UBC Faculty Research and Publications

Ty1-copia elements reveal diverse insertion sites linked to polymorphisms among flax (Linum usitatissimum… Galindo-González, Leonardo; Mhiri, Corinne; Grandbastien, Marie-Angèle; Deyholos, Michael K Dec 7, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12864_2016_Article_3337.pdf [ 1.11MB ]
Metadata
JSON: 52383-1.0340319.json
JSON-LD: 52383-1.0340319-ld.json
RDF/XML (Pretty): 52383-1.0340319-rdf.xml
RDF/JSON: 52383-1.0340319-rdf.json
Turtle: 52383-1.0340319-turtle.txt
N-Triples: 52383-1.0340319-rdf-ntriples.txt
Original Record: 52383-1.0340319-source.json
Full Text
52383-1.0340319-fulltext.txt
Citation
52383-1.0340319.ris

Full Text

RESEARCH ARTICLE Open AccessTy1-copia elements reveal diverse insertionsites linked to polymorphisms among flax(Linum usitatissimum L.) accessionsLeonardo Galindo-González1* , Corinne Mhiri2, Marie-Angèle Grandbastien2 and Michael K. Deyholos3AbstractBackground: Initial characterization of the flax genome showed that Ty1-copia retrotransposons are abundant, withseveral members being recently inserted, and in close association with genes. Recent insertions indicate a potentialfor ongoing transpositional activity that can create genomic diversity among accessions, cultivars or varieties. Thepolymorphisms generated constitute a good source of molecular markers that may be associated with phenotype ifthe insertions alter gene activity. Flax, where accessions are bred mainly for seed nutritional properties or for fibers,constitutes a good model for studying the relationship of transpositional activity with diversification and breeding.In this study, we estimated copy number and used a type of transposon display known as Sequence-SpecificAmplification Polymorphisms (SSAPs), to characterize six families of Ty1-copia elements across 14 flax accessions.Polymorphic insertion sites were sequenced to find insertions that could potentially alter gene expression, and apreliminary test was performed with selected genes bearing transposable element (TE) insertions.Results: Quantification of six families of Ty1-copia elements indicated different abundances among TE familiesand between flax accessions, which suggested diverse transpositional histories. SSAPs showed a high level ofpolymorphism in most of the evaluated retrotransposon families, with a trend towards higher levels ofpolymorphism in low-copy number families. Ty1-copia insertion polymorphisms among cultivars allowed ageneral distinction between oil and fiber types, and between spring and winter types, demonstrating theirutility in diversity studies. Characterization of polymorphic insertions revealed an overwhelming associationwith genes, with insertions disrupting exons, introns or within 1 kb of coding regions. A preliminary test onthe potential transcriptional disruption by TEs of four selected genes evaluated in three different tissues,showed one case of significant impact of the insertion on gene expression.Conclusions: We demonstrated that specific Ty1-copia families have been active since breeding commenced inflax. The retrotransposon-derived polymorphism can be used to separate flax types, and the close association ofmany insertions with genes defines a good source of potential mutations that could be associated withphenotypic changes, resulting in diversification processes.Keywords: Ty1-copia, Transposable elements, Flax, Cultivars, Sequence-Specific Amplification Polymorphism (SSAP)* Correspondence: galindo@ualberta.ca1Department of Biological Sciences, University of Alberta, Edmonton, AB T6G2E9, CanadaFull list of author information is available at the end of the article© The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Galindo-González et al. BMC Genomics  (2016) 17:1002 DOI 10.1186/s12864-016-3337-3BackgroundTransposable elements (TEs) are DNA fragments that canmove between genomic locations using a cut and pastemechanism (DNA transposons), or a copy and pastemechanism via an RNA intermediate (retrotransposons).Transposition can result in alterations of gene expressionand diversification between individuals, populations andspecies. TEs are commonly activated upon stresses thatinclude tissue culture, wounding, microbial elicitors andpathogen attack [1–8]. Polyploidization (whether spontan-eous or induced) also mobilizes transposable elements,resulting in genome restructuring, and genetic and epi-genetic effects on gene activity (reviewed in [9]). Selectivebreeding can also affect TE activity. For example, in vege-tatively propagated grape clones, TE insertional polymor-phisms constitute the largest class of mutations [10].Genetic diversity associated with TE polymorphisms hasbeen commonly explored in plant varieties and speciessuch as pepper and tomato [2, 11], barley [12], strawberry[13], coffee [14], blue agave [15] and cashew [16].We previously showed that more than 20% of the flax(Linum usitatissimum) genome is made of TEs [17, 18].The main group represented in the genome are LTR(Long Terminal Repeat) retrotransposons, from which theTy1-copia elements are the most abundant. Ty1-copiahave five main domains encoding proteins required forthe retrotransposition cycle: group-specific antigen(GAG), protease (PR), integrase (INT), reverse tran-scriptase (RT) and ribonuclease H (RH). Because oftheir retrotransposition mechanism, LTRs are identicalat the time of TE insertion [19], and thus sequence varia-tions in them can be used as a molecular clock of inser-tion. LTRs act as promoter sequences since they containcis-acting elements that respond to different stress elici-tors [20–25]. Ty1-copia elements can spread randomlythroughout the genome and are more often associatedwith genes than Ty3-gypsy elements [17, 26, 27]. Therefore,Ty1-copia elements can alter gene regulation [28–30],promote transduction events of one or more genes to othergenomic locations [31–33], or result in epigenetic genesilencing [34, 35]. Additionally, they can also fall insidegenes disrupting gene function or altering splice patterns[36–39].Numerous Ty1-copia members have been recentlyinserted in flax, as inferred from their LTR similarity andgene domain conservation (at least 83 Ty1-copia elementshave 100% LTR similarity) [17]. Furthermore, Ty1-copiaelements have had increasing activity in the flax genomestarting five million years ago [17]. These observationsindicate that Ty1-copia elements could generate polymor-phisms among closely related flax cultivars.Flax is a valuable source of bioproducts derived fromthe seed (i.e. linseed) and stem fiber [40]. Its breeding foreither seed or fiber traits in diverse climates has resultedin diverse cultivars and an array of agrobotanic charac-teristics that have been artificially selected [40]. Whileflax grown for human consumption (seeds are used fornutrition but also for oil derived industrial products), isthe same species as the flax grown mainly to manufacturelinen, they represent two different flax types (oil and fiber)and the products are usually obtained from cultivars (oraccessions) that have been bred to have mainly one of thetwo characteristics [40]. Additionally, flax is a summer an-nual crop in temperate climates and is usually sown dur-ing spring, but winter cultivars have been bred that can besown in the autumn in milder climates. Flax is thereforean interesting system for studying the relationship of TEsto continuous and divergent selection practices.The current study aims to uncover the impact ofspecific Ty1-copia retrotransposon families on diversifi-cation of flax cultivars. We measured the level of poly-morphism among a set of flax cultivars, and analyzedtheir relationship using a TE-based marker system.Since TE insertions within genes are more likely tointerfere with gene function, we characterized the na-ture of target sequences of polymorphic insertions tofind out if they were closely associated with genes, andmeasured the effect of retrotransposon insertion ontranscript expression in selected genes.Several strategies have been devised to find TE inser-tional polymorphisms (reviewed in [41]). Previously, Inter-Retrotransposon Amplified Polymorphism (IRAP) wasused to study flax cultivars and species [42]. Here we useda type of transposon display (TD), known as Sequence-Specific Amplification Polymorphism (SSAP) [43, 44], toevaluate Ty1-copia retrotransposon insertions in 14 flaxaccessions of either oil or fiber types, and spring or wintertypes. In SSAP, TEs and flanking DNA are preferentiallyamplified using a PCR primer that anneals to a sequencespecific to a particular TE family (usually an LTR), anda second primer that anneals to an adaptor ligated to arestriction enzyme site. Our study showed that familiesof flax Ty1-copia TEs have high levels of polymorphismbetween cultivars, indicating recent activity since orga-nized breeding commenced in the last century. Whilethe copy number of each family did not vary greatly be-tween cultivars, some families of TEs were consistentlymore abundant than others across multiple cultivars.Analysis of sequence insertion sites demonstrated thatmany of these Ty1-copia elements inserted within or inclose proximity to genes. Finally, we found one case wherean insertion of a TE in an exon of a Laccase gene de-creased gene expression in roots. Our study demonstratesthat TEs from the Ty1-copia group have been part ofthe diversification associated with breeding, and thatthey may play a role in modifying gene expression pat-terns in the flax genome, which can lead to diversifiedphenotypes.Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 2 of 18ResultsComparison of TE copy number between flax accessionsTo compare the abundance of TE families between flaxcultivars, we designed reverse transcriptase (RT) primers(Additional file 1A) from six selected Ty1-copia elementsrepresentative of six retrotransposon families, and usedquantitative PCR (qPCR) to measure their abundance in14 diverse flax accessions belonging to either oil or fiber,or spring and winter types (Table 1). The retrotransposonfamilies were selected because previous analysis showedthem to have LTRs with high similarity and conservedprotein domains [17], suggesting the elements had beenrecently active, and may therefore be expected to bepolymorphic between cultivars. Families were namedaccording to previously suggested conventions (seeMethods [45]) as: RLC_Lu0, RLC_Lu1, RLC_Lu2,RLC_Lu6, RLC_Lu8 and RLC_Lu28. From each familya representative sequence which showed conservedsites for primer design among family members was se-lected (see additional selection characteristics of repre-sentative sequences in the Methods section). Four ofthe six representative sequences from the selectedretrotransposon families had 100% similarity in theirLTRs and two had LTRs over 99% similar, indicatinginsertion of these elements in the last 200,000 years(Additional file 2). Similarly, we identified the five ex-pected protein domains from Ty1-copia elements inhalf of the representative sequences, and four domainsin the other half (Fig. 1).Quantitative PCR showed differences in TE familyabundances (Fig. 2). When averaged across cultivars,family RLC_Lu2 presented the lowest copy number perhaploid genome (17.7), with families RLC_Lu1, RLC_Lu8and RLC_Lu0 following with 22.5, 25.3 and 38.0 copiesrespectively. Finally, family RLC_Lu28 had 50.1 copiesper haploid genome, while family RLC_Lu6 had the lar-gest average copy number of all with 84.2 copies. Inorder to validate these experimental findings, the copynumber of each TE family was also estimated for the se-quenced CDC (Crop Development Center) Bethune cul-tivar (herein named Bethune) by BLAST alignments ofprimer pairs used for quantitative PCR results. The ex-pected copy number of TEs in each family estimated byBLAST alignments, was supported by the quantitativePCR results, showing a correlation of 0.85, and demon-strating the validity of the analysis. Non-parametric testssupported a significant variation in copy number foreach family between cultivars in all cases (p ≤ 0.0027,Kruskal-Wallis test). Moreover, adjusted p-values(Dunn’s test) for all pairwise comparisons showed sig-nificant differences between some accessions for all TEfamilies tested with the exception of family RLC_Lu6(Fig. 2). No single cultivar or group of cultivars showeda clear expansion of all TE families, but the statisticalanalysis demonstrated a variability that could beexploited through the use of SSAPs.Identification of polymorphic TE insertions using SSAPHaving demonstrated significant variation in TE copynumber between flax accessions, we used SSAP to iden-tify individual insertions that were polymorphic betweenTable 1 Cultivars used for transposon displayCultivar TypeStormont Cirrus fiber springAurore fiber springBelinka fiber springDrakkar fiber springEvea fiber springHermes fiber springViolin fiber winterAdelie fiber winterrdf* oil springBethune oil springLutea oil springBlizzard oil winterOleane oil winterOliver oil winter*rdf is a mutant derived from Bethune and therefore cannot be classified as acultivar per-se. It should be referred as an accessionFig. 1 Diagrams of representative Ty1-copia TEs. Long terminalrepeat (LTR), group-specific antigen domain (GAG), protease (PR),integrase (INT), reverse transcriptase (RT), RNase H (RH)Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 3 of 18the accessions. We identified seven LTR primers thatconsistently amplified distinct bands from the same sixTy1-copia families used for copy number quantification(Additional file 1B). Two primers were used for familyRLC_Lu1 because they generated distinct patterns andresulted in additional polymorphisms. The sevenprimers were used on to amplify DNA from each of the14 flax accessions (Table 1), to generate the SSAP pro-files (an example is shown in Additional file 3). A totalof 219 bands were scored, from which 140 were poly-morphic (63.9% - Table 2). The primers with the lowestnumber of average bands were LTR-RLC_Lu1-primer1and LTR_RCL_Lu1-primer2 (8.1 and 8.5 respectively);however, they also showed the highest rate of polymorph-ism (96.6 and 90% respectively). Conversely, LTR-RLC_Lu6-primer3 produced the highest number of bandsacross cultivars, with an average of 49.6, but showed thelowest rate of polymorphism (25%). The number of ex-pected TEs from the Bethune genomic sequence analysis(Additional file 1A) was also correlated (r = 0.80) with thenumber of scored bands in the same cultivar (Table 2),showing consistency between methods.SSAP bands were converted into a binary matrix (bandpresence = 1, absence = 0), which was used to constructFig. 2 Absolute quantification of Ty1-copia retrotransposon families. The quantification was performed in 14 flax cultivars, based on amplificationfrom their reverse transcriptase (RT) domains. The log10 of molecule copy number (mcn) was calculated using an online tool (see text) thataccounts for plasmid+insert size. This value was used along Ct to generate standard curves to calculate molecule copy numbers for RTs, whichwere normalized to ETIF1 to find absolute copy number. Families depicted are: a RLC_Lu0, b RLC_Lu1, c RLC_Lu2, d RLC_Lu6, e RLC_Lu8, fRLC_Lu28. Error bars = standard deviation. Numbers above represent significant differences of the respective cultivar to other cultivars (Dunn’smultiple comparison test p≤ 0.05) which are numbered as: 1. Adelie, 2. Aurore, 3. Belinka, 4. Bethune, 5. Blizzard, 6. Drakkar, 7. Evea, 8. Hermes, 9.Lutea, 10. Oleane, 11. Oliver, 12. rdf, 13. Stormont Cirrus. 14. Violin. The average copy number for all cultivars in each TE family is also indicatedGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 4 of 18a maximum likelihood (ML) tree using IQ-TREE v1.4.4[46]. For the most part, oil (linseed)-types were moresimilar to each other than they were to fiber-types, withthe exception of the winter fiber variety, Violin (Fig. 3).A grouping pattern was also discerned for the dichot-omy between spring and winter types, with the excep-tion of Adelie, a winter fiber type which seemed closerto spring fiber types, and Lutea, a spring oil type whichwas closer to the winter oil types (Fig. 3).No single band could distinguish all linseed-types fromall fiber-types, or all winter-types from spring-types, how-ever, the definition of a cultivar as either a winter-type orspring-type can sometimes be ambiguous depending onthe breeding program from which it originated [40].Within the linseed-types, Bethune and rdf (reducedfiber) had the most SSAP sequenced bands in common,followed by Oliver and Blizzard (Additional file 4).Bethune and rdf had 4 bands which were only presentin those two accessions: band 25 (LTR-RLC_Lu1-primer1),band 15 (LTR-RLC_Lu1-primer2), band 5 (LTR-RLC_Lu2-primer1), and band 4 (LTR-RLC_Lu8-pri-mer1). At the same time, Oliver and Blizzard had 2 bandsthat were exclusively in those two cultivars: band 17(LTR-RLC_Lu0-primer3), and band 14 (LTR-RLC_Lu1-primer1).In the case of the fiber-types, Evea, Drakkar, andHermes shared the most bands. A band derived fromLTR-RLC_Lu2-primer1 (band 10) was present in sevenof the eight fiber-types tested and was absent from thelinseed types (Additional file 4); the same was true forband 6 from LTR-RLC_Lu28-primer1 but the band wasalso present in rdf and Bethune (linseed types). Onemore band from LTR-RLC_Lu28-primer1 was commonto eight fiber-types (band 12), although in this case, onelinseed-type (Lutea) also had the band. One single bandwas present only in all spring-fiber types (LTR-RLC_Lu2-primer1, band 4), but this one was also present in Lutea(Additional file 4).Analysis of flax genomic sequences targeted bypolymorphic insertionsFrom the 140 polymorphic bands detected using SSAPs,99 were successfully excised and sequenced. Most of thefailed sequencing occurred with the highest molecularweight bands, which were either difficult to re-amplify ordid not otherwise produce high quality sequence. Some ofthe resulting sequences were redundant, probably becausethe restriction enzyme used for the SSAPs found crypticsites in larger fragments; alternatively, different cultivarscould have the same TE insertion accumulating mutationsthat generate a new restriction site, resulting in bands withdifferent electrophoretic mobility, but which representedthe same insertion event. After filtering for residual redun-dancies, sequences where the restriction site fell inside orTable 2 SSAP band scoring and polymorphic bandsAccession LTR-RLC_Lu0-primer-3LTR-RLC_Lu1-primer-1LTR-RLC_Lu1-primer-2LTR-RLC_Lu2-primer-1LTR-RLC_Lu6-primer-3LTR-RLC_Lu8-primer-1LTR-RLC_Lu28-primer-1Totalsrdf 18 12 13 12 50 8 20 133Bethune 18 10 14 12 50 8 20 132Lutea 14 4 6 12 48 8 20 112Oleane 14 4 5 9 48 8 22 110Blizzard 16 10 9 9 49 10 17 120Oliver 17 10 10 11 49 8 20 125Violin 16 12 11 12 49 10 24 134Adelie 14 7 8 14 50 10 20 123S. Cirrus 13 8 6 17 48 5 23 120Evea 14 10 9 13 52 8 23 129Drakkar 15 9 8 14 50 12 23 131Hermes 14 8 9 13 50 8 24 126Belinka 20 4 5 13 50 8 22 122Aurore 18 6 6 11 51 9 23 124Average numberof bands15.8 8.1 8.5 12.3 49.6 8.6 21.5 17.8Scored positions 28 29 30 22 56 22 32 219Polymorphicpositions17 28 27 16 14 19 19 140%Polymorphism 60.7 96.6 90.0 72.7 25.0 86.4 59.4 63.9Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 5 of 18just besides a TE, and sequences where no LTR couldbe identified, 66 unique insertion sites where found(Additional files 4 and 5). Each insertion was classifiedaccording to its Ty1-copia family and was mapped to thegenome assembly according to the annotation depositedin phytozome (Additional file 4). Of the 66 insertions, 14(21.2%) interrupted annotated exons, 30 (45.5%) were inintrons, 11 (16.7%) were within 1 kb of a gene openingreading frame (upstream or downstream), and 11 (16.7%)were characterized as intergenic (where the TE wasinserted at a distance of more than 1 kb from any anno-tated gene). Altogether, more than 83.3% of the cloned TEinsertions mapped within genes, or within 1 kb of a gene.For insertions in introns, exons, or within 1 kb of theCDS, the inferred transcription sense strand of the TE andgene were the same in 30 cases. Conversely, in 28 cases,the TE and associated gene were transcribed from oppos-ite strands (Additional file 4).To validate the results of the SSAPs, we conductedgenomic PCR assays of 28 selected insertions (Additionalfile 6) on each of the 14 flax accessions. We designedfrom each TE insertion site sequence (Additional file 5)one specific primer complementary to the flanking gen-omic DNA (listed in Additional file 1C) and performedPCR with the corresponding LTR-RLC SSAP primer(Additional file 1B). Nineteen (67.9%) of the validatedinsertions showed a perfect match or nearly perfectmatch to the polymorphisms initially assessed withSSAPs, while the remaining bands had different levelsof disagreement (Additional file 6).We used Gene Ontology (GO) categories to classify thegenes that were found to be associated with polymorphicTE insertions. The genes represented 15 cellular compo-nents, 12 molecular functions, and 14 biological processes(Additional file 7). Nine genes where classified as respon-sive to stress, four in DNA or RNA metabolism, nine in cellorganization and biogenesis, nine corresponded to proteinmetabolism, five were related to transcription, three totransport, five to development, six involved in signal trans-duction and one was related to electron transport or energyprocesses. None of the categories were enriched when theArabidopsis orthologs were compared to the background ofall annotated genes using AgriGO (data not shown).qRT-PCR analysis of selected genes with polymorphic TEinsertionsTo assess effects of TE insertions on gene expression, weselected four flax genes from Additional file 4 that wereFig. 3 ML tree using 14 flax cultivars. The ML tree was built using a general time reversible model for binary data and 1000 bootstrap replicates(bootstrap support is shown for each branch). The colored groupings reflect different flax types: orange (fiber spring - FS), purple (fiber winter -FW), green (oil spring - OS), blue (oil winter - OW). Two biological replicates were used per cultivar with the exception of Bethune witheight replicatesGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 6 of 18expected to be constitutively expressed under normalconditions, so that any effects of TE insertion on geneexpression could be detected. In making this selection,we relied partly on flax RNA-seq data of control plantsfrom an experiment on the flax-fusarium interactionperformed in our lab (Galindo-González & Deyholos,in preparation), and on comparisons to the presumptiveArabidopsis orthologs of our flax genes, since extensivetranscript expression data is available for Arabidopsis(ThaleMine - [47, 48]). Four flax genes were selected forqRT-PCR analysis: Pyruvate carboxylase (Lus10022077), aLaccase-13-related gene (Lus10026400), and two Rabgap/TBC domain containing proteins (Lus10036500 andLus10040349). The two Rabgap/TBC domain containingproteins had 83.6% nucleotide identity, and bore the TEinsertions in different regions (Additional file 4). Addition-ally, in order to study potential positional effects of the TEinsertions, we selected genes harbouring TEs in exons orintrons and TE-gene associations in sense or antisenseorientation (Fig. 4). Primers were designed downstreamthe insertion following the theoretical gene transcriptionorientation (Fig. 4). Five flax accessions that were poly-morphic for insertions in these four genes were selectedto be assayed on three different tissues (leaves, root andstem) by qRT-PCR: Lutea (TE insertion in exon 8 ofpyruvate carboxylase), Oleane (insertion in intron 6 offirst Rabgap/TBC domain containing protein); StormontCirrus (TE insertion in exon 3 of the Laccase-13-related gene), and an insertion in a second Rabgap/TBC domain containing protein of intron 4, which wasalso present in Bethune (Fig. 4); and Oliver, which hadno TE insertions in any these four genes. The results ofthe qRT-PCR analysis showed several differences (somewith statistical significance) in the relative expression ofthe genes tested (Fig. 5). Only the laccase gene harbouringa TE insertion (Stormont Cirrus accession) displayed asignificant decrease in root gene expression (p < 0.005)compared to homologous laccase gene of the other fourcultivars evaluated that did not bear the insertion(Fig. 5). There was also a significant difference betweenthe expression of Lutea/Oleane when compared withOliver/Bethune, but the expression in Stormont Cirruswas lower than in any other of the four cultivars. Thethree other genes did not show decrease in gene ex-pression in agreement with the accession containingthe TE insertion.Fig. 4 Diagrams of genes bearing Ty1-copia TE insertions. The location of the primers used to test for changes in gene expression is displayed.Gene expression was tested in five flax cultivars (Lutea, Oleane, S. Cirrus, Bethune and Oliver), which were polymorphic for the insertions. Thename of the TE family is above each represented TE. Orientation of the genes and TEs is as depicted after mapping using phytozome. Genes aredrawn according to scale while TEs (not to scale) are depicted only to show insertion positionsGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 7 of 18DiscussionTE activity and genomic copy numberQuantitative PCR using six Ty1-copia families in 14 flaxcultivars demonstrated copy number variation betweenTE families, but also within each family between culti-vars. Artificial selection through plant breeding involvessubjecting plants to diverse stress conditions (e.g. drought,cold), and growing them under agricultural and laboratorypractices which are not always common in natural envi-ronments. Mobilization of TEs has accompanied theprocesses of breeding, as evidenced by the level of inter-varietal polymorphisms found using different transposon-derived markers [10, 12, 13, 42, 49].A first clue that genomes diverge with respect to trans-poson history is a difference in abundance of specific TEs.Our approach to assess copy numbers of the selected TEfamilies in flax followed a previous report that quantifiedthe highly abundant BARE-1 retrotransposons from barleyin several cultivars [50]. We found significant differencesin copy numbers between cultivars for each TE familyexamined and differences when testing for multiple com-parisons in five out of six TE families (Fig. 2). One of themost extreme examples of genome diversification due toactive TEs in a plant genomes was demonstrated by ampli-fications of the mPing MITE (Miniature Inverted-repeatTE) in different rice landraces, where the element wasactively transposing and ranged from 50 to more than1000 copies [51]. While the differences in abundance ofthe Ty1-copia families we tested in flax were not as largeas reported for mPing in rice, they were nevertheless, indi-cative of recent activity of these TEs.In general, the absolute copy numbers we reported(Fig. 2), probably underestimate the actual abundance ofthe Ty1-copia families, due to frequent recombination andhigh mutation rates expected among LTR elements [52].This results in modification of binding sites for qPCRprimers or loss of internal retrotransposon domains, withconcomitant creation of solo LTRs [53–55], whichwould not be accounted for by our method, based onamplification of the internal RT gene. Furthermore, theexpected number of annealing sites from our BLASTanalysis (Additional file 1A) is almost always lower thanthe calculated copy number by qPCR (Fig. 2) in Bethune(this happens for 5 out of 6 families). This is probably a re-sult of unassembled genome regions which are yet to bereported in the database (in general regions which aredifficult to assemble are rich in repeats such as TEs).Nonetheless, the copy numbers we reported for eachfamily for Bethune (Fig. 2), were positively correlatedwith the number of TEs identified by BLAST alignmentof primer binding sites to the Bethune genome assem-bly (r = 0.85), showing the validity of our approach.There was also a proportional high correlation betweenthe TEs counted by BLAST alignment and the numberFig. 5 Normalized gene expression of four genes bearing TEinsertions in three different tissues. Each ΔCt corresponds to theaverage of four biological replicates, each with three qRT-PCRtechnical replicates. PYR (Pyruvate carboxylase – Lus10022077),RAB1 (Rabgap/TBC domain containing protein 1 – Lus10036500),RAB2 (Rabgap/TBC domain containing protein 2 – Lus10040349),LAC (Laccase-13-related – Lus10026400). All pairwise comparison ineach gene and tissue, were performed using unpaired two-tailed t-tests, and significant differences were calculated after Bonferronicorrection (p < 0.005). Numbers above represent significantdifferences of the respective cultivar to other cultivars which arenumbered as: 1. Lutea, 2. Oleane, 3. Stormont Cirrus, 4. Oliver, 5.Bethune. Error bars = standard deviation. The red outline in rootsdepicts the only expression pattern in agreement with the insertionof a TE (TE inserted in the Laccase gene of S. Cirrus and absent in therest of the cultivars)Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 8 of 18of SSAP bands found for Bethune (r = 0.80). In this casethe number of SSAP bands for Bethune (Table 2), wasgenerally lower than the number of expected hits(Additional file 1A), and calculated TE copy numbers(Fig. 2) in each family, because this transposon displaytechnique is only efficient for insertions located closeto a restrictions site, and thus only reveals a subset ofall insertions.The highest estimated copy numbers and SSAP bandsacross all cultivars were found on family RLC_Lu6(Compare Fig. 2 and Table 2). Interestingly, this familyalso had the lowest proportion of polymorphic bands(Table 2), and most of its flanking sequences were notrelated to gene regions (Additional file 4). An explan-ation for the insertion pattern and abundance of familyRLC_Lu6 could be related to a lower level of negativeselection, since TEs inserted in regions of low geneabundance, may not be as detrimental for the genome.Likewise, a lower level of polymorphism reflects in-activity, and it is therefore likely that family RLC_Lu6expanded in the flax genome before breeding of thesecultivars, and has been mostly quiescent since. In fact,this family’s representative sequence has the largestLTR divergence (Additional file 2), supporting thishypothesis.Opposite to what happened with family RLC_Lu6, thereare compelling clues that the lower copy number familiesfrom our study were active in the recent past as demon-strated by the differences in TE copy numbers betweencultivars (Fig. 2), level of SSAP polymorphism (Table 2),their LTR similarity (Additional file 2), and that they relatemore closely to genes (Additional file 4). This is in agree-ment with low copy number TEs catalogued as beinginserted closer to genes [56], and also being more active inrecent past than high-copy number TEs in plants likemaize [57, 58]. Analysis of the maize genome suggest thatthe transition from low copy to high copy number TEsshould be placed in the 10–100 copies range [56], which isin agreement with the difference between RLC_Lu6 andour low copy number families.SSAP markers associate with flax typesSSAPs were performed with the same six families forwhich we measured TE copy number. Based on 140polymorphic bands, we produced a ML tree, in whichaccessions showed associations reflecting the divisionbetween fiber and oil types, with the exception of Violin,a fiber winter type that was closer to the oil winter typesthat to its other fiber winter partner (Adelie) (Fig. 3).Violin is a cultivar that behaves in the field like a dualpurpose (oil/fiber) winter flax, and is genetically closerto an oil type, while Adelie has characteristics of a springfiber (Jean-Paul Trouvé - personal communication). Thiswould explain the close relationship of Violin to winterlinseed types in the ML tree, as well as the close rela-tionship of Adelie to the fiber spring types. Previously,the division of fiber and oil types was supported bymolecular studies looking at genes closely linked to thedistinct phenotypes of these groups. The sad2 gene, in-volved in fatty acid metabolism, was used to determinean ancestral state of domestication of oil over fiber flax[59]. Additional candidate genes related to fiber devel-opment or oil metabolism that can distinguish betweenfiber and linseed varieties, were also found in a geneticdiversity study by Soto-Cerda and collaborators [60].The TE markers found in our study are therefore agood source of molecular variation that can be linkedto genes involved in the divergence of flax types (seemore below).Analysis of TE insertions and potential impact on genesCharacterization of insertion sites from polymorphicbands of the SSAPs in flax cultivars, evidenced a highpercentage of association of TEs to genes. PolymorphicTE insertions can result in genome divergence throughgenome restructuring, gene mutation and regulationchanges (e.g. LTRs upstream of genes can change expres-sion patterns), which at the same time depend on the TE’sinsertion site preference, and regulation of their transpos-ition by host mechanisms (e.g. epigenetic control). Whilethe mechanisms for insertion site selection are still notcompletely understood, insertional bias is evident forcertain TE families. Young Copia-like retrotransposonshave been shown to insert more randomly than Gypsy-likeelements in Arabidopsis, and are associated with euchro-matic gene-rich regions [26, 27]. Similarly, we previouslyfound that in flax, recently inserted Ty1-copia elementswere non-randomly associated with gene regions andconstituted the largest superfamily of TEs in the flaxgenome [17]. Our results here confirmed that numerousTy1-copia TEs are biased towards insertion close to or in-side genes. GO (Gene Ontology) classification (Additionalfile 7), however, showed no bias towards specific func-tional categories of genes.TD has been often used to find polymorphic markers tostudy intraspecific genetic diversity [12, 13, 42, 61, 62], butthese types of studies rarely characterize polymorphicinsertion sites with detail at the sequence level. Wesuccessfully sequenced 66 non-redundant insertions indifferent genomic locations. Analysis of these insertionsites showed some interesting genes that could be relatedto agronomic traits, and represent potential candidates forfuture studies. For example, band 11 from LTR-RLC_Lu0-primer3 (Additional file 4), was characterized as a TEinsertion on intron 19 of Pyruvate dehydrogenase E1. Thisgene is involved in fatty acid biosynthesis, and has beenidentified as potentially associated with divergent selectionbetween flax types [60]. We also found a TE insertion onGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 9 of 18exon 2 of a Pinoresinol-lariciresinol reductase 3 gene(PLR 3) for cultivars Bethune and the associated mu-tant accession rdf (LTR-RLC_Lu2-primer1, band 5 -Additional file 4). PLR 1 is a key enzyme in flax lignanbiosynthesis [63, 64]; lignans act as antioxidants, as wellas having beneficial effects on human health [65].Another interesting gene annotated as a Laccase-13-related, had a TE insertion on exon 3 (LTR-RLC_Lu1-primer1, band 18 - Additional file 4) that was presentin six fiber cultivars. Laccases catalyze the last step inlignin biosynthesis [66], and downregulation of ligninbiosynthetic genes (including laccases) has been associ-ated with the hypolignification of bast fibers, a desirablecharacteristic for easier harvesting [67, 68]. Anothergene with a TE insertion, was 1-aminocyclopropane-carboxylate synthase 2-related (LTR-RLC_Lu8-primer1,band 7 - Additional file 4), which was previously identi-fied as Lu-ACS5 (1-aminocyclopropane-carboxylatesynthase 5) [69]. ACS enzymes are involved in ethylenesynthesis, and a previous study of ACS gene expressionin flax roots showed that transcript abundance of ACS5did not change in response to treatment with auxin an-tagonists, although transcripts of four other ACS genesdid. Whether or not this might be related to the TEinsertion is yet to be investigated. We also found a TEoverlapping with a WRKY27 transcription factor (LTR-RLC_8-primer1, band 11 - Additional file 4). Mutantsof this gene in Arabidopsis have delayed wilting uponinfection with the bacterial pathogen Ralstonia solana-cearum, showing that the gene might be a negativeregulator of defense response [70].Our results showed that 83.3% of insertions disruptexons, or introns or are otherwise in close proximity tocoding regions (Additional file 4). Fourteen (21.2%) ofthe characterized insertions in our study disruptedexons. While the most common result of an exon dis-ruption is loss of gene function, this loss can result in adesirable agronomic trait. As an example, glutinous riceis the product of a retrotransposon disrupting an exonof the granule-bound starch synthase gene [71]. We alsofound 30 TEs mapped to introns. Intron insertions canresult in different patterns: An LTR-retrotransposon in-sertion in different introns of a MADS-box transcriptionfactor of different apple varieties causes transcript sup-pression leading to seedless fruits [72], and waxy kernelphenotypes in maize result from alternative splicing pat-terns caused by retrotransposon insertions in introns ofan amylose biosynthesis gene [73]. Regulation of expres-sion can also result from TEs that do not disrupt thecoding sequence; we found 10 insertions within 1 kb ofgenes. Examples of the impact of extragenic insertionsinclude: insertions of LTR retrotransposons adjacent toMYB genes involved in anthocyanin biosynthesis result-ing in skin color variation in grape cultivars [74], and inthe production of blood oranges [28]; insertion of a retro-transposon in the 5′ UTR region of a vernalization gene(VRN1), which allows winter wheat to grow as a spring-type wheat [75]; and an increase in disease resistance torice blast due to an insertion of an LTR retrotransposon inthe promoter of the Pit resistance gene [29].We found a particular example of a TE carrying an F-box domain protein between its two LTRs. This TE(band 8 from LTR-RLC_Lu6-primer3 in Additional file 4)has a recognizable RNAse H (ribonuclease H) domainnear the 3′ LTR and therefore represents a functionalretrotransposon that has acquired a gene. This genecapture and capacity to mobilize the gene is known astransduplication, and has been widely seen with over3000 Pack-MULEs (Mu-like Elements) in rice that havecaptured over 1000 genes [76]. Retrotransposons in riceand sorghum have also been shown to capture numer-ous genes [77].TE impact on flax gene expressionA preliminary assay was performed to test the effects offlax TEs on gene expression, using qRT-PCR on fourgenes with insertions in either exons or introns. Onlythe Laccase gene demonstrated a significant decreasedtranscript abundance in roots that correlated to thepresence of the TE insertion: the cultivar StormontCirrus harbouring a TE insertion had lower relativetranscript abundance than the other cultivars withoutthe insertion (Fig. 5). Observation of the qRT-PCRbands on a gel (not shown), demonstrated that the ex-pected band was present in all cultivars, which wouldmean that likely scenarios for repression would be: i)anti-sense gene transcripts generated from readout ofthe TE inserted in opposite orientation of the gene(Fig. 4), that could potentially be used for the gener-ation of small RNAs tagging the gene for inactivationvia methylation [34, 78], or, ii) the generation of a dif-ferent splice form, which conserves the exon tested byqRT-PCR, but has reduced transcript abundance as aconsequence of the modification [71, 73]. Other statis-tical differences where found, but none of these were inagreement with the presence or absence of a TE inser-tion. We believe these differences could be related tocultivar specific differential expression (Fig. 5).For the Pyruvate carboxylase gene, we expected thatthe TE insertion in exon 8 would result in transcript al-terations for Lutea but this was not the case. For both ofthe Rabgap/TBC domain genes which had insertions onintrons, no impact on gene expression was evidenced.These results show no common mechanisms bywhich these insertions may alter gene expression. Inser-tion on exons would be expected to be directly disrup-tive but only in one of two cases a change was noticed.Opposite orientation of TE-derived transcripts couldGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 10 of 18create epigenetic-mediated gene silencing [34, 78], butthis should depend on actual transcriptional activationof our TEs which might not be happening under ourconditions. And TEs inserted in introns could changegene expression or splice forms, but this does not al-ways happen, and alternate transcripts can be createdfrom one single gene with an insertion [73]. Finally, ifthe TE insertion is present in just one allele of the gene,the TE effects can be masked by the other allele func-tioning normally.In future studies, stress conditions or treatments whichupregulate genes with inserted TEs might proof to be abetter strategy to discern if gene expression levels areaffected by the TE, for two reasons: i) the stress cangenerate a higher response of the host gene that can bemore distinct that a low constitutive expression if theTE really alters gene expression, and ii) the stress mayalso upregulate the transcription of the TE, increasingthe chance of readout transcripts that can be used forthe production of small RNAs that can mediate silen-cing. These examinations should be coupled with ex-periments: i) to assess the production of small RNAsand methylation state of the gene, and ii) revise if TEinsertions are homozygous and if different splice formsare produced from the host gene.ConclusionsBased on our findings, we can conclude that there havebeen recent active retrotransposition events since breed-ing started for the tested cultivars. The TE markers foundusing SSAPs were useful to separate the major flax types,and their level of polymorphism further showed that theyhave an impact on diversifying flax cultivars. While not allflax TEs examined fall in gene-rich regions [17], we nowknow that the transposition of most studied families hereis biased towards these regions and their study constitutesa good source of novel mutations that can be used to findpotential linkage to diversifying phenotypes, which is thebasis for creating new cultivars. In fact, strong proof ofTE-mediated diversification exists in closely related spe-cies of Arabidopsis [35] and rice [52] and in cultivars ofrice [51] and maize [79]. No matter what the adaptive fateof these insertions may be, the mobilization of TEs amongflax cultivars constitutes a powerful tool in diversity stud-ies. However, understanding how these insertions influ-ence genome restructuring and shape gene evolutionrequires studying related cultivars and species to deter-mine what insertions may be under purifying selectionand which ones are being positively selected as part of thenormal functioning of the genome. The TE insertionsfound here, in different gene regions and in different ori-entations, open the door to study their potential influenceon gene regulation on a case by case basis.MethodsPlant materialFor determination of TE family copy numbers, eightplants from each of 14 flax cultivars or accessions (Table 1)were grown in a growth chamber at the University ofAlberta under the following settings: seeds sown in podswith a 50/50 soil/sand mix, 16 h of light/8 h of dark (0.132μMoles of light), 22 °C, 50% humidity. Aerial sections(stems and leaves) were harvested after 2 weeks of growthand instantly frozen with liquid nitrogen in 2 mL tubes.These cultivars were selected based on their expectedbroad genetic base (this fact is more common for springtypes). Only two representatives of fiber-winter wereselected due to availability, because these cultivars areuncommon and not widely grown or bred.For SSAPs, 14 flax cultivars were used (Table 1).Plants were grown in greenhouse conditions (14 h oflight, 24 °C day/20 °C night, 40% humidity) at theNational Institute for Agronomic Research (INRA) inVersailles, France. Seeds were sown in pods with a 50/50soil/sand mix, and left to grow for 2 weeks before aerialsections were collected in 2 mL tubes and instantly frozenin liquid nitrogen.For testing the expression of genes bearing polymorphicTE insertions among cultivars additional plants of eachcultivar were grown in the same growth chamber at theUniversity of Alberta under the same conditions used fordetermination of TE copy number. Stems, 5–10 youngleaves (including the apical meristem) and roots wereharvested after 2–3 weeks of growth.Nucleic acids extraction and cDNA synthesisThe samples for SSAPs were ground with a plastic pestlemaintaining the tube in liquid nitrogen until achieving afine powder. Samples for SSAP validation, transposonfamilies copy number determination and gene expressionwere ground adding an autoclaved 5.6 mm stainless steelbead, and using a Retsch MM301 mixer mill (Retsch,Haan, Germany) with two cycles of 1 min at 20 Hz.DNA extraction was performed using the DNeasyPlant Mini Kit (QIAGEN, Venlo, The Netherlands).Sample quantification was performed with a NanodropND-1000 spectrophotometer (Thermo Fisher Scientific,Waltham, MA, USA).RNA was extracted using the RNeasy Plant Mini Kit(QIAGEN, Venlo, The Netherlands), and quantity wasassessed using a Nanodrop ND-1000 spectrophotometer(Thermo Scientific, Waltham, MA, USA). A DNAse treat-ment was performed for 30 min at 37 °C after extractionwith DNase I (Thermo Scientific, Waltham, MA, USA).For cDNA synthesis 500 ng of DNAse treated RNA wereused to perform reverse transcription using the RevertAidH Minus Reverse transcriptase under the manufacturerGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 11 of 18specifications and using oligo dT (18) (Thermo Scientific,Waltham, MA, USA). To test for residual contamin-ation of DNA a PCR was performed with primers fromthe eukaryotic translation initiation factor 3E (ETIF3E)which has constitutive expression in the tested tissues(Additional file 1E). The PCR was run with 1× buffer,2 mM MgCl2, 0.2 mM dNTPs, 0.4 μM of each primer,5 ng of cDNA and 1.5 units of Taq polymerase (ThermoScientific, Waltham, MA, USA). Cycling conditions were94 °C for 2 min, followed by 35 cycles of 94 °C for 30 s,60 °C for 30 s and 72 °C for 1 min, finalizing with anextension at 72 °C for 5 min.TE primersRetrotransposon sequences were obtained from ourprevious study on transposable elements of flax [17]. Todesign Ty1-copia primers, TE families were first defined:family membership of a Ty1-copia element was estab-lished with a threshold similarity of at least 80% in atleast 80% of the aligned sequence, following previouslyestablished rules for family membership [45]. The com-parison was performed on the 554 non-redundant re-verse transcriptase (RT) domain sequences, which werefirst predicted using RepeatExplorer [80, 81], and thenused as input for CD-HIT-est [82, 83] using an identitycutoff and minimal alignment coverages of 0.8. Familieswere named using the suggested designation of class,order and superfamily [45], followed by a species desig-nation, and a number corresponding to the specific TE(e.g. Retrotransposon-LTR-Copia from Linum usitatissi-mum family 0, representative sequence 1 = RLC_Lu0-1).Selected families with evidence of recent insertion (highsimilarity among its LTRs and conserved domain proteins)were selected for primer design. To calculate the insertionage of the TEs, first LTR pairs from each element wherealigned using ClustalW from MEGA v6.06 [84, 85] andthe Kimura two-parameter method [86] was used to cal-culate nucleotide substitution. Then, the age of insertionwas estimated as t = K/2r, where K corresponds to thenucleotide substitution per site and r corresponds to thenucleotide substitution rate which in this case was takenfrom a previous study used for dating LTR retrotranspo-sons in Arabidopsis [26]. Presence of the main proteindomains in Ty1-copia elements: GAG (group-specificantigen), PR (protease), INT (integrase), RT (reverse tran-scriptase) and RH (Ribonuclease H), was assessed usingconserved domains from NCBI [87, 88], and RepeatEx-plorer [80, 81].To design reverse transcriptase (RT) primers to assessTE copy number (see below), the RT nucleotide se-quences from all members in each TE family werealigned using ClustalW [84] from MEGA v6.06 [85]using the following parameters: a gap opening penalty of15 and a gap extension penalty of 6.66 for both pairwiseand multiple alignments, DNA weight matrix – IUB,transition weight of 0.5, negative matrix off and delaydivergent sequences that have less than 40% similarity.For each family, one representative sequence bearingconserved sites for primer design from all (or most)family members, was used as input for Primer3 [89, 90]with the following parameters: primer size range be-tween 18 and 24 bp, temperature between 57 and 63°,product size 100–200 bp, and GC content between 40and 60% (the rest of the parameters were left by de-fault). Candidate representative sequences from eachalignment were chosen based on preliminary bioinformat-ics analysis showing protein domain conservation (at least4 out of 5 of expected domains (GAG, PR, INT, RT andRH), and high LTR similarity to all members of the family(not shown). The reference gene used to normalize copynumber was ETIF1 (eukaryotic translation initiation factor1) which has been previously tested in flax quantitativegene expression [91]. Selected RT primer pairs (Additionalfile 1A), were aligned to the flax genome (Bethune) usingBLAST to get an estimate of the expected copy numbersper TE family. From two primer pairs designed per family(six families in total) for qPCR (see below), the one with abetter standard curve was selected in each family. Theexpected number of reported hits to the flax genome(Additional file 1A) reflects matches where both primersfrom the pair have 100% identity and 100% coverage tothe match site, with no gaps. Additionally, for all matchesreported in all families, the distance between the primerpair reflects the range of the expected size displayed inAdditional file 1A.To design primers for SSAPs, LTRs from all familymembers in each of the selected families were alignedfollowing the same alignment parameters as for theRT sequences when designing primers to assess TEfamily copy number. After filtering largely divergentand redundant sequences from the alignment, thecomplete retrotransposon sequences were checked forthe presence of internal EcoRI sites in order tominimize the chance of amplifying retrotransposon in-ternal regions. The LTR alignments were then scannedfor a region that is conserved among most aligned ele-ments of the family, to maximize the generation SSAPbands. A representative sequence was selected in eachfamily (Additional file 2), and the conserved regionwas then used as input for Primer3 [89, 90] along withthe primer corresponding to the EcoRI adapter withthe following parameters: primer size range between20 and 24 bp, temperature between 55 and 62° andGC content between 30 and 70% (the rest of the pa-rameters were left by default). A total of 19 primerswere designed for six TE families, but only seven wereused for the final experiment (Additional file 1B).Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 12 of 18Transposon family copy numberTo find the absolute copy number of TEs from each ofthe families used in the SSAPs (see below), we per-formed qPCR on DNA samples from the 14 cultivarsusing reverse transcriptase (RT) TE primers (Additionalfile 1A). The amplifications were then compared tostandard curve dilutions of the cloned amplified RTfragments (see below).PCR to amplify the RT regions to be cloned wasperformed with 1× Taq buffer, 2 mM of MgCl2, 0.2 mMdNTPs, 0.2 μM of each primer and 1 unit of recombin-ant Taq polymerase (Thermo Fisher Scientific, Waltham,MA, USA). Cycling conditions were as following: 94 °Cfor 2 min followed by 35 cycles of 94 °C for 30 s, 60 °Cfor 30 s and 72 °C for 1 min, with a final extension at72 °C for 5 min. PCR was run on a 1% agarose gel at90 V for 60 min and the expected amplicon size wasassessed. Then, the bands were eluted using the WizardSV gel and PCR clean-up system (Promega, Madison,WI, USA). Eluted products were quantified using aNanodrop ND-1000 spectrophotometer (Thermo FisherScientific, Waltham, MA, USA). For sequencing, 150 ngof the eluted product was used along with a primer at afinal concentration of 0.25 μM (forward or reverseprimers corresponding to the same primers used forPCR). Sequencing reactions were performed with theBigDye terminator v3.1 cycle sequencing kit (AppliedBiosystems - Thermo Fisher Scientific, Waltham, MA,USA) using a 3730 Genetic Analyzer equipment (AppliedBiosystems -Thermo Fisher Scientific, Waltham, MA,USA). Sequencing products were aligned with the originalRT sequences to confirm that amplification products wereas expected.To clone the amplification products, ~5–8 ng of theinsert (this varied depending on the amplicon size) werecloned into the PGEM-T vector II system (Promega,Madison, WI, USA) to create a 3:1 (insert:vector) molarratio. Ligation products were transformed into JM109high-efficiency competent cells (Promega, Madison, WI,USA), following the manufacturer recommendations.One hundred microliters of the transformed cultureswere plated into LB-agar plates with 2% X-gal, 20%IPTG and 50 ng/μL of ampicillin. Cultures were incu-bated overnight (ON) at 37 °C and white colonies wereselected as positive for the insertion.Selected colonies were grown in LB supplied with100 ng/μL of ampicillin. Tubes were placed in a shaker at200 rpm ON (minimum of 12 h) at 37 °C. Plasmids wereextracted from concentrated bacterial cultures using theQIAprep Spin Miniprep Kit (QIAGEN, Venlo, TheNetherlands) and concentrations were measured using aNanodrop ND-1000 spectrophotometer (Thermo FisherScientific, Waltham, MA, USA). To confirm the identityof the cloned products, inserts were reamplified fromplasmids using the same conditions previously mentioned,and resequenced using 575 ng of plasmid and the genericT7 and Sp6 primers matching the vector.Five nanograms of DNA from eight samples of eachof the 14 cultivars, and a 1:10 8-serial dilution (5 to 5 ×10−7 ng) of each plasmid with the different TE familyinserts, were used for qRT-PCR with 5 μL of SYBR green(Molecular Probes – Thermo Fisher Scientific, Waltham,MA, USA) and 2.5 μL of the mixed primer pair (3.2 μM),in a 10 μL reaction (three technical replicates per eachsample or dilution); three additional reactions withwater instead of DNA were used as controls in each pri-mer tested (these showed no amplicons after reactionswere completed). Samples were aliquoted in 384-wellplates using a Biomek 3000 Laboratory AutomationSystem (Beckman Coulter, Brea, CA, USA), and the qRT-PCR was run using a QuantStudio 6 Flex Real-Time PCRsystem (Applied Biosystems-Life Technologies, Carlsbad,CA, USA). Cycling conditions were 94 °C for 2 min,followed by 35 cycles of 94 °C for 30 s, 60 °C for 30 s and72 °C for 45 s. A melting curve stage was added: 95 °C for15 s, 60 °C for 1 min and 95 °C for 15 s.To find out the molecule copy number (mcn) in thedilution series, we used the amount of DNA from eachpoint of the serial dilution and the size of the plasmidplus insert [92], and performed a log10 transformation.The standard curve was built by plotting the Ct values(average of technical replicates) against the log10 ofmcn. The linear equation for the slope y = mx + b wasused to determine the log10 mcn intercepts (x) for the Ctvalues of the eight replicates in each of the 14 cultivarsfor all primers tested. The log10 mcn values where thenback-transformed using the power function (10x). Thesame procedure was conducted to find out the copynumbers for the reference gene. The copy numbers ofthe reverse transcriptases of each family for each samplewere normalized to the copy numbers of the referencegene (ETIF1) and the average absolute copy number andstandard deviation (from the 8 replicates), were calcu-lated and plotted for each cultivar and TE family.Statistical differences for each TE copy number amongcultivars, were determined using the non-parametric testof Kruskal-Wallis, followed by multiple comparisons usingDunn’s test, using GraphPad Prism version 6.0 (GraphPadSoftware, La Jolla California USA). Correlation coefficientswere calculated in excel for the relationship between theexpected copy number of TEs in each family estimatedusing the primer pairs to BLAST against the flax genome,and either, the calculated copy numbers from qPCR, orthe number of scored bands in the SSAPs.Sequence-Specific Amplification Polymorphism (SSAP)One hundred nanograms of each DNA sample wereused for restriction digestion at 37 °C for 16 h, with 10Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 13 of 18units of EcoRI and supplemented with 0.03 mg of BSA and1× restriction-ligation buffer (10 μL of the digestion wereused to check the restriction in a 1% agarose gel). A 1 μMmixture of EcoRI adapter 1 and 2 (Additional file 1B), wereligated to the digested ends for 16 h, using 0.2 mM ATP,1× restriction-ligation buffer and 0.004 units of T4 DNAligase (Invitrogen, Carlsbad, CA, USA). Ligations were cen-trifuged and diluted with 80 μL of 1× TE.To confirm ligation efficiency, a cold PCR (with non-radioactively labelled TE primer) was performed using1× Taq Buffer, 2 mM MgCl2, 0.2 mM of each dNTP,0.4 μM of the specific TE primer and 0.4 μM of theEcoRI primer 00 (Additional file 1B), 1.5 units of recom-binant Taq DNA polymerase (Thermo Fisher Scientific,Waltham, MA, USA) and 5 μL of the diluted restriction-ligation. Cycling conditions were 94 °C for 5 min,followed by 35 cycles of 94 °C for 30 s, 56 °C for 30 sand 72 °C for 1 min, finalizing with an extension at 72 °Cfor 5 min.Retrotransposon primer labelling with P33 was per-formed using the LTR TE specific primers (Additionalfile 1B) at a final concentration of 4 μM, 1× kinasebuffer A, 0.5 units of T4 kinase (Thermo Fisher Scien-tific, Waltham, MA, USA) and 1 μCi of gamma ATP33(PerkinElmer Health Sciences, Boston, MA, USA). Thecycling conditions to label the primer were: 1 h at 37 °Cand 15 min at 70 °C to inactivate the kinase enzyme (theoligo was kept at −20 °C until used in the SSAP PCR).SSAP PCR was performed with 1× buffer, 2 mMMgCl2, 0.2 mM of each dNTP, 0.4 μM of adapter primer,0.16 μM of specific radioactively labeled primer, 1.5 unitsof recombinant Taq DNA polymerase (Thermo Scien-tific, Thermo Fisher Scientific, Waltham, MA, USA) and2.5 ng of the restriction-ligation product. Cycling condi-tions were as following: 94 °C for 5 min followed by13 cycles of 94 °C for 30 s, 65 °C for 30 s and 72 °C for2 min; then 25 cycles of 94 °C for 30 s, 56 °C for 30 sand 72 °C for 2 min; finishing at 72 °C for 10 min. AfterPCR the product was diluted 1:1 with 2× AFLP loadingbuffer and kept at −20 °C until running the gel. PCRproducts were separated in 6% denaturing polyacryl-amide gels on a Bio-Rad Sequi-Gen GT electrophoresissystem (Bio-Rad, Hercules, CA, USA). After the run, thegel was dried and adhered to Whatman paper, and ex-posed from 1 to 3 days to Kodak Biomax XAR films(Carestream Health Inc., Rochester, New York, USA)and then developed for band scoring.Band scoring and maximum likelihood (ML) treeExposed films displaying the SSAP band patterns werecaptured as images (.tif files) and used as input in GelA-nalyzer [93] where bands were digitally scored aspresent = 1 or absent = 0. The scored bands were used tocreate a binary matrix that was formatted as a nexus file,and utilized as input to generate a maximum likelihoodtree using IQ-TREE v1.4.4 [46]. The parameters usedwere: −m TEST (to test for an optimal substitutionmodel of evolution) and –b 1000 (to do 1000 bootstrapassays). The general time reversible model for binarydata was selected to produce a consensus ML tree whichwas visualized an formatted in MEGA v6.06 [85].Band recovery and sequencingTo recover the polymorphic bands, the exposed film wasoverlaid on the original dried gels on Whatman (bothfilm and gel were pinned previously on the corners toallow matching). A clean scalpel was used to cut themapped band on surface of the gel-Whatman assembly,and the detached piece was placed in a 1.5 mL tube with35 μL of nuclease free water. The band in water was vor-texed for 1 min and spun down for incubation at 37 °Cfor 15–16 h. The liquid was recovered to a new 1.5 mLtube and 5 μL were used for a PCR with 1× Taq buffer,2 mM of MgCl2, 0.2 mM dNTPs, 0.2 μM of each primerand 1 unit of recombinant Taq polymerase (ThermoFisher Scientific, Waltham, MA, USA). Primers for thePCR corresponded to the LTR specific primer for theband along with the EcoRI adapter primer (Additionalfile 1B). Cycling conditions were as following: 94 °C for2 min followed by 35 cycles of 94 °C for 30 s, 56 °C for30 s and 72 °C for 2 min, and a final extension at 72 °Cfor 10 min. The total PCR (25 μL) was run on a 1%agarose gel at 80 V for 60 min and the bands were elutedusing the Wizard SV gel and PCR clean-up system(Promega, Madison, WI, USA). Eluted products werequantified using a Nanodrop ND-1000 spectrophotometer(Thermo Fisher Scientific, Waltham, MA, USA). For se-quencing, 75 to 225 ng of the eluted product was used(depending on the band size) along with a primer at a finalconcentration of 0.25 μM (forward or reverse primers cor-responded to the same primers used for PCR). Sequencingreactions were performed with the BigDye terminatorv3.1 cycle sequencing kit (Applied Biosystems - ThermoFisher Scientific, Waltham, MA, USA) using a 3730 Gen-etic Analyzer equipment (Applied Biosystems -ThermoFisher Scientific, Waltham, MA, USA).Sections corresponding to the LTRs from sequencedSSAP bands were first compared to the correspondingLTR reference sequence used for primer design (Additionalfile 5) to see if amplification matched the expected family.Since some LTR variation exists between members ofthe same family we allowed an identity of over 90% formatches between reference and sequenced bands for allfamilies, with the exception of family RLC_Lu6 wereidentity of LTRs from the reference Ty1-copia elementwas the lowest (Additional file 2), and an 80% identitywas allowed). Then, the complete sequence includingLTR plus flanking region was compared to the flaxGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 14 of 18genome deposited in phytozome [94, 95] using blastnand Gbrowse to determine the insertion site of the TE.Accepted blastn results bore 100% identity (or close to100% - sequencing errors might change a few bases) forthe full sequenced band (LTR+flanking region) whenthe band was initially present (SSAP) in the cultivarBethune, which represents the reference genome. Whenthe band was not present in this cultivar then the simi-larity to the LTR and the flanking region was found indifferent sections of the genome as expected.Once mapped on the genome, the IDs of the flax geneswith associated TE insertions, were used to find theclosest Arabidopsis thaliana ortholog from a databasepreviously obtained by performing blast analysis of flaxtranscripts against the peptide TAIR database (release 10).Arabidopsis ortholog IDs were then used to perform func-tional Gene Ontology (GO) classification using the GOannotation search from TAIR [96], to find out what cat-egories of genes were predominantly affected by TE inser-tions. An additional enrichment analysis was performedusing AgriGO [97, 98] by comparing the Arabidopsisorthologs to the background of all Arabidopsis genesusing a Fisher test, a Yekutieli multiple test adjustmentand a minimum of 1 mapping read.Validation of TE insertionsOnce the TE insertions were mapped to the genome,primers were designed from the flanking regions of thetransposable element insertion site to validate the poly-morphism encountered with the initial SSAP gels. For theTEs that fell within genes we looked for paralogs and per-formed an alignment to select allele-specific primerswhich would not bind related genes. Twenty eight primerswere designed with Primer3 [89, 90] under the same pa-rameters cited above to be compatible with the originalLTR-derived primers, but with a product size range of200–1000 bp (Additional file 1C), and were used to per-form amplification in eight replicate plants in each one ofthe 14 cultivars. Five nanograms of DNA from each ofeight samples per cultivar was used for PCR on 384-wellplates using 1× Taq buffer, 2 mM of MgCl2, 0.2 mMdNTPs, 0.2 μM of each primer and 1 unit of recombinantTaq polymerase (Thermo Scientific - Thermo Fisher Sci-entific, Waltham, MA, USA) in a 10 μL reaction. Cyclingconditions were as following: 94 °C for 2 min followed by35 cycles of 94 °C for 30 s, 60 °C for 30 s (this temperaturevaried according to the primer used – see Additional file1C) and 72 °C for 1 min, with a final extension at 72 °Cfor 5 min. Bands were visualized in 1% agarose gels run at90 V for 60 min.Expression of genes with TE insertionsWe selected four genes to test their expression in fivecultivars that were polymorphic for the respective TEinsertion (Fig. 4). The primer pairs per gene werenamed according to their gene of origin: Pyruvate carb-oxylase (PYR), Rabgap/TBC domain containing protein-1(RAB1), Laccase-13-related (LAC), and Rabgap/TBCdomain containing protein-2 (RAB2) (Fig. 4 andAdditional file 1D).cDNA from three tissues (leaf + apical meristem, stemand roots) from four biological replicates (differentplants), was used to evaluate the primer pairs of each geneusing qRT-PCR. Seven reference genes were tested forstability among tissues and replicates [91] (Additionalfile 1E). While all seven genes were stable, the threewith higher stability according to Bestkeeper [99] andGeNorm [100] were GAPDH (glyceraldehyde 3-phosphatedehydrogenase), ETIF5A (eukaryotic translation initiationfactor 5 A), and EF1A (elongation factor 1-α). These wereused to generate the geometric mean for relative quantifi-cation of the test genes using the ΔCt of the reference –the test gene. Statistical differences in each gene amongcultivars were calculated using unpaired two-tailed t-testsafter a Bonferroni correction for multiple comparisons(p < 0.005), using GraphPad Prism version 6.0 (GraphPadSoftware, La Jolla California USA).Samples were aliquoted in 384-well plates (with threetechnical replicates per sample and tissue combination)using a Biomek 3000 Laboratory Automation System(Beckman Coulter, Brea, CA, USA), and the qRT-PCRwas run using a QuantStudio 6 Flex Real-Time PCR sys-tem (Applied Biosystems-Life Technologies, Carlsbad,CA, USA). Sample reactions were done in 10 μL with5 μL of SYBR-green (Molecular Probes – Thermo FisherScientific, Waltham, MA, USA), 2.5 μL of the mixed pri-mer pair (3.2 μM) and 2.5 μL of a 1:50 dilution of thesynthesized cDNA. Cycling conditions were: 95 °C for2 min followed by 40 cycles of 95 °C for 30 s, 60 °C for1 min. A melting curve stage was added: 95 °C for 15 s,60 °C for 1 min and 95 °C for 15 s.Additional filesAdditional file 1: (A-E). Primers. Sets of primers for all experimentsperformed. (XLSX 18 kb)Additional file 2: Insertion age and domains of representativesequences from selected Ty1-copia families. Characterization of six Ty1-copia elements which represent the six families investigated in this study.(XLSX 11 kb)Additional file 3: SSAP example of retrotransposon family RLC_Lu1. TheSSAP was run for cultivars: 1. Bethune, 2. Lutea, 3. Stormont Cirrus, 4.Adelie, 5. Aurore, 6. Belinka, 7. Blizzard, 8. Drakkar, 9. Evea, 10. Hermes, 11.Oleane, 12. Oliver, 13. rdf, 14. Violin. (PPTX 606 kb)Additional file 4: Mapping of insertion sites of SSAP bands sequenced.(DOC 187 kb)Additional file 5: Sequences from SSAP eluted bands. The LTRsequences from representative members of the six TE familiesinvestigated are presented along the sequenced sections of the SSAPbands (Genbank accession numbers: KX364308 to KX364373). BoxedGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 15 of 18sequences correspond to the LTR of representative elements of each TEfamily (Additional file 2), in direct and in reverse orientation; some LTRboundaries were adjusted from the original predictions of LTR finder(Additional file 2) after mapping analysis. The LTR primer region is shownin blue. The sequenced LTR region present in SSAP bands is shown ingreen. The polymorphic bands are named according to the originalnumber given when eluting the band from the gel and match Additionalfile 4 descriptions. Sequences of the EcoRI adaptor primer and of the LTRprimer have been trimmed from the sequences. Names of the TEs areexplained in the Methods section. (DOCX 33 kb)Additional file 6: Comparison of selected SSAP band scores and PCRvalidation in 14 flax accessions. SSAP band polymorphisms were selectedfor validation using conventional PCR (see Methods). 1 = present, 0 = absent,W =weak band at expected size, (?) = weak band at non-expected size,1+L = expected band plus an additional lower band, 1+H = expectedband plus an additional higher band, P = polymorphic among replicatesof same cultivar. Colors from accessions represent flax types as in Fig. 3.(DOCX 41 kb)Additional file 7: GO functional categories of flax closest orthologs inArabidopsis. The flax gene sequences from Additional file 4 where usedto search the Arabidopsis closest ortholog, and these were used forfunctional classification using Gene Ontology (see Methods). (XLSX 9 kb)AbbreviationsACS: Aminocyclopropane-carboxylate synthase; CDC: Crop DevelopmentCenter; EF1A: Elongation factor 1-α; ETIF1: Eukaryotic translation initiationfactor 1; ETIF3E: Eukaryotic translation initiation factor 3 E; ETIF5A: Eukaryotictranslation initiation factor 5 A; GAG: Group-specific antigen;GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; GO: Gene Ontology;INRA: National Institute for Agronomic Research (Institut National de laRecherche Agronomique); INT: Integrase; IRAP: Inter-RetrotransposonAmplified Polymorphism; LAC: Laccase; LTR: Long Terminal Repeat; Lu: Linumusitatissimum; mcn: Molecule copy number; MITE: Miniature Inverted-repeatTE; ML: Maximum likelihood; MULE: Mu-like Element; MYB: Myeloblastosis;PLR: Pinoresinol-lariciresinol reductase; PR: Protease; PYR: Pyruvatecarboxylase; qPCR: Quantitative PCR; RAB: Rabgap/TBC domain containingprotein; rdf: Reduced fiber; RLC: Retrotransposon-LTR-Copia; RNAse H/RH: Ribonuclease H; RT: Reverse transcriptase; SSAP: Sequence-SpecificAmplification Polymorphism; TD: Transposon display; TE: Transposableelement; VRN: VernalizationAcknowledgementsWe would like to thank Troy Locke from the Molecular Biology Service Unitat the University of Alberta, for his advice in qPCR and qRT-PCR experiments.Julien Daniel from the Host-Retrotransposon Interactions laboratory at InstitutJean-Pierre Bourgin - INRA, for his help with plant harvesting and nucleicacid extraction.FundingGenome Canada ABC Program grant TUFGEN; Natural Sciences andEngineering Research Council (NSERC - Canada) Discovery Grant 2014-03596to MKD; NSERC Alexander Graham Bell Canada Graduate Scholarship (CGSD), NSERC Michael Smith Foreign Study Supplement (CGS- MSFSS), andAlberta Innovates – Technology Futures (AITF) top-up award to LGG.The IJPB benefits from the support of the Labex Saclay Plant Sciences-SPS(ANR-10-LABX-0040-SPS) to MAG.Availability of data and materialSequences from SSAP bands are available in the Genbank repository underaccession numbers KX364308 to KX364373. All remaining datasets supportingthe results and conclusions of this article are included within the article and itsadditional files (Additional files 1, 2, 3, 4, 5, 6 and 7).Authors’ contributionsLGG designed and conducted the experiments, performed all analyses andwrote the manuscript. CM helped with experimental design, supervisedexperiments, and edited the final version of the manuscript. MAG helpedwith experimental design, supervised research and edited the final version ofthe manuscript. MKD supervised the research and edited the final version ofthe manuscript. All authors read and approved the final manuscript.Competing interestsThe authors declare they do not have any competing interests.Consent for publicationNot applicable.Ethics approval and consent to participateNot applicable.Author details1Department of Biological Sciences, University of Alberta, Edmonton, AB T6G2E9, Canada. 2Institut Jean-Pierre Bourgin, INRA, AgroParisTech, CNRS,Université Paris-Saclay, RD10, 78026 Versailles Cedex, France. 3Department ofBiology, University of British Columbia, Okanagan campus, Kelowna, BC V1V1V7, Canada.Received: 9 June 2016 Accepted: 23 November 2016References1. Hirochika H, Sugimoto K, Otsuki Y, Tsugawa H, Kanda M. Retrotransposonsof rice involved in mutations induced by tissue culture. Proc Natl Acad SciU S A. 1996;93:7783–8.2. Tam SM, Mhiri C, Vogelaar A, Kerkveld M, Pearce SR, Grandbastien M-A.Comparative analyses of genetic diversities within tomato and peppercollections detected by retrotransposon-based SSAP, AFLP and SSR. TheorAppl Genet. 2005;110:819–31.3. Okamoto H, Hirochika H. Efficient insertion mutagenesis of Arabidopsis bytissued culture-induced activation of the tobacco retrotransposon Tto1.Plant J. 2000;23:291–304.4. Melayah D, Bonnivard E, Chalhoub B, Audeon C, Grandbastien M-A. Themobility of the tobacco Tnt1 retrotransposon correlates with itstranscriptional activation by fungal factors. Plant J. 2001;28:159–68.5. Mhiri C, De Wit PJGM, Grandbastien M-A. Activation of the promoter of theTnt1 retrotransposon in tomato after inoculation with the fungal pathogenCladosporium fulvum. Mol Plant-Microbe Interact. 1999;12:592–603.6. Liu ZL, Han FP, Tan M, Shan XH, Dong YZ, Wang XZ, et al. Activation of arice endogenous retrotransposon Tos17 in tissue culture is accompanied bycytosine demethylation and causes heritable alteration in methylationpattern of flanking genomic regions. Theor Appl Genet. 2004;109:200–9.7. Kimura Y, Tosa Y, Shimada S, Sogo R, Kusaba M, Sunaga T, et al. OARE-1, aTy1-copia retrotransposon in oat activated by abiotic and biotic stresses.Plant Cell Physiol. 2001;42:1345–54.8. Grandbastien MA. LTR retrotransposons, handy hitchhikers of plantregulation and stress response. Biochim Biophys Acta. 2015;1849:403–16.9. Parisod C, Alix K, Just J, Petit M, Sarilar V, Mhiri C, et al. Impact oftransposable elements on the organization and function of allopolyploidgenomes. New Phytol. 2010;186:37–45.10. Carrier G, Le Cunff L, Dereeper A, Legrand D, Sabot F, Bouchez O, et al.Transposable elements are a major cause of somatic polymorphism in Vitisvinifera L. PLoS One. 2012;7:e32973.11. Tam SM, Lefebvre V, Palloix A, Sage-Palloix A-M, Mhiri C, Grandbastien M-A.LTR-retrotransposons Tnt1 and T135 markers reveal genetic diversity andevolutionary relationships of domesticated peppers. Theor Appl Genet.2009;119:973–89.12. Soleimani VD, Baum BR, Johnson DA. Genetic diversity among barleycultivars assessed by sequence-specific amplification polymorphism. TheorAppl Genet. 2005;110:1290–300.13. He P, Ma Y, Dai H, Li L, Liu Y, Li H, et al. Development of Ty1-copiaretrotransposon-based S-SAP markers in strawberry (Fragaria x ananassaDuch.). Sci Hortic (Amsterdam). 2012;137:43–8.14. Hamon P, Duroy P-O, Dubreuil-Tranchant C, Mafra D’Almeida Costa P, DuretC, Razafinarivo NJ, et al. Two novel Ty1-copia retrotransposons isolated fromcoffee trees can effectively reveal evolutionary relationships in the Coffeagenus (Rubiaceae). Mol Genet Genomics. 2011;285:447–60.15. Bousios A, Saldana-Oyarzabal I, Valenzuela-Zapata AG, Wood C, Pearce SR.Isolation and characterization of Ty1-copia retrotransposon sequences in theblue agave (Agave tequilana Weber var. azul) and their development asSSAP markers for phylogenetic analysis. Plant Sci. 2007;172:291–8.Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 16 of 1816. Syed NH, Sureshsundar S, Wilkinson MJ, Bhau BS, Cavalcanti JJ V, Flavell AJ.Ty1-copia retrotransposon-based SSAP marker development in cashew(Anacardium occidentale L.). Theor Appl Genet. 2005;110:1195–202.17. Galindo González L, Deyholos MK. Identification, characterization anddistribution of transposable elements in the flax (Linum usitatissimum L.)genome. BMC Genomics. 2012;13:644.18. Wang Z, Hobson N, Galindo L, Zhu S, Shi D, Mcdill J, et al. The genome offlax (Linum usitatissimum) assembled de novo from short shotgun sequencereads. Plant J. 2012;72:461–73.19. Kumar A, Bennetzen J. Plant retrotransposons. Annu Rev Genet.1999;33:479–532.20. Mhiri C, Vernhettes S, Casacuberta JM. The promoter of the tobacco Tnt1retrotransposon is induced by wounding and by abiotic stress. Plant MolBiol. 1997;33:257–66.21. Salazar M, González E, Casaretto JA, Casacuberta JM, Ruiz-Lara S. The promoterof the TLC1.1 retrotransposon from Solanum chilense is activated by multiplestress-related signaling molecules. Plant Cell Rep. 2007;26:1861–8.22. Suoniemi A, Narvanto A, Schulman AH. The BARE-1 retrotransposon istranscribed in barley from an LTR promoter active in transient assays. PlantMol Biol. 1996;31:295–306.23. Woodrow P, Pontecorvo G, Ciarmiello LF, Fuggi A, Carillo P. Ttd1a promoteris involved in DNA-protein binding by salt and light stresses. Mol Biol Rep.2011;38:3787–94.24. Cavrak VV, Lettner N, Jamge S, Kosarewicz A, Bayer LM, Mittelsten Scheid O.How a retrotransposon exploits the plant’s heat stress response for itsactivation. PLoS Genet. 2014;10:e1004115.25. Takeda S, Sugimoto K, Otsuki H, Hirochika H. A 13-bp cis-regulatory elementin the LTR promoter of the tobacco retrotransposon Tto1 is involved inresponsiveness to tissue culture, wounding, methyl jasmonate and fungalelicitors. Plant J. 1999;18:383–93.26. Pereira V. Insertion bias and purifying selection of retrotransposons in theArabidopsis thaliana genome. Genome Biol. 2004;5:R79.27. Lockton S, Gaut BS. The contribution of transposable elements to expressedcoding sequence in Arabidopsis thaliana. J Mol Evol. 2009;68:80–9.28. Butelli E, Licciardello C, Zhang Y, Liu J, Mackay S, Bailey P, et al.Retrotransposons control fruit-specific, cold-dependent accumulation ofanthocyanins in blood oranges. Plant Cell. 2012;24:1242–55.29. Hayashi K, Yoshida H. Refunctionalization of the ancient rice blast diseaseresistance gene Pit by the recruitment of a retrotransposon as a promoter.Plant J. 2009;57:413–25.30. Studer A, Zhao Q, Ross-Ibarra J, Doebley J. Identification of a functionaltransposon insertion in the maize domestication gene tb1. Nat Genet.2011;43:1160–3.31. Xiao H, Jiang N, Schaffner E, Stockinger EJ, Van Der Knaap E. Aretrotransposon-mediated gene duplication underlies morphologicalvariation in tomato fruit. Science. 2008;319:1527–31.32. Bureau TE, White SE, Wessler SR. Transduction of a cellular gene by a plantretroelement. Cell. 1994;77:479–80.33. Elrouby N, Bureau TE. Bs1, a new chimeric gene formed by retrotransposon-mediated exon shuffling in maize. Plant Physiol. 2010;153:1413–24.34. Kashkush K, Feldman M, Levy AA. Transcriptional activation ofretrotransposons alters the expression of adjacent genes in wheat. NatGenet. 2002;33:102–6.35. Hollister JD, Smith LM, Guo Y, Ott F, Weigel D, Gaut BS. Transposableelements and small RNAs contribute to gene expression divergencebetween Arabidopsis thaliana and Arabidopsis lyrata. Proc Natl Acad Sci U SA. 2011;108:2322–7.36. Tsuchiya T, Eulgem T. An alternative polyadenylation mechanism cooptedto the Arabidopsis RPP7 gene through intronic retrotransposondomestication. Proc Natl Acad Sci U S A. 2013;110:E3535–43.37. Costa JH, De Melo DF, Gouveia Z, Cardoso HG, Peixe A, Arnholdt-Schmitt B.The alternative oxidase family of Vitis vinifera reveals an attractive model tostudy the importance of genomic design. Physiol Plant. 2009;137:553–65.38. Pouteau S, Spielmann A, Meyer C, Grandbastien M-A, Caboche M. Effects ofTnt1 tobacco retrotransposon insertion on target gene transcription. MolGen Genet. 2000;228:233–9.39. Kanazawa A, Liu B, Kong F, Arase S, Abe J. Adaptive evolution involvinggene duplication and insertion of a novel Ty1/copia-like retrotransposon insoybean. J Mol Evol. 2009;69:164–75.40. Muir A, Wescott N, editors. Flax, the genus Linum. London: Taylor & Francisgroup; 2003.41. Kalendar R, Flavell AJ, Ellis THN, Sjakste T, Moisy C, Schulman AH. Analysis ofplant diversity with retrotransposon-based molecular markers. Heredity(Edinb). 2010;106:520–30.42. Smykal P, Bacova-Kerteszova N, Kalendar R, Corander J, Schulman AH,Pavelek M. Genetic diversity of cultivated flax (Linum usitatissimum L.)germplasm assessed by retrotransposon-based markers. Theor Appl Genet.2011;122:1385–97.43. Waugh R, McLean K, Flavell AJ, Pearce SR, Kumar A, Thomas BBT, et al.Genetic distribution of Bare-1-like retrotransposable elements in the barleygenome revealed by sequence-specific amplification polymorphisms (S-SAP). Mol Gen Genet. 1997;253:687–94.44. Syed NH, Flavell AJ. Sequence-specific amplification polymorphisms (SSAPs):a multi-locus approach for analyzing transposon insertions. Nat Protoc.2006;1:2746–52.45. Wicker T, Sabot F, Hua-Van A, Bennetzen JL, Capy P, Chalhoub B, et al. Aunified classification system for eukaryotic transposable elements. Nat RevGenet. 2007;8:973–82.46. Nguyen L, Schmidt HA, Von Haeseler A, Minh BQ. IQ-TREE : a fast andeffective stochastic algorithm for estimating maximum-likelihoodphylogenies. Mol Biol Evol. 2014;32:268–74.47. ThaleMine [Internet]. Available from: https://apps.araport.org/thalemine/begin.do. Accessed 25 Aug 201548. Krishnakumar V, Hanlon MR, Contrino S, Ferlanti ES, Karamycheva S, Kim M,et al. Araport: the Arabidopsis information portal. Nucleic Acids Res.2015;43:D1003–9.49. Huang X, Lu G, Zhao Q, Liu X, Han B. Genome-wide analysis of transposoninsertion polymorphisms reveals intraspecific variation in cultivated rice.Plant Physiol. 2008;148:25–40.50. Soleimani VD, Baum BR, Johnson DA. Quantification of the retrotransposonBARE-1 reveals the dynamic nature of the barley genome. Genome.2006;396:389–96.51. Naito K, Cho E, Yang G, Campbell MA, Yano K, Okumoto Y, et al. Dramaticamplification of a rice transposable element during recent domestication.Proc Natl Acad Sci U S A. 2006;103:17620–5.52. Ma J, Bennetzen JL. Rapid recent growth and divergence of rice nucleargenomes. Proc Natl Acad Sci U S A. 2004;101:12404–10.53. Devos KM, Brown JKM, Bennetzen JL. Genome size reduction throughillegitimate recombination counteracts genome expansion in Arabidopsis.Genome Res. 2002;12:1075–9.54. Vitte C, Panaud O. Formation of solo-LTRs through unequal homologousrecombination counterbalances amplifications of LTR retrotransposons inrice Oryza sativa L. Mol Biol Evol. 2003;20:528–40.55. Ma J, Devos KM, Bennetzen JL. Analyses of LTR-Retrotransposon structuresreveal recent and rapid genomic DNA loss in rice. Genome Res. 2004;14:860–9.56. Baucom RS, Estill JC, Chaparro C, Upshaw N, Jogi A, Westerman RP, et al.Exceptional diversity, non-random distribution, and rapid evolution ofretroelements in the B73 maize genome. PLoS Genet. 2009;5:e1000732.57. Meyers BC, Tingey SV, Morgante M. Abundance, distribution, andtranscriptional activity of repetitive elements in the maize genome. GenomeRes. 2001;11:1660–76.58. Diez CM, Meca E, Tenaillon MI, Gaut BS. Three groups of transposableelements with contrasting copy number dynamics and host responses inthe maize (Zea mays ssp. mays) genome. PLoS Genet. 2014;10:e1004298.59. Allaby RG, Peterson GW, Merriwether DA, Fu Y-B. Evidence of thedomestication history of flax (Linum usitatissimum L.) from genetic diversityof the sad2 locus. Theor Appl Genet. 2005;112:58–65.60. Soto-Cerda BJ, Diederichsen A, Ragupathy R, Cloutier S. Geneticcharacterization of a core collection of flax (Linum usitatissimum L.) suitablefor association mapping studies and evidence of divergent selectionbetween fiber and linseed types. BMC Plant Biol. 2013;13:78.61. Zerjal T, Rousselet A, Mhiri C, Combes V, Madur D, Grandbastien MA, et al.Maize genetic diversity and association mapping using transposableelement insertion polymorphisms. Theor Appl Genet. 2012;124:1521–37.62. Smýkal P. Development of an efficient retrotransposon-based fingerprintingmethod for rapid pea variety identification. J Appl Genet. 2006;47:221–30.63. Hano C, Martin I, Fliniaux O, Legrand B, Gutierrez L, Arroo RRJ, et al.Pinoresinol–lariciresinol reductase gene expression and secoisolariciresinoldiglucoside accumulation in developing flax (Linum usitatissimum) seeds.Planta. 2006;224:1291–301.64. Renouard S, Tribalatc M-A, Lamblin F, Mongelard G, Fliniaux O, Corbin C, etal. RNAi-mediated pinoresinol lariciresinol reductase gene silencing in flaxGalindo-González et al. BMC Genomics  (2016) 17:1002 Page 17 of 18(Linum usitatissimum L.) seed coat: Consequences on lignans andneolignans accumulation. J Plant Physiol. 2014;171:1372–7.65. Westcott ND, Muir AD. Flax seed lignan in disease prevention and healthpromotion. Phytochem Rev. 2003;2:401–17.66. Zhao Q, Nakashima J, Chen F, Yin Y, Fu C, Yun J, et al. LACCASE is necessaryand nonredundant with PEROXIDASE for lignin polymerization duringvascular development in Arabidopsis. Plant Cell. 2013;25:3976–87.67. Huis R, Morreel K, Fliniaux O, Lucau-Danila A, Fenart S, Grec S, et al. Naturalhypolignification is associated with extensive oligolignol accumulation inflax stems. Plant Physiol. 2012;158:1893–915.68. Chantreau M, Portelette A, Dauwe R, Kiyoto S, Crônier D, Morreel K, et al.Ectopic lignification in the flax lignified bast fiber1 mutant stem isassociated with tissue-specific modifications in gene expression and cellwall composition. Plant Cell. 2014;26:4462–82.69. Zhao Y, Hasenstein KH. Primary root growth regulation: the role of auxinand ethylene antagonists. J Plant Growth Regul. 2009;28:309–20.70. Mukhtar MS, Deslandes L, Auriac M-C, Marco Y, Somssich IE. The Arabidopsistranscription factor WRKY27 influences wilt disease symptom developmentcaused by Ralstonia solanacearum. Plant J. 2008;56:935–47.71. Hori Y, Fujimoto R, Sato Y, Nishio T. A novel wx mutation caused byinsertion of a retrotransposon-like sequence in a glutinous cultivar of rice(Oryza sativa). Theor Appl Genet. 2007;115:217–24.72. Yao J-L, Dong Y-H, Morris BAM. Parthenocarpic apple fruit productionconferred by transposon insertion mutations in a MADS-box transcriptionfactor. Proc Natl Acad Sci U S A. 2001;98:1306–11.73. Varagona MJ, Purugganan M, Wessler SR. Alternative splicing induced byinsertion of retrotransposons into the maize waxy gene. Plant Cell. 1992;4:811–20.74. Kobayashi S, Goto-Yamamoto N, Hirochika H. Retrotransposon-inducedmutations in grape skin color. Science. 2004;304:982.75. Chu C-G, Tan CT, Yu G-T, Zhong S, Xu SS, Yan L. A novel retrotransposoninserted in the dominant Vrn-B1 allele confers spring growth habit intetraploid wheat (Triticum turgidum L.). G3. 2011;1:637–45.76. Jiang N, Bao Z, Zhang X, Eddy SR, Wessler SR. Pack-MULE transposableelements mediate gene evolution in plants. Nature. 2004;431:569–73.77. Jiang S-Y, Ramachandran S. Genome-wide survey and comparative analysisof LTR retrotransposons and their captured genes in rice and sorghum.PLoS One. 2013;8:e71118.78. Kashkush K, Khasdan V. Large-scale survey of cytosine methylation ofretrotransposons and the impact of readout transcription from longterminal repeats on expression of adjacent rice genes. Genetics. 2007;177:1975–85.79. Wang Q, Dooner HK. Remarkable variation in maize genome structureinferred from haplotype diversity at the bz locus. Proc Natl Acad Sci U S A.2006;103:17644–9.80. Novak P, Neumann P, Pech J, Steinhaisl J, Macas J. RepeatExplorer: a Galaxy-based web server for genome-wide characterization of eukaryotic repetitiveelements from next-generation sequence reads. Bioinformatics. 2013;29:792–3.81. RepeatExplorer [Internet]. Available from: http://www.repeatexplorer.org/.Accessed 13 Nov 201382. Huang Y, Niu B, Gao Y, Fu L, Li W. CD-HIT Suite: a web server for clusteringand comparing biological sequences. Bioinformatics. 2010;26:680–2.83. CD-HIT [Internet]. Available from: http://weizhongli-lab.org/cdhit_suite/cgi-bin/index.cgi. Accessed 13 Nov 201384. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivityof progressive multiple sequence alignment through sequence weighting,position-specific gap penalties and weight matrix choice. Nucleic Acids Res.1994;22:4673–80.85. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: molecularevolutionary genetics analysis version 6.0. Mol Biol Evol. 2013;30:2725–9.86. Kimura M. A simple method for estimating evolutionary rates of basesubstitutions through comparative studies of nucleotide sequences. J MolEvol. 1980;16:111–20.87. Marchler-Bauer A, Derbyshire MK, Gonzalez NR, Lu S, Chitaz F, Geer JY, et al.CDD: NCBI’s conserved domain database. Nucleic Acids Res. 2015;43:D222–6.88. CDD: NCBI’s conserved domain database [Internet]. Available from: http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi. Accessed 26 Jan 201689. Primer3web [Internet]. Available from: http://bioinfo.ut.ee/primer3/.Accessed 10 Dec 201390. Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, et al.Primer3–new capabilities and interfaces. Nucleic Acids Res. 2012;40:e115.91. Huis R, Hawkins S, Neutelings G. Selection of reference genes forquantitative gene expression normalization in flax (Linum usitatissimum L.).BMC Plant Biol. 2010;10:71.92. SciencePrimer.com [Internet]. Available from: http://scienceprimer.com/copy-number-calculator-for-realtime-pcr. Accessed 5 Aug 201593. GelAnalyzer.com [Internet]. Available from: http://www.gelanalyzer.com/.Accessed 21 Aug 201494. Phytozome [Internet]. Available from: http://phytozome.jgi.doe.gov/pz/portal.html. Accessed 13 June 201595. Goodstein DM, Shu S, Howson R, Neupane R, Hayes RD, Fazo J, et al.Phytozome: a comparative platform for green plant genomics. NucleicAcids Res. 2012;40:1178–86.96. Gene Ontology at TAIR [Internet]. Available from: https://www.arabidopsis.org/tools/bulk/go/index.jsp. Accessed 29 Feb 201697. AGRIGO [Internet]. Available from: http://bioinfo.cau.edu.cn/agriGO/.Accessed 1 Mar 201698. Du Z, Zhou X, Ling Y, Zhang Z, Su Z. agriGO: a GO analysis toolkit for theagricultural community. Nucleic Acids Res. 2010;38:W64–70.99. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP. Determination of stablehousekeeping genes, differentially regulated target genes and sampleintegrity: BestKeeper–Excel-based tool using pair-wise correlations.Biotechnol Lett. 2004;26:509–15.100. Vandesompele J, De Preter K, Poppe B, Van Roy N, De Paepe A. Accuratenormalization of real-time quantitative RT -PCR data by geometricaveraging of multiple internal control genes. Genome Biol. 2002;3:1–12.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Galindo-González et al. BMC Genomics  (2016) 17:1002 Page 18 of 18

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0340319/manifest

Comment

Related Items