commu-nity’s efforts over the past decade to build astrong justification for why trees should be viewedas model systems in plant biology, and the USDepartment of Energy (DOE), Office of Sci-ence’s decision to sequence the first forest treegenome — Populus — the IPGC was founded.

Sequencing the genome of a forest tree willundoubtedly usher in an exciting period of sci-entific discovery for the forestry community. TheUS DOE’s Joint Genome Institute, with fundingthrough the Office of Biological and Environmen-tal Research, will provide a 3× draft sequenceof the female black cottonwood (P. trichocarpa)clone ‘Nisqually-1’ in early 2003 and a second 3×draft in late 2003. Alignment of assembled con-tigs into chromosomal units will then be based oninformation from physical and genetic maps. It isanticipated that annotation will take place usingPopulus-specific gene finding models trained fromfull-length cDNA sequences provided by the UmeaPlant Science Center, Genome British Columbia,Oak Ridge National Laboratory and INRA. All



genomic sequence data will be made freely avail-able to the scientific community. Scientists world-wide will soon have an opportunity to examine fun-damental mechanisms that determine growth anddevelopmental processes in long-lived organisms.However, this same opportunity will also challengethe forest biologist to consider how such a resourceshould best be developed and used.

As a means to advance the capabilities of theforest science community in areas of functionaland comparative genomics, the IPGC was formed.Multinational academic, government and privategroups, working together, will help ensure that thetools, techniques and resources required to enter amolecular era are available to the forest researchcommunity. The IPGC will seek to promote cot-tonwood, aspen and hybrid poplar as model organ-isms (Wullschleger et al., 2002) and to foster thedevelopment of a collaborative worldwide networkof groups interested in tree biology. One of thebenefits associated with IPGC will be the coordi-nated assembly of institutions and scientists work-ing on poplar genomics. It will be the goal of thisgroup to extend and lead post-sequence researchactivities in poplar, and to develop a collabora-tively written science plan for the benefit of thescience community. An international Populus Sci-ence Plan will include a far-reaching summaryof the applications and opportunities associatedwith the arrival of the Populus genome sequence.Science teams are now being formed to exam-ine genetic and genomic resources currently avail-able to Populus researchers, to identify areas inwhich tools, techniques and additional resourcesmust be developed, and to assess applications andopportunities for future research. The IPGC hasenlisted over 200 members from 29 countries tohelp in this endeavour. In general, the Science Planwill focus on technical areas identified as criti-cal to the future success of functional and com-parative genomics in Populus. Currently, teamsare being formed to address the following areas:genetic resources, physical mapping and markerdevelopment, tissue culture and transformation,gene expression/microarrays, metabolic character-ization and metabolomics, protein characterizationand proteomics, high-throughput phenotyping, andinformatics, annotation and database developments.A time-line for development and delivery of anInternational Populus Science Plan suggests that afinal science plan will be available towards the end

of 2003. Scientists from all disciplines interestedin participating in the consortium or in helping todevelop the science plan are invited to register onthe IPGC website (http://www.ornl.gov/ipgc).

Conclusions and future prospects

The Forest Tree Workshop, organized within thePAG XI meeting, was considered very successfulby its participants. The workshop brought togetherscientists with different backgrounds and providedan excellent opportunity to review the differentaspects of forest tree genomics. It clearly showsthat the genomic (r)evolution is providing for-est tree researchers with large amounts of data,providing enormous opportunities to study andunderstand the genetics of ecologically and eco-nomically important traits for trees. In practice,genomic information should enable scientists togain new insights into the variability and expres-sion of genetic sequences to better evaluate theinherent nature of our forest genetic resources, e.g.this knowledge should be useful in breeding pro-grams to accelerate the efficiency of selection andto deliver improved varieties for abiotic and bioticstress adaptation, wood production and wood qual-ity. It will also be germane to conservation pro-grams, where information on the genetic control ofthe physiological functions of trees should makeit possible to preserve critically important geno-types. Another important area of application forfunctional genomics in particular is the adapta-tion of trees to environmental changes of all sorts,and physiological responses to biotic and abioticstresses.

Although we are beginning to understand thegenetic architecture of quantitative traits, we stillknow very little of the actual loci responsible forquantitative variation. EST sequencing and expres-sion analysis are discovering remarkable variationin gene sequence and gene expression, thus iden-tifying many good candidate genes. Now the chal-lenge is to discover whether such variations reallymatter and whether they affect phenotypic varia-tion in natural populations. If we believe that sucha challenge may be tackled through multidisci-plinary cross-overs between scientists involved intree genomics, we also believe that our communityshould work closely with physiologists, ecologists,wood technologists, entomologists, pathologists,


http://www.ornl.gov/ipgc


etc. to define the right traits to analyse. The impor-tance of multidisciplinary approaches becomes evi-dent as we undertake linkage disequilibrium, orassociation studies, between DNA sequence poly-morphisms at candidate gene loci and quantitativetraits. When carried out in well-selected naturalpopulations, such studies are considered to offera highly promising avenue to identify molecularmechanisms contributing to phenotypic variation.Ultimately, they are expected to lead to the identi-fication of quantitative trait nucleotides. However,given the abundance of candidate genes, the prob-lem now lies in the screening of the most relevantcandidates and in associating genetic variation withthe right traits and phenotypes. In order to achievethis goal, we will need to broaden our expertiseto identify the best candidate genes and the rightcandidate traits.

References

Allona I, Quinn M, Shoop E, et al. 1998. Analysis of xylemformation in pine by cDNA sequencing. Proc Natl Acad SciUSA 95: 9693–9698.

Cervera MT, Plomion C, Malpica C. 2000. Molecular markersand genome mapping in woody plants. In Molecular Biologyof Woody Plants , vol 1, Jain SM, Minocha SC (eds). KluwerAcademic: Dordrecht; 375–394.

Chagne D, Brown G, Lalanne C, et al. 2003. Comparative genomeand QTL mapping between maritime and loblolly pines. MolBreed (in press).

Colebatch G, Trevaskis B, Udvardi M. 2002. Functional genomics:tools of the trade. New Phytol 153: 27–36.

Cooke JEK, Brown KA, Wu R, Davis JM. 2003. Gene expressionassociated with N-induced shifts in resource allocation in poplar.Plant Cell Environ (in press).

Flint J, Mott R. 2001. Finding the molecular basis of quantitativetraits: successes and pitfalls. Nature Rev Genet 2: 437–445.

Hertzberg M, Aspeborg H, Schrader J, et al. 2001. A transcrip-tional roadmap to wood formation. Proc Natl Acad Sci USA 98:14732–14737.

MacKay JJ, O’Malley DM, Presnell T, et al. 1997. Inheritance,gene expression and lignin characterization in a mutant pinedeficient in cinnamyl alcohol dehydrogenase. Proc Natl AcadSci USA 94: 8255–8260.

Mauricio R. 2001. Mapping quantitative trait loci in plants: usesand caveats for evolutionary biology. Nature Rev Genet 2:370–381.

Plomion C, Le Provost G, Stokes A. 2001. Wood formation intrees. Plant Physiol 127: 1513–1523.

Sala OE, Chapin FS, Armesto JJ, et al. 2000. Global biodiversityscenarios for the year 2100. Science 287: 1770–1774.

Sewel MM, Neale DB 2000. Mapping quantitative traits in foresttrees. In Molecular Biology of Woody Plants , vol 1, Jain SM,Minocha SC. (eds). Kluwer Academic: Dordrecht; 407–424.

Shield DC, O’Halloran AM. 2002. Integrating genotyping datawith transcriptomic and proteomic data. Comp Funct Genom3: 22–27.

Sterky F, Regan S, Karlsson J, et al. 1998. Gene discovery in thewood-forming tissues of poplar: analysis of 5692 expressedsequence tags. Proc Natl Acad Sci USA 95: 13330–13335.

Tseng GC, Oh M-K, Rohlin L, Liao JC, Wong WH. 2001. Issuesin cDNA microarray analysis: quality filtering, channelnormalization, models of variation, and assessment of geneeffects. Nucleic Acid Res 29: 2549–2557.

Whetten R, Sun YH, Zhang Y, Sederoff R. 2001. Functionalgenomics and cell wall biosynthesis in loblolly pine. Plant MolBiol 47: 275–291.

Wullscleger SD, Jansson S, Taylor G. 2002. Genomics and forestbiology: Populus emerges as the perennial favorite. Plant Cell14: 2651–2655.


Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation http://www.hindawi.com

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttp://www.hindawi.com

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research


International Journal of

Microbiology

conference review report on the forest trees workshop at...

Documents