``arabidopsis ko strategies 11

ReviewTransposons as tools for functional genomicsSrinivasan Ramachandran, Venkatesan Sundaresan*

Institute of Molecular Agrobiology, National University of Singapore, 1, Research Link, Singapore 117 604, Singapore

Received 19 December 2000; accepted 25 January 2001

Abstract Transposons have been used extensively for insertional mutagenesis in several plant species. These include specieswhere highly active endogenous systems are available such as maize and Antirrhinum majus, as well as species whereheterologous transposons have been introduced through transformation, such as Arabidopsis thaliana and tomato. Much of thepast use of transposons has been in traditional forward genetics approaches, to isolate and molecularly characterize genesidentified by mutant phenotypes. With the rapid progress in the genome projects of different plants, large-scale transposonmutagenesis has become an important component of functional genomics, permitting assignment of functions to sequencedgenes through reverse genetics. Different strategies can be pursued, depending upon the properties of the transposon such as themechanism and control of transposition, and those of the host plant such as transformation efficiency. The successful use of thesestrategies in A. thaliana has made it possible to develop databases for reverse genetics, where screening for the knockout of agene of interest can be performed by computer searches. The extension of these technologies to other plants, particularlyagronomically important crops such as rice, is now feasible. 2001 ditions scientifiques et mdicales Elsevier SAS

gene tagging / reverse genetics / transposons

Ac, Activator element / Ds, Dissociation element / dSpm, defective Suppressor-mutator element / En, Enhancer / FST,flanking sequence tag / I, Inhibitor / Mu, Mutator / Spm, Suppressor-mutator

1. INTRODUCTION

The advent of large-scale sequencing projects inseveral plants, together with the anticipated comple-tion of whole genome sequences for Arabidopsisthaliana and possibly rice as well, has resulted in anexplosion of gene sequence information in plants. Asmore and more sequences are available in the data-bases, it becomes critical to assign functions to thethousands of new genes identified. There are severalapproaches being attempted to tackle this problem.

The most direct approach to determine the functionsof the sequenced genes in an organism is to disrupt orgenerate mutations in the genes and analyse theconsequences. Methods that have been developed inplants for this purpose include gene replacement, senseand anti-sense suppression, and insertional mutagen-esis. Although recently it has been shown that targetedgene replacement by homologous recombination ispossible in A. thaliana, the frequency is very low [39],which makes the method laborious for generating largenumbers of gene knockouts. Recently, a nuclear gene

of Nicotiana tobaccum has been mutated by site-specific base substitution using self-complementarychimeric oligonucleotides [9] but this method is alsotoo labour intensive for use in functional genomics.Strategies for sense and anti-sense suppression havebeen developed for the inactivation of known genes [8,37] but as these strategies require generation of severalindependent transgenic lines for each gene they arecurrently limited to the study of single genes. Cur-rently the most widely used approach for large-scalegene function analysis in plants is random insertionalmutagenesis. Either T-DNA (reviewed in [2, 42]) ortransposons (reviewed in [50, 77, 84]) can be used asinsertional mutagens in plants.

Insertional mutagenesis using Agrobacterium medi-ated T-DNA integration into plant genomes (primarilyin A. thaliana, and more recently in rice) has proved tobe very successful [2, 34, 42]. This approach has theadvantage of simplicity as each transformant yields astable insertion in the genome and does not needadditional steps to stabilize the insert. Several groupshave used this approach in A. thaliana to generate tensof thousands of independent lines that can be used forreverse genetics. Recently, a National A. thalianaKnockout Facility has been established at the Univer-

*Correspondence and reprints: fax +65 872 7007.E-mail address: [email protected] (V. Sundaresan).

Plant Physiol. Biochem. 39 (2001) 243252 2001 ditions scientifiques et mdicales Elsevier SAS. All rights reservedS0981942801012438/REV

sity of Wisconsin, USA, with access to 60 480 inser-tion lines [42]. Modified T-DNA insertions have beenused in A. thaliana as gene [3], promoter traps [46]and in activation tagging [85]. Recently, Jeon et al.[34] have also been using T-DNA insertions forfunctional genomics in rice. Despite the extremelysuccessful use of T-DNA in A. thaliana, there arehowever a few disadvantages to this approach. Theintegration of the T-DNA is generally complex, result-ing in tandem direct and inverted repeats and deletionsin one or more borders. Such rearrangements can makethe subsequent molecular analysis difficult in manycases, and adversely affect the success of large-scalestrategies such as flanking sequence databases(described later). Secondly, complex and multipleinsertions are more likely to lead to artefactual patternsof reporter gene expression when using entrapmentvectors such as gene and enhancer traps. Finally, whilethe T-DNA approach is extremely useful for plantspecies where quick and efficient transformation meth-ods are available, it may not be feasible in those plantspecies where the transformation methods are slow orlabour intensive.

For these reasons, insertional mutagenesis usingtransposable elements offers some advantages overT-DNA mutagenesis. The insertions generated by trans-posons are generally single intact elements, which lendthemselves easily to molecular analysis. Such inser-tions are also less likely to result in artefactual patternsof expression if the transposon is being used as a genetrap or enhancer trap. An additional advantage is thatmany transposons can be excised from the disruptedgene in the presence of transposase. Such excisionscan result in phenotypic reversion to the wild type orgive rise to alleles with weaker phenotypes. Thisproperty of many transposons provides ready confir-mation that the mutation was really tagged by thetransposon, as well as the possibility of generating anallelic series. In addition, another property of severaltransposons to preferentially insert into geneticallylinked sites [7, 33, 36], can be used to perform localmutagenesis in a particular region of interest byre-mobilizing the transposon [31, 35, 69]. Finally,transposons can be used for insertional mutagenesis inplant species where transformation is inefficient, sincethe generation of new insertions occurs through cross-ing or propagation rather than through transformation.

In this review, we discuss the most widely usedtransposons and how these elements have been exploitedsuccessfully in heterologous plant species as an inser-tional mutagen and as a tool for reverse genetics forfunctional genomics.

2. ENDOGENOUS TRANSPOSABLEELEMENTS IN DIFFERENT PLANT SPECIES

Transposable elements like Activator (Ac),Suppressor-mutator/Enhancer (Spm/En) and Mutator(Mu) were originally discovered and molecularly char-acterized in maize (table I [15, 16, 52, 53, 63, 64, 67]).Subsequently, a number of endogenous transposableelements in other plant species that are similar ordistantly related to the maize Ac element have beenidentified including Tam3 in Antirrhinum majus [14,28], Tag1 in A. thaliana ecotype Landsberg [79] andslide1 in tobacco [25]. Also Spm/En-like transposableelements have been reported in A. majus (Tam1 [58]),rice (Tnr3 [56]) and in Petunia (Psl [72]).

Endogenous transposable elements have been usedto clone genes in maize, A. majus and Petunia hybrida.In maize, all three families (Ac, Spm and Mu) oftransposable elements have been widely used as tagsto isolate genes (reviewed in [24, 83]). The Tamelements from A. majus (reviewed in [24]) and thedTph elements from P. hybrida [73] have provedsimilarly valuable in their host species. Some familiesof endogenous elements are present in high-copynumbers in their hosts. For example elements of theMu family in maize and dTph family in P. hybridaexist in more than 100 copies per genome. This is veryadvantageous for large-scale mutagenesis, as a fewthousand lines will be sufficient to cover the wholegenome. On the other hand, there are a few disadvan-tages in using high-copy number endogenous trans-posons. First, since there is a number of insertions perline, there will be a continuous transposition of ele-ments. If there was a significant frequency of germinalexcisions, this would result in mutations that are due tothe excision footprints, and therefore not tagged.

Table I. Endogenous transposons in different plants.

Element Plant Similar to Reference

Activator, Ac Maize [15, 16, 52]Tam3 Antirrhinum

majusAc of maize [14, 28]

Tph1 Petunia hybrida Ac of maize [23]Tag1 Arabidopsis

thalianaDistantly relatedto Ac of maize

[19, 79]

Slide Tobacco Ac of maize [25]Spm/En Maize [53, 63, 64, 65]Tam1 A. majus Spm/En [58]Tnr3 Rice Spm/En [56]Psl P. hybrida Spm/En [72]Mutator Maize [67]

244 S. Ramachandran, V. Sundaresan / Plant Physiol. Biochem. 39 (2001) 243252

Secondly, confirmation that a mutation is indeed causedby a particular copy of the transposon may requireseveral out-crosses to remove other elements, which isa time consuming process. Finally, the use of gene/enhancer trap elements is not feasible, as the patternsof expression due to several insertions of the elementsper plant will be additive and therefore inaccurate.

One of the most successfully used high-copy num-ber transposon families is the Mu transposable elementfamily consisting of the autonomous element MuDRand the non-autonomous elements Mu1 to Mu8(reviewed in [83]). These elements have been used toclone several genes in maize [10]. The Mu elementshave been particularly effective tool for gene taggingfor the following reasons: (a) higher forward mutationfrequency when compared to Ac or Spm element (20 to50 % higher) (reviewed in [83, 84]); (b) equal trans-position to linked and unlinked sites when comparedto other transposable elements [47]; (c) characteristi-cally late excisions which stabilizes new alleles insibling progeny and reduces the probability of foot-print mutations. However, transfer of the Mu systeminto heterologous plant species has not yet beensuccessfully accomplished.

Retrotransposons, which transpose through a RNAintermediate, have been shown to generate spontane-ous mutations in maize [82] and are the major class oftransposable elements in most plants (reviewed in[11]). Retrotransposons are potentially very useful forgene tagging as these elements generate stable inser-tions, and unlike many DNA transposons they inte-grate into unlinked sites. However, the relatively lowfrequencies of transpositions of most plant retrotrans-posons have restricted their uses for large-scale genetagging. Recently however, the Tos17 retrotransposonhas been developed as a promising system for rice[29]. The endogenous rice Tos17 retrotransposonsappear to be inactive during normal growth conditions,but they are reactivated by tissue culture, resulting inhigh transposition frequencies suitable for insertionalmutagenesis [29].

3. TRANSPOSONSIN HETEROLOGOUS SYSTEMS

The first report that maize transposable elementAc/Ds can be active in a heterologous system camefrom studies with transgenic tobacco (Nicotiana tobac-cum) by Baker et al. [4, 5]. Since then, the maizeAc/Ds elements have been shown to transpose activelyand been exploited in tagging studies in a number of

heterologous species including A. thaliana, rice, tomato,P. hybrida, flax, carrot, lettuce, potato, etc. (table II).Similarly maize Spm/dSpm (En/I) has been utilizedsuccessfully in tobacco, potato and A. thaliana (tableII). Apart from maize transposable elements, the A.thaliana transposon Tag1 which is distantly related tothe maize Ac, has been shown to be active in tobacco[19] and in rice [48].

In order to better regulate the transpositional eventsand obtain stable insertions, two-component systemsare preferred. In a typical two-component system, atransgenic line will be generated which harbours animmobilized (wings clipped) autonomous element(Ac or Spm/En) that will provide the transposasesource. Second line transgenic for the non-autonomouselement (Ds or dSpm/I), which cannot transpose unlessthe transposase source is available, will be generated.Selectable markers such as antibiotic resistance orherbicide resistance markers can be engineered in thenon-autonomous elements to select for the presence oftransposed elements. In order to monitor the transpo-sition events, the non-autonomous transposon can beinserted between a promoter and the marker gene sothat excision results in expression of the selectablemarker gene (reviewed in [76]). Plants homozygousfor transposase and for the dependent element are usedas starter lines and crossed to facilitate transposition.Lines carrying stable single insertions can be obtainedsubsequently by segregation of the transposase source,which can be simplified by the use of negativeselection markers [76] (figure 1B).

The ability of Spm/En elements to amplify throughcontinued propagation in A. thaliana has been exploitedto increase the number of insertions per plant [74, 88].

Table II. Transposons activity in heterologous plants.

Plant of origin Transposontype

Activity inheterologousplants

Reference

Maize Ac/Ds A. thaliana [7, 81]Rice [32, 57, 70]Tomato [35, 54, 90]Petunia [13, 26, 66]Flax [17, 44]Tobacco [5, 18, 86]Carrot [81]Lettuce [89]Potato [40]

Maize Spm/En A. thaliana [1, 12]Potato [21]Tobacco [51, 62]

Arabidopsis Tag1 Rice [48]

S. Ramachandran, V. Sundaresan / Plant Physiol. Biochem. 39 (2001) 243252 245

In one strategy, the autonomous Spm/En element wastransformed into A. thaliana and propagated for fivegenerations by a single seed descent. This yielded15 000 insertions in 3 000 lines [88]. Alternatively, atwo component (in-cis) En-I (Spm/dSpm) system hasbeen introduced into A. thaliana, where a mobile I(dSpm) element and an immobilized trans-active En(Spm) transposase are contained in the same T-DNA.This in-cis element system was used to generate2 592 A. thaliana lines containing multiple insertions(approx. twenty inserts/line) of mobile I element [74].This approach of amplifying the number of insertionsper plant has the very significant advantage that it

permits near-saturation mutagenesis of a plant genomewith a relatively small number of plants. However,there are also some disadvantages in using thisapproach, arising from the continuous nature of thetransposition events. First, the PCR strategies to iden-tify a gene knockout are complicated by somaticinsertion events that will result in the detection ofinsertions which are not transmitted through the germline. Second, the presence of footprints in genes due toimprecise excisions will lead to mutations that are nottagged.

Most transposable elements, including Ac/Ds andEn/Spm, have a tendency to preferentially transpose togenetically linked sites [7, 33, 36]. This feature can beadvantageous for directed tagging of a specific targetgene or for performing local (regional) insertionalmutagenesis in a selected region of a chromosomewhen a transposable element is inserted close to thetarget gene or within the derived chromosomal region(figure 1A) [31, 35, 69]. In another instance, Ac/Dstransposons and cDNA scanning methods were usedtogether to perform regional insertional mutagenesison genes from CIC7E11/8B11 and 5CIC5F11/CIC2B9loci on A. thaliana chromosome V [31, 69]. Thisallows cloning of cDNAs from a small region in thegenome. The flanking sequences of insertions showedthat 1420 % of the transpositions were located inabout 1 Mb of genomic DNA surrounding Ds donorsites.

When random saturation mutagenesis is desired, thehigh fraction of closely linked site transpositions posesa serious limitation. In principle, this limitation can beovercome by using many starter lines, with Ds/dSpminsertions distributed evenly over different chromo-somes. Alternately, in the Ac-Ds system of Sundaresanet al. [77], the propensity of Ds elements to transposeto linked sites is overcome by selection against thedonor site using a negative selection marker. Bysimultaneous selection against the transposase sourceand the donor site, a population of stable, unlinked,transposon insertion lines can be generated (figure 1B)[61, 77]. In the modified Spm/En system designed byTissier et al. [78], the Spm transposase and the mobiledSpm elements are contained within the same T-DNAtransformed into A. thaliana. This system eliminatesthe need for crossing, and also reduces the number ofprogeny required for the selection by a factor of four,as the negative selection is applied against only asingle locus as opposed to two loci when the trans-posase is introduced separately. However, the mainte-nance of starter lines becomes problematic, as thedSpm elements will continually transpose in the pres-

Figure 1. Schematic diagram of different approaches to transposontagging. In both A and B, the Ds element is mobilized in the presenceof Ac transposase. A, Directed tagging; B, random tagging. 1, 2 and 3represent different chromosomes, and X, Y, Z different genes. C,Insertion of gene trap or enhancer trap element. SA, splice acceptor;TA, minimal promoter; E, enhancer sequence in the genome; X,unknown gene and GUS, glucuronidase. D, Insertion of activationtagging element. X denotes an unknown gene. TE, transposableelement; LB, left border of T-DNA; RB, right border of T-DNA; PSM,positive selection marker; NSM, negative selection marker.


ence of the Spm transposase, so that this systemrequires a relatively large number of independentstarter lines obtained through transformation.

4. SPECIALIZED TAGGING SYSTEMS

4.1. Gene and enhancer trap elementsInsertional mutagenesis by transposon tagging is

useful when disruption of a gene leads to an obviousphenotype. But in eukaryotic systems, disruptions ofgenes frequently do not result in visible phenotypesdue to functional redundancy, or they may result inearly lethality that obscures late-acting functions whenthe same gene has multiple functions in development.To overcome these difficulties, modified transposonscalled gene trap or enhancer trap transposable ele-ments have been developed (reviewed in [43, 75, 76]).Such modified elements were first used in Drosophilamelanogaster and mouse [71, 87], and were subse-quently extended to plant transposon systems [7577].

Gene/enhancer trap elements carry a reporter genewhose expression pattern reflects the activity andregulation of the disrupted genes (figure 1C). Theenhancer traps contain a reporter gene driven by aminimal or weak promoter in the dependent element.The reporter gene expression is achieved (by utilizingthe endogenous enhancer sequences) only when thetransposition occurs within or close to a gene (figure1C). A variation of this vector called a promoter trapcontains a promoterless reporter gene, which willexpress only when the transposon is inserted down-stream of an active endogenous plant promoter. Incontrast, gene trap vectors are designed to contain oneor more splice acceptor sequences preceding the reportergene. This allows expression of the reporter only whenit inserts in a transcribed region. The splice donor sitesfrom the intron of the endogenous gene and the spliceacceptor sequences, which precede the reporter gene,will be spliced to generate a fusion transcript (reviewedin [50, 75, 76]). Thus far only the Ac/Ds transposonsystem has been modified for enhancer and gene trapapproaches.

4.2. Activation tagging elementsGene disruptions through insertional mutagenesis

nearly always generate recessive loss-of-function muta-tions. Such mutations do not always produce anobvious phenotype due to various factors such asfunctional redundancy (see next section). In suchcases, increasing the expression level or ectopically

expressing a gene can provide dominant gain-of-function mutations that produce informative mutantphenotypes. A strategy to generate mutants as a con-sequence of increased or expanded expression of atagged gene is known as activation tagging. Theprinciple involves using strong enhancers in T-DNA ora transposon resulting in ectopic expression or over-expression of nearby genes through transcriptionalactivation. The first successful activation tagging inplants was reported by Hayashi et al. [27], whointroduced four copies of enhancer elements from thecauliflower mosaic virus 35S promoter (CaMV 35S)into T-DNA, generating gain-of-function mutationsand novel phenotypes. This approach has been used togenerate a large-scale tagging population by Wiegel etal. [85]. The principle of activation tagging has alsobeen adapted for transposable elements by Wilson etal. [87]. They used the Ds element with a completeCaMV 35S promoter pointing outwards from theelement (figure 1D), and generated a population ofinsertions through crosses with Ac transposase. Thisapproach was used to identify dominant mutations atvarious loci in A. thaliana, including TINY, LateElongated Hypocotyl (LHY), and Short Internodes(SHI) [22, 68, 87]. A limitation of this approach is thatthe observed frequency of dominant mutations throughactivation tagging has been significantly lower thanthat of recessive mutations arising from insertionalinactivation [85], suggesting that many genes may beover-expressed without resulting in observable pheno-types. Nevertheless, activation tagging has proven tobe a valuable complementary approach for the identi-fication of gene functions.

5. FORWARD VS. REVERSE GENETICS

Cloning of genes that produce mutant phenotype orfunction forms the basis of forward genetics (i.e.phenotype or function to gene). On the other hand, ifthe gene sequence is known and the biological func-tion of that gene is not known, a knockout mutant canbe generated and analysed to determine its function.This forms the basis for reverse genetics, i.e. fromgene to phenotype or function (reviewed in [42, 50,60]). Since genome sequencing projects for variousplant species are progressing rapidly, more and moresequences encoding predicted genes are available inthe databases. Reverse genetics strategies will be ofgreat importance for the purpose of assigning func-tions to predicted genes. As discussed earlier, genedisruption by transposons constitutes a powerful toolfor reverse genetics.


6. STRATEGIESFOR REVERSE GENETICS SCREENING

In order to perform reverse genetic screens effi-ciently, it is necessary to generate a large population oftransposon tagged mutants. The number of lines to bescreened is dependent on the genome size and thenumber of genes of a given plant species, the type oftransposon used (single or multi-copy) etc. A popula-tion containing insertions with a reasonable probabil-ity of finding at least one insertion in any given geneis important for the success of this approach.

In order to screen for an insertion in a particulargene, a PCR-based strategy has been applied in D.melanogaster for site-specific selection of insertionsusing P elements [6, 38]. In this approach, a gene-specific primer and an insertion-specific primer areused for PCR amplification. Several insertion lines arepooled together and the DNA extracted is then used asa template for PCR reactions. Samples of 20 to100 insertion lines are pooled to extract genomicDNA, and a gene-specific primer and an insertion-specific primer were used for PCR [55, 78]. Any poolshowing a positive signal is re-screened using DNAfrom individual lines, to identify the line carrying theinsertion of interest. Several pools can be combined toform a super pool if the number of insertion lines inthe population is very high [78]. Alternatively athree-dimensional matrix pooling strategy has beentested for P. hybrida lines carrying the multi-copydTph1 transposon [41]. The leaf material from apopulation of 1 000 plants were pooled according to athree-dimensional matrix (columns, rows and blocks).DNA samples were extracted from ten blocks of100 lines. Similarly DNA is extracted from ten col-umns and ten rows each of them containing 100 lines.Altogether thirty DNA samples were used to performPCR, which can identify a single plant that containsthe insertion in the specific gene. This method isadvantageous as it requires fewer amplifications, anddirectly identifies the single plant with insertion (figure2 [41]).

An alternate approach to identify the genes thathave been tagged in a population of insertion linesinvolves random amplification of the DNA flankingthe insertions. Several different methods can be fol-lowed within this approach. Transposon display [80]and amplification of insertion mutagenized sites (AIMS)[20] are techniques used to identify tagged genes infamilies with multi-copy transposons like dTph1 inPetunia and Mu in maize. For example, the maize Bx1gene and the Fbp1 gene fragment from Petunia wereisolated by this method [20, 80]. The transposon

display and AIMS techniques utilize adapters that areligated to the genomic DNA, which has been digestedwith restriction enzyme(s). The appropriate restrictionenzyme(s) are selected which recognizes one site inthe insertion element and the other in the flankingregion. A chimeric adapter and transposon primer isused from one end and an adapter primer from theother end to amplify the flanking region in the trans-poson display technique. On the other hand, adapterprimers are used as forward and reverse primers foramplification in AIMS approach. Flanking regionsobtained by amplifications are displayed which can becloned and sequenced subsequently ([20, 80] andreviewed in [50]).

A method called inverse display of insertions (IDI)has been developed by Tissier et al. [78] to isolate theflanking sequences spanning the dSpm/I element. Theinverse PCR (iPCR) method [59] has been adapted for

Figure 2. Flow chart for the use of insertional mutants in a reversegenetics approach to functional genomics. The steps involved from theidentification of an insertion in a gene of interest to the phenotypiccharacterization of the mutant are detailed.


this strategy. Since the selection of restriction sites (nottoo close or too far from the insertion site) is a criticalfactor for a successful iPCR, three different restrictionenzymes were used to greatly increase the probabilityof amplifying the flanking region. DNA pools frominsertion lines were used as template and iPCR reac-tions were performed. These products were mixed andspotted in grid arrays on a nylon membrane, andhybridized with labelled probes from the genes ofinterest [78].

Alternatively, the DNA flanking the insertions canbe amplified and sequenced individually to cataloguethe insertions by chromosomal location [61, 78]. Thisstrategy is especially useful when substantial genomesequence information is available, as in the case of A.thaliana. A protocol such as iPCR, ligation-mediatedPCR or thermal asymmetric interlaced PCR (TAIL-PCR; [49]) can be used to amplify and sequence theflanking region of single- or low-copy insertion lines.With an appropriately constructed database of suchflanking sequences, it will be feasible to identifyinsertions in a particular gene by simple computersearches, removing the necessity for the more tediouspooling and hybridization protocols. In addition, evenif no hit is found, the availability of a transposoninsertion close to a gene of interest can be quicklyascertained by a computer search. As described previ-ously, transposons from the Ac/Ds and Spm/En fami-lies are very useful for mutagenesis of closely linkedgenes, and a collection of sequenced insertions pro-vides launch pads for tagging most of the genes withinthe genome.

7. GENE SEQUENCETO FUNCTIONAL ANALYSIS

Once a knockout in a desired gene is identifiedthrough reverse genetics strategies among the popula-tion of insertional mutants, it becomes necessary tofunctionally characterize the mutant (figure 2). Thefirst step in the characterization process is to obtain amutant that has a single insertion in the gene ofinterest. If a two-component transposable elementsystem (Ac/Ds for example) has been used in generat-ing a population of insertion mutants, then obtainingsingle insertion line is rather straight forward. On theother hand, if multi-copy transposable elements areused to perform the reverse genetics screen, then it isnecessary to do several out-crosses to make sure thatonly a single insertion is in the gene. The next step forcharacterization is the identification of phenotypescaused by gene knockout. If there is an observable

phenotype caused by the single stable insertion, thenthe gene functions may be deduced through detailedanalysis of the phenotype (figure 2). However, thereare many instances, where a gene knockout does notshow an observable phenotype. This could be due tothe fact that several genes may be required andexpressed only under specific conditions, such aspathogen infection or environmental stresses. For muta-tions in such genes, in order to detect a phenotype itwill be necessary to subject the plants to conditions inwhich the gene is required. These conditions mayinclude challenges with pathogens and the use ofspecific growth media or growth conditions. Forexample, a T-DNA insertion in the AKT-1 potassiumchannel gene of A. thaliana was identified by a reversegenetics approach using pooled PCR and hybridiza-tion. On most nutritional media, the akt-1 mutantplants were indistinguishable from wild type plants,but on media containing a low potassium concentra-tion of 100 M or less, the growth of the mutant plantswas defective [30].

Another reason for not observing a clear phenotypewhen a gene is knocked out is functional redundancywith other genes. In such cases, creation of double ortriple mutants of the functionally redundant genes willuncover the phenotype and permit characterization oftheir functions (figure 2). For example, A. thalianagenes SHATTERPROOF1 (SHP1) and SHATTER-PROOF2 (SHP2) regulate fruit dehiscence or podshatter. Both genes are functionally redundant, asneither single mutant produces a novel phenotype.However, shp1 and shp2 double mutants produce fruitsor pods that do not shatter [45]. Since the A. thalianagenome-sequencing project is almost complete and thesequences are available, it is feasible to identify all theclosely related members of a gene family in thisspecies. Together with computer searches of flankingsequence databases, it will soon be possible to con-struct combinations of mutants to reveal novel func-tions that have gone undetected using the currentgenetics methodologies.

8. CONCLUSION

In the post-genome era, sequences of many plantgenomes will be available and functional assessmentof the genes identified will depend on many approachesincluding insertional mutagenesis, polymorphismanalysis, expression microarrays, and bioinformatics.A large population of insertional mutants generated bytransposon mutagenesis can be used to dissect out


gene functions in combination with the sequencesobtained from the EST and genome sequencing projectsin many plant species, through various strategies suchas pooled PCR screening and the construction offlanking sequence databases. Recent advances demon-strate the feasibility of using transposon-based func-tional genomics approaches for the study of anyspecies of flowering plant.

REFERENCES

[1] Aarts M.G., Dirkse W.G., Stiekema W.J., Periera A.,Transposon tagging of a male sterility gene in A.thaliana, Nature 363 (1993) 715717.

[2] Azpiroz-Leehan R., Feldmann K.A., T-DNA insertionmutagenesis in A. thaliana: going back and forth,Trends Genet. 13 (1997) 146152.

[3] Babiychuk E., Fuangthong M., Van Montagu M.,Inze D., Kushnir S., Efficient gene tagging in Arabi-dopsis thaliana using a gene trap approach, Proc. Natl.Acad. Sci. USA 94 (1997) 1272212727.

[4] Baker B., Coupland G., Federoff N.V., Starlinger P.,Schell J., Phenotypic assay for excision of the maizecontrolling element Ac in tobacco, EMBO J. 6 (1987)15471554.

[5] Baker B., Schell J., Lorz H., Federoff N.V., Transpo-sition of the maize controlling element Activator intobacco, Proc. Natl. Acad. Sci. USA 83 (1986)48444848.

[6] Ballinger D.G., Benzer S., Targeted gene mutation inDrosophila, Proc. Natl. Acad. Sci. USA 86 (1989)94029406.

[7] Bancroft I., Dean C., Transposition pattern of themaize element Ds in Arabidopsis thaliana, Genetics134 (1993) 12211229.

[8] Baulcombe D.C., RNA as a target and an initiator ofpost-transcriptional gene silencing in transgenic plants,Plant Mol. Biol. 32 (1996) 7982.

[9] Beetham P.R., Kipp P.B., Sawycky X.L., Arntzen C.J.,May G.D., A tool for functional plant genomics;Chimeric RNA/DNA oligonucleotides cause in vivogene-specific mutations, Proc. Natl. Acad. Sci. USA 96(1999) 87748778.

[10] Bennetzen J.L., The Mutator transposable elementsystem of maize, Curr. Top. Microbiol. Immunol. 204(1996) 195229.

[11] Bennetzen J.L., The contributions of retroelements toplant genome organization, function and evolution,Trends Microbiol. 4 (1996) 347353.

[12] Cardon G.H., Frey M., Saedler H., Gierl A., Mobilityof the maize transposable element En/Spm in Arabi-dopsis thaliana, Plant J. 3 (1993) 773784.

[13] Chuck G., Robbins T., Nijjar C., Raltson E., Courtney-Gutterson N., Dooner H., Tagging and cloning of aPetunia flower color gene with the maize transposableelement Activator, Plant Cell 5 (1993) 371378.

[14] Coen E.D., Carpenter R., Martin C., Transposable

element generates novel spatial patterns of gene expres-sion in Antirrhinum majus, Cell 24 (1986) 285296.

[15] Federoff N.V., Furtek D.B., Nelson O.E. Jr, Cloning ofthe bronze locus in maize by a simple and generaliz-able procedure using the transposable element Activa-tor (Ac), Proc. Natl. Acad. Sci. USA 81 (1984)38253829.

[16] Federoff N.V., Wessler S., Shure M., Isolation of thetransposable maize controlling elements Ac and Ds,Cell 35 (1983) 235242.

[17] Finnegan E.J., Lawrence G.J., Dennis E.S., Ellis J.G.,Behavior of modified Ac elements in flax callus andregenerated plants, Plant Mol. Biol. 22 (1993) 625633.

[18] Fitzmaurice W.P., Nguyen L.V., Wernsman E.A.,Thompson W.F., Conkling M.A., Transposon taggingof the sulfur gene of tobacco using engineered maizeAc/Ds elements, Genetics 153 (1999) 19191928.

[19] Frank M.J., Liu D., Tsay Y.F., Ustach C., CrawfordN.M., Tag1 is an autonomous transposable elementthat shows somatic excision in both Arabidopsis andtobacco, Plant Cell 9 (1997) 17451756.

[20] Frey M., Stettner C., Gierl A., A general method forgene isolation in tagging approaches: amplification ofinsertion mutagenized sites (AIMS), Plant J. 13 (1998)717721.

[21] Frey M., Tavatzis S.M., Saedler H., The maize En-I/Spm element transposes in potato, Mol. Gen. Genet.217 (1989) 172177.

[22] Fridborg I., Kuusk S., Moritz T., Sundberg E., TheArabidopsis dwarf mutant shi exhibits reduced gibber-ellin responses conferred by overexpression of a newputative zinc finger protein, Plant Cell 11 (1999)10191031.

[23] Gerats A.G., Huits H., Vrijlandt E., Marana C.,Souer E., Beld M., Molecular characterization of anon-autonomous transposable element (dTph1) of Petu-nia, Plant Cell 2 (1990) 11211128.

[24] Gierl A., Saedler H., Plant-transposable elements andgene tagging, Plant Mol. Biol. 19 (1992) 3949.

[25] Grappin P., Audeon C., Chupeau M.C., Grandbas-tien M.A., Molecular and function characterization ofSlide, an Ac-like autonomous transposable elementfrom tobacco, Mol. Gen. Genet. 252 (1996) 386397.

[26] Haring M.A., Gao J., Volbeda T., Rommens C.M.T.,Nijkamp H.J.I., Hille J., A comparative study of Tam3and Ac transposition in transgenic tobacco and Petuniaplants, Plant Mol. Biol. 13 (1989) 189201.

[27] Hayashi H., Csaja I., Lubenow H., Schell J.,Walden R., Activation of a plant gene by T-DNAtagging: auxin-independent growth in vitro, Science258 (1992) 13501353.

[28] Hehl R., Nacken W.K., Krause A., Saedler H., Som-mer H., Structural analysis of Tam3, a transposableelement from Antirrhinum majus, reveals homologiesto the Ac element from maize, Plant Mol. Biol. 16(1991) 369371.

[29] Hirochika H., Retrotransposons of rice: their regula-tion and use for genome analysis, Plant Mol. Biol. 35(1997) 231240.

[30] Hirsch R.E., Lewis B.D., Spalding E.P., Suss-


man M.R., A role for the AKT1 potassium channel inplant nutrition, Science 280 (1998) 918921.

[31] Ito T., Seki M., Hayashida N., Shibata D., Shi-nozaki K., Regional insertional mutagenesis of geneson Arabidopsis thaliana chromosome V using theAc/Ds transposon in combination with a cDNA scan-ning method, Plant J. 17 (1999) 433444.

[32] Izawa T., Ohnishi T., Nakano T., Ishida N., Enoki H.,Hashimoto H., Itoh K., Terada R., Wu C., Miyazaki C.,Endo T., Iida S., Shimamoto K., Transposon tagging inrice, Plant. Mol. Biol. 35 (1997) 219229.

[33] James D.W. Jr, Lim E., Keller J., Plooy I., Ralston E.,Dooner H.K., Direct tagging of the Arabidopsis FattyAcid Elongation 1 (FAE1) gene with the maize trans-poson Activator, Plant Cell 7 (1995) 309319.

[34] Jeon J.H., Lee S., Jung K.H., Jun S.H., Jeong D.H.,Lee J., Kim C., Jang S., Yang K., Nam J., An K.,Han M.J., Sung R.J., Choi H.S., Yu J.H., Choi J.H.,Cho S.Y., Cha S.S., Kim S.I., An G., T-DNA inser-tional mutagenesis for functional genomics in rice,Plant J. 22 (2000) 561570.

[35] Jones D.A., Thomas C.M., Hammond-Kosack K.E.,Balint-Kurti P.J., Jones J.D., Isolation of the tomatoCf-9 gene for resistance to Cladosporium fulvum bytransposon tagging, Science 266 (1994) 789793.

[36] Jones J.D.G., Carland F.C., Lim E., Raltson E.,Dooner H.K., Preferential transposition of the maizeelement Activator to linked chromosomal locations intobacco, Plant Cell 2 (1990) 701707.

[37] Jorgensen R.A., Que Q., Stam M., Do unintendedantisense transcripts contribute to sense co-suppressionin plants?, Trends Genet. 15 (1999) 1112.

[38] Kaiser K., Goodwin S.F., Site selected transposonmutagenesis of Drosophila, Proc. Natl. Acad. Sci. USA87 (1990) 16861690.

[39] Kempin S.A., Liljegren S.J., Block L.M., Roun-sley S.D., Yanofsky M.F., Lam E., Targeted disruptionin Arabidopsis, Nature 38 (1997) 802803.

[40] Knapp S., Coupland G., Uhrig H., Starlinger P.,Salamini F., Transposition of the maize transposableelement Ac in Solanum tuberosum, Mol. Gen. Genet.213 (1988) 285290.

[41] Koes R., Souer E., Van Houwelingen A., Mur L.,Spelt C., Quattrocchio F., Wing J., Oppedijk B.,Ahmed S., Maes T., Gerats T., Hoogeveen P.,Meesters M., Kloos D., Mol J.N.M., Targeted geneinactivation in petunia by PCR-based selection oftransposon insertion mutants, Proc. Natl. Acad. Sci.USA 92 (1995) 81498153.

[42] Krysan P.J., Young J.C., Sussman M.R., T-DNA as aninsertional mutagen in Arabidopsis, Plant Cell 11(1999) 22832290.

[43] Kumar C.S., Narayanan K.K., Plant transposable ele-ments and functional genomics, Plant Biotech. 15(1998) 159165.

[44] Lawrence G.J., Finnegan E.J., Ayliffe M.A., Ellis J.G.,The L6 gene for flax rust resistance is related to theArabidopsis bacterial resistance gene RPS2 and thetobacco viral resistant gene N, Plant Cell 7 (1995)11951206.

[45] Lilijegren S.J., Ditta G.S., Eshed Y., Savidge B.,Bowman J.L., Yanofski M.F., Shatterproof MADS-boxgenes control seed dispersal in Arabidopsis, Nature404 (2000) 766770.

[46] Lindsey K., Wei W., Clarke M.C., McArdle H.F.,Rooke L.M., Topping J.F., Tagging genomic sequencesthat direct transgene expression by activation of apromoter trap in plants, Transgenic Res. 2 (1993)3347.

[47] Lisch D., Chomet P., Freeling M., Genetic character-ization of Mutator system in maize: behavior andregulation of Mu transposons in a minimal line, Genet-ics 139 (1995) 17771796.

[48] Liu D., Zhang S., Fauquet C., Crawford N.M., TheArabidopsis transposon Tag1 is active in rice, under-going germinal transposition and restricted, late somaticexcision, Mol. Gen. Genet. 262 (1999) 413420.

[49] Liu Y.G., Whittier R.F., Thermal asymmetric inter-laced PCR: automatable amplification and sequencingof insert end fragments from P1 and YAC clones forchromosome walking, Genomics 25 (1995) 674681.

[50] Maes T., Keukeleire P.D., Gerats T., Plant tagnology,Trends Plant. Sci. 4 (1999) 9096.

[51] Masson P., Federoff N.V., Mobility of the maizesuppressor-mutator element in transgenic tobacco cells,Proc. Natl. Acad. Sci. USA 86 (1989) 22192223.

[52] McClintock B., Mutable loci in maize, Carnegie Inst.Wash. Year Book 46 (1947) 155169.

[53] McClintok B., Mutations in maize and chromosomalaberrations in neurospora, Carnegie Inst. Wash. YearBook 53 (1954) 254260.

[54] Meissner R., Chague V., Zhu Q., Emmanuel E.,Elkind Y., Levy A.A., Technical advance: a highthroughput system for transposon tagging and pro-moter trapping in tomato, Plant J. 22 (2000) 265274.

[55] Meissner R.C., Jin H., Corminelli E., Denekamp M.,Fuertes A., Greco R., Kranz H.D., Penfield S.,Petroni K., Urzainqui A., Function search in a largetranscription factor gene family in Arabidopsis: Assess-ing the potential of reverse genetics to identify inser-tional mutations in r2r3 myb genes, Plant Cell 11(1999) 18271840.

[56] Motohashi R., Ohtsubo E., Ohtsubo H., Identificationof Tnr3, a Suppressor-mutator/enhancer-like transpos-able element from rice, Mol. Gen. Genet. 250 (1996)148152.

[57] Murai N., Li Z.J., Kawagoe Y., Hayashimoto A.,Transposition of the maize Activator element in trans-genic rice plants, NucleicAcids Res. 19 (1991) 617622.

[58] Nacken W.K., Piotrowiak R., Saedler H., Sommer H.,The transposable element Tam1 from Antirrhinummajus shows structural homology to the maize trans-poson En/Spm and has no sequence specificity ofinsertion, Mol. Gen. Genet. 228 (1991) 201208.

[59] Ochman H., Geber A.S., Hartl D.L., Genetic applica-tions of an inverse polymerase chain reaction, Genetics120 (1988) 621623.

[60] Parinov S., Sundaresan V., Functional genomics inArabidopsis: large-scale insertional mutagenesis


complements the genome sequencing project, Curr.Opin. Biotech. 11 (2000) 157161.

[61] Parinov S., Sevugan M., Ye D., Yang W.C., Kuma-ran M., Sundaresan V., Analysis of flanking sequencesfrom Dissociation insertion lines:Adatabase for reversegenetics in Arabidopsis, Plant Cell 11 (1999)22632270.

[62] Periera A., Saedler H., Transcriptional behavior of themaize En/Spm element in transgenic tobacco, EMBOJ. 8 (1989) 13151321.

[63] Periera A., Cuypers H., Gierl A., Schwarz-Sommers Z., Saedler H., Molecular analysis of theEn/Spm transposable element system of Zea mays,EMBO J. 5 (1986) 835841.

[64] Peterson P., The En mutable system in maize, Theor.Appl. Genet. 40 (1970) 367377.

[65] Peterson P.A., A mutable pale green locus in maize,Genetics 38 (1953) 682683.

[66] Robbins T.P., Jenkin M., Courtney-Gutterson N.,Enhanced frequency of transposition of the maizetransposable element Activator following excision fromT-DNA in Petunia hybrida, Mol. Gen. Genet. 244(1994) 491500.

[67] Robertson D.S., Characterization of a mutator systemin maize, Mutat. Res. 51 (1978) 2128.

[68] Schaffer R., Ramsay N., Samach A., Corden S.,Putterill J., Carre I.A., Coupland G., The late elongatedhypocotyl mutation of Arabidopsis disrupts circadianrhythms and the photoperiodic control of flowering,Cell 93 (1998) 12191229.

[69] Seki M., Ito T., Shibata D., Shinozaki K., Regionalinsertional mutagenesis of specific genes on theCIC5F11/CIC2B9 locus of Arabidopsis thaliana chro-mosome 5 using the Ac/Ds transposon in combinationwith the cDNA scanning method, Plant Cell Physiol.40 (1999) 624639.

[70] Shimamoto K., Miyazaki C., Hashimoto H., Izawa T.,Itoh K., Terada R., Inagaki Y., Iida S., Trans-activationand stable integration of the maize transposable ele-ment Ds cotransfected with the Ac transposase gene intransgenic rice plants, Mol. Gen. Genet. 239 (1993)354360.

[71] Skarnes W.C., Entrapment vectors: a new tool formammalian genetics, Bio/Technology 8 (1990)827831.

[72] Snowden K.C., Napoli C.A., Psl: a novel Spm-liketransposable element from Petunia hybrida, Plant J. 14(1998) 4354.

[73] Souer E., Quattrocchio F., de Vetten N., Mol J.,Koes R., A general method to isolate genes tagged bya high copy number transposable element, Plant J. 7(1995) 677685.

[74] Speulman E., Metz P.L.J., van Arkel G., te LintelHekkert B., Stiekema W.J., PereiraA.,Atwo-componentenhancer-inhibitor transposon mutagenesis system forfunctional analysis of the Arabidopsis genome, PlantCell 11 (1999) 18531866.

[75] Springer P., Gene Traps: Tools for plant developmentand genomics, Plant Cell 12 (2000) 10071020.

[76] Sundaresan V., Horizontal spread of transposon

mutagenesis: new uses for old elements, Trends PlantSci. 1 (1996) 184190.

[77] Sundaresan V., Springer P., Volpe T., Harward S.,Jones J.D.G., Dean C., Martienssen R., Patterns ofgene actions in plant development revealed by enhancertrap and gene trap transposable elements, Genes Dev.9 (1995) 17971810.

[78] Tissier A.F., Marillonnet S., Kimyuk V., Patel K.,Torres M.A., Murphy G., Jones J.D.G., Multiple inde-pendent defective Suppressor-mutator transposon inser-tions in Arabidopsis: A tool for functional genomics,Plant Cell 11 (1999) 18411852.

[79] Tsay Y.F., Frank M.J., Page T., Dean C., Craw-ford N.M., Identification of a mobile endogenoustransposon in Arabidopsis thaliana, Science 260 (1993)342344.

[80] Van den Broeck D., Maes T., Sauer M., Zethof J., DeKuekeleire P., DHauw M., Van Montagu M., Ger-ats T., Transposon display identifies individual trans-posable elements in high copy number lines, Plant J.13 (1998) 121129.

[81] Van Sluys M.A., Tempe J., Federoff N., Studies on theintroduction and mobility of the maize Activator ele-ment in Arabidopsis thaliana and Daucus carota,EMBO J. 6 (1987) 38813889.

[82] Varagona M.J., Purugganan M., Wessler S.R., Alter-nate splicing induced by insertion of retrotransposonsinto maize waxy gene, Plant Cell 4 (1992) 811820.

[83] Walbot V., Strategies for mutagenesis and gene cloningusing transposon tagging and T-DNA insertionalmutagenesis, Annu. Rev. Plant Physiol. Plant Mol.Biol. 43 (1992) 4982.

[84] Walbot V., Saturation mutagenesis using maize trans-posons, Curr. Opi. Plant Biol. 3 (2000) 103107.

[85] Weigel D., Hoon Ahn J., Blazquez M.A., Bor-evitz J.O., Christensen S.K., Frankhauser C., Ferran-diz C., Karadailsky I., Malancharuvil E.J., Neff M.M.,Nguyen J.T., Sato S., Wang Z- Y., Xia Y., Dixon R.A.,Harrison M.J., Lamb C.J., Yanofsky M.F., Chory J.,Activation tagging in Arabidopsis, Plant Physiol. 122(2000) 10031013.

[86] Whitham S., Dinesh-Kumar S.P., Choi D., Hehl R.,Corr C., Baker B., The product of the tobacco mosaicvirus resistance gene N: similarity to Toll and theinterleukin-1 receptor, Cell 78 (1994) 11011115.

[87] Wilson K., Long D., Swinburne J., Coupland G., ADissociation insertion causes a semidominant muta-tion that increases expression of TINY, an Arabidopsisgene related to APETALA2, Plant Cell 8 (1996) 659671.

[88] Wisman E., Cardon G.H., Fransz P., Saedler H., Thebehaviour of the autonomous maize transposable ele-ment En/Spm in Arabidopsis thaliana allows efficientmutagenesis, Plant Mol. Biol. 37 (1998) 989999.

[89] Yang C.H., Ellis J.G., Michelmore R., Infrequenttransposition of Ac in Lettuce, Lactuca sativa, PlantMol. Biol. 22 (1993) 793805.

[90] Yoder J.I., Palys J., Alpert K., Lassner M., Ac trans-position in transgenic tomato plants, Mol. Gen. Genet.213 (1988) 291296.


``arabidopsis ko strategies 11

Documents