Court Filings

A general method to isolate genes tagged by a high copy number transposable element

A general method to isolate genes tagged by a high copy number transposable element
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  The P/ant Journal (1995) 7(4), 677-685 TECHNICAL ADVANCE A general method to isolate genes tagged by a high copy number transposable element Erik Souer, Francesca Quattrocchio, Nick de Vatten, Joseph Mol end Ronald Koes* Department of Genetics, Vrije Universiteit, Institute for Molecular Biological Sciences, BioCentrum Amsterdam, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands Summary The Petunia hybrida line W138 contains more than 200 copies of the transposable element dTphl. In W138 pro- geny these elements give rise to new unstable mutations at high frequency. With the aim of isolating these mutated genes a method was developed to isolate dTphl flanking sequences unique for mutant plants. This method is based on differential screening of cloned inverse polymerese chain reaction (IPCR) products srcinating from the mut- ated plant. It directly yields e probe for the mutated gene which can be used to screen pre-existing cDNA and genomic libraries. This method may be generally applicable to isolate genes tagged by other high copy number transposable elements, like Mutator (Mu) or Disso- ciation (Ds) in Zea mays. Introduction Introduction of mutations in plants can be achieved by a variety of techniques. An efficient way to induce mutations is by chemical treatment or X-ray irradiation. Commonly used chemicals like EMS introduce point mutations, while X-ray irradiation leads to gross chromosomal changes. Therefore, isolation of the mutated gene is not easily achieved in these cases. An alternative is to introduce mutations by insertion of a known sequence, such as a T- DNA or a transposable element. Until now, T-DNA tagging has been used in Arabidopsis thaliana mainly because of the ease with which a large number of transformants can be generated (Feldmann, 1991). However, less than 40% of the mutants obtained are tagged by the T-DNA and isolation of flanking sequences is further complicated by the complexity of the T-DNA inserts (Castle et al., 1993). Gene tagging with endogenous transposable elements has been successfully used to isolate genes from Antirrhinum Received 21 September 1994; revised 14 December 1994; accepted 22 December 1994. *For correspondence fax +31 20 4447137). majus and Zea mays (see e.g. Coen et al., 1990; Federoff et al., 1984; Goodrich et al., 1992; Martienssen et al., 1989; Martin et aL, 1985; O'Reilly et al., 1985; Theres et al., 1987; for a review see Walbot, 1992). Usually tagged genes are cloned after one copy of the transposon could be linked to the mutation on Southern blots probed with the transpos- able element (Martienssen et al., 1989; Theres et al., 1987; Walbot, 1992). Subsequently, a genomic library has to be constructed of the plant DNA fraction containing this transposon copy. Clones hybridizing to the transposable element possibly contain the mutated gene. In both Z. mays and Petunia hybrida plant lines are available that contain high copy numbers of transposable elements which give rise to many unstable mutations in selfed progeny (Doodeman et al., 1984; Gerats eta/., 1990; Robertson, 1978). The Mutator line of Z. mays contains numerous copies of Mutator transposable elements of which Mul appears to be the most active (Bennetzen, 1984; Talbert et al., 1989). Also, Ds elements can be present in high copy numbers in Z. mays lines (Geiser eta/., 1982). The high copy numbers of these transposable elements in combination with a high transposition activity increase the chances of hitting genes of interest. However, at the same time, these features create difficulties in isolation of the mutated gene, because detection of a co-segregating band on genomic Southern blots containing DNA of a population segregating for the mutation is not easily achieved. Various techniques have been used to overcome this problem. In some cases it is possible to reduce the number of hybridiz- ing copies of the transposable element by using a small part of the element as a probe (Federoff eta/., 1984; Paz-Ares et al., 1986; Wienand et aL, 1986). Methylation- sensitive restriction enzymes have been used, based on the idea that inactive elements are methylated thereby only visualizing active elements on genomic Southern blots (Chandler and Walbot, 1986). The mutation can also be outcrossed to a line that lacks transposable elements (McCarty et al., 1989) or tagged by two different transpos- able elements (O'Reilly et al., 1985). The P. hybrida inbred line W138 gives rise to numerous unstable mutations in selfed progeny (Doodeman et al., 1984). Among these mutants are plants which are disturbed in flower pigmentation, plant and flower development, fertility or chlorophyll synthesis. Until now, out of 13 unstable alleles of flower pigmentation genes that have been analysed, 12 contained an insertion of a dTphl element. The different dTphl copies were all 284 bp in size 677  678 Erik Souer et al. and their sequences differed at most in nine positions (Kroon et al., 1994; Souer, van Houwelingen, Oppedijk, Mol and Koes, unpublished results). Cloning of new genes tagged by dTph I is, however, complicated by the presence of more then 200 dTph 1 elements in this line (Gerats et al., 1990). Outcrossing of the mutation to low copy number transposable element lines is time-consuming because it takes many generations. Because all dTphl elements are highly homologous, using a small part of the element as a probe will not reduce the number of hybridizing bands on genomic Southern blots. We therefore developed an alternative method to clone genes tagged by insertion of dTphl transposable elements. Using a combination of inverse polymerase chain reaction (IPCR) and differential screening of amplification products we were able to isolate dTphl flanking sequences unique for mutant plants. The mutant gene fragment obtained by this method can be used directly to identify wild-type genomic and cDNA clones from existing libraries, thereby avoiding the neces- sity of cloning size-fractionated genomic fragments. In principle, this method will be useful for isolating genes tagged by high copy number transposable elements, like Ds and Mu in Z. Mays or dTph l in P. hybrida. Results Cloning sequences flanking a dTph 1 insertion in an unstable an3 allele Some P. hybrida lines contain more than 200 copies of dTphl transposable elements (Gerats et aL, 1990; Huits, 1993). To isolate genes interrupted by a dTphl element one needs to identify sequences containing dTphl which are unique to mutant plants. This is conventionally done by genomic Southern blot analysis of transposon-containing restriction fragments, which discriminates hose sequences on size (Walbot, 1992). We tested this on a family of plants containing an unstable allele of the An3 locus (an3-$205). The An3 locus encodes the enzyme flavanone 3~- hydroxylase (F3H) and in an3-$205 the second exon of this gene is interrupted by a dTphl element (Souer, van Houwelingen, Oppedijk, Mol and Koes, unpublished results; see Figure 3a). For the wild-type we used two F 2 families, 206 and 200, srcinating from the same parental plant as an3-S205which did not segregate for the mutation. In this way the wild-type 206 and 200 families and the mutant 205 family are as isogenic as possible. As shown in Figure 1, at most two copies specific for the mutant 205 plant can be detected by these analyses. We also attempted to identify sequences containing dTphl which are unique for a mutant with a seedling-lethal phenotype. Since we could not isolate sufficient amounts of DNA from mutant seedlings we had to use plants heterozygous for the mutant allele. Therefore, the DNA used for Southern Rgure 1. Genomic Southern blot analyses o detect dTph copies specific for mutant plants. Two F amilies, 227 and 205, were compared with related wild-type F families. 206, 200 and 205 were F families derived rom a single parental plant. The same holds or 171, 170 and 227. The DNA was digested with Bcll and hybridized o dTphl. The open arrowheads o the right ndicate dTphl copies present n 205 and absent n 206. The closed arrowheads ndicate dTph copies present n 227 but absent n 171. blot analysis srcinated from an F 2 family ( 227) segregat- ing for the mutation. The wild-type DNA was isolated from two F 2 families ( 171 and 170) from the same parental plant that consisted of wild-type plants solely. On a genomic Southern blot at most five dTphl copies specific for the family containing the mutant locus could be detected (Figure 1). More extensive Southern blot analysis showed that none of these copies corresponded to the mutated gene (data not shown). When dealing with high copy number transposable elements discrimination on the basis of size is complicated, because many fragments will not be separated from each other. As an alternative we developed a method which differen- tiates flanking sequences by cross-hybridization. The basic principle of this method is outlined in Figure 2. Briefly, dTphl flanking sequences from mutant and wild-type plants are amplified by inverse polymerase chain reaction (IPCR) and the amplification products derived from the mutant plant are cloned in an M13 vector to obtain a library of dTphl flanking sequences. This library is subsequently hybridized to IPCR amplification products generated from  Gonomic DNA Genomic DNA from wild-type plant from mutant plant x B =, e A idonoin~l,~,,/ I ~ ~ select mutant specific clones mutable gene Figure 2. Cloning transposable element flanking sequences unique for mutant plants. Total genomic DNA of mutated and wild-type plants is digested with restriction enzyme X and ligated to form monomeric circles. After IPCR amplification using transposon-specific primers, products from the mutant are cloned in M13mp18. Duplicate filters are hybridized with either IPCR products from mutant or wild-type plants. Fragments A and B represent transposon copies that did not move in the two plants. Fragment C srcinates from the tagged gene and will show up as a differentially hybridizing clone, Gene cloning using high copy transposable elements 679 mutant and wild-type plants. The differentially hybridizing clones are isolated for further analysis. We first tested if we would be able to clone part of the unstable an3 allele (an3- 205) segregating in the 205 family by this approach. Since the f3h sequence is known (Britsch et al., 1991), we could select a restriction enzyme that would generate a fragment containing the dTphl flanking region whose size was suitable for IPCR amplifica- tion. Total genomic DNA of the plant families 205 and 206 was digested with EcoRl. In the case of 205 a genomic f3h fragment containing dTph 1 was generated as shown in Figure 3(a). The next step was to amplify dTphl flanking sequences by IPCR. In a first experiment we used primer out1, which is complementary to the terminal inverted repeat of dTphl. Figure 3(b) shows that this yields a range of amplification products (lanes 1-4), also when unligated restriction fragments were used as a template. Apparently these fragments do not result from IPCR but Figure 3. Cloning of an f3h gene fragment from a dTph I insertion mutant. (a) Structure of the an3-$205 allele containing the two EcoRI sites used to generate IPCR ragments. Black bars represent exon sequences, spaced by thin line introns. A dTphl insertion, indicated by the triangle, is present in the second exon of the f3h gene. Open bars represent the terminal inverted repeats of dTph 1. The an3-S206allele is identical exceptthat t lacks he dTph 1 insertion. Primers used in the experiment are shown above the transposable element. The fragment eventually cloned is indicated. (b) Analyses of dTph I flanking sequences generated by IPCR. Amplification products of genomic DNA digested with EcoRl: anes 1-4, primer out1; lanes 5-8, primers out2/out3; lanes 9-12, primers out2/out3 followed by amplification with primer out1. Lanes 1, 5 and 9, 205 ligated, lanes 2, 6 and 10, 205 unligated; lanes 3, 7 and 11, 206 ligated; lanes 4, 8 and 12, 206 unligated. The top panels show the ethidium bromide-stained agarose gels, the lower panels Southern blots from the same gels after hybridization with the f3h cDNA. Smaller amplification products hybridizing to the f3h cDNA were probably single stranded fragments. Sizes are indicated in kilobases. (c) Differential hybridization of M13 plaques containing cloned IPCR products. The IPCR products of an3-$205 plants were cloned in M13mp18. Triplicate plaque lifts were hybridized with IPCR products of an3-$205 plants, IPCR products of an3-$206 plants and the f3h cDNA, respectively. Note that clone 1 contains part of the f3h gene as indicated in (a). Clone 2 probably corresponds to another dTph I insertion unique to an3-$205 plants.  680 Erik Souer et al. from PCR amplification of linear molecules. Because the pattern of fragments varies between different plants, we assume that they are flanking sequences of two closely spaced dTphl elements or a single element close to a sequence resembling the out1 primer. Only when ligated DNA srcinating from the mutant plant was used as a template, relatively low amounts of f3h specific amplifica- tion products were obtained (about 10 pg in 1-2 p,g of amplification products) in addition to the previously men- tioned PCR amplification products of linear molecules (Figure 3b, lane 1). To improve the yield of dTphl flanking sequences we used nested primers. As the dTph I element consists almost completely of repeated and inverted repeated sequences, it was difficult to choose nested IPCR primers. From the few non-repeated parts of the element we made primers out2 and out3 (Figure 3a). After IPCR amplification with primers out2 and out3 no amplified products were detected, either by ethidium bromide stain- ing or by hybridization with the f3h cDNA (Figure 3b, lanes 5-8). Nevertheless, when re-amplifying part of this reaction with primer out1, increased amounts of dTphl flanking sequences were generated in comparison with amplifica- tion with out1 alone (Figure 3b, compare lanes 1 and 9). Apparently, primers out2 and out3 do amplify dTphl flanking sequences but with a relatively low efficiency. Nested IPCR products amplified from the mutant family ( 205) were cloned into the EcoRI site of M13mp18. Triplic- ate filters were prepared from these M13mp18 clones and hybridized with 32p-labelled IPCR fragments originating from 205 and 206 and with the f3h cDNA. Because we selected for flanking sequences that could be amplified from circular EcoRI fragments, only a limited number of plaques needed to be screened. Out of 80 plaques hybridiz- ing to the 205 probe, seven were differential in that they did not hybridize to the 206 probe. From those seven differentials two hybridized to the f3h cDNA. In Figure 3(c) an example is shown where plaque no. 1 is a differential that also hybridizes to the f3h cDNA. Plaque no. 2 is also differential, but does not hybridize to the f3h cDNA. Presumably it corresponds to another dTphl insertion unique for the 205 family. One clone hybridizing to the f3h cDNA was analysed by restriction mapping and found to contain the expected 3' flanking sequence of the dTphl element in the f3h gene (Figure 3a). Cloning novel genes tagged by the insertion of dTph 1 We attempted to clone dTphl flanking sequences unique for the mutant with a seedling-lethal phenotype. The mutant probe used in this experiment srcinated from the family S227 segregating for the mutation. The wild-type DNA was isolated from the closely related family S171 that consisted of wild-type plants solely. When dTphl flanking sequences of unknown identity Table 1. Fifteen classes of dTphl flanking sequences solated by differential screening of IPCR amplification products Class No. of clones Enzyme Band shift a I 14 EcoRI + II 9 BsfYI - III 7 EcoRI or Xbal b + IV 2 Bc/ + Bg/ll + V 1 Sstl + Vl 9 Hcll + VII 2 EcoRI + VIII 10 Hcll - IX 1 Bcll+ Bglll + X 5 BstYI - Xl 2 EcoRI - Xtl 1 Bcll + Bglll + XlII 5 EcoRI or Xbal b n.d. XlV 1 BstYI _c XV 6 BstYI + a+, Band shift detected on a genomic Southern blot; -, no band shift detected on a genomic Southern blot; n.d., not done. b Cross-hybridizing differentials separately cloned from Xbal and EcoRI digestions. c Middle repetitive. have to be amplified, the restriction enzyme that will generate a fragment smaller than 2 kb cannot be predicted. To overcome this problem, multiple digests were per- formed using six different restriction enzymes or combina- tions of restriction enzymes generating compatible ends. None of these restriction enzymes cut within dTph 1. After nested IPCR amplification of each of the digests, the frag- ments srcinating from the mutant plant family were cloned in M13mp18. At least 75 out of approximately 5000 recom- binant plaques hybridized to the mutant IPCR probe but not to the wild-type IPCR probe. After cross-hybridization experiments among the 75 clones, they could be divided into 15 unique classes (Table 1). One fragment of each class was hybridized to Southern blots containing restricted genomic DNA from plants segregating for the mutation. Figure 4 shows, as an example, the hybridization of a class XV clone illustrating a band shift of approximately 300 bp. The shifted band was seen in the mutant pool and not in the wild-type pool (Figure 4, lanes 1 and 2) and it corres- ponds well to the 284 bp size of dTphl. Furthermore, the shifted band was detected in multiple F 1 plants (Figure 4, lanes 3-10) indicating that it resulted from a sporogenic event. In general, most of the differential dTphl flanking sequences corresponded to low copy DNA in P. hybrida, except for the fragment of class XIV which was middle repetitive (Table 1). At least nine out of 15 fragments corresponded to DNA flanking a dTphl transposable ele- ment in the mutant plants and not in the srcinal wild-type plants. The class V clone showed 100% linkage to the gene responsible for the aberrant seedling phenotype in 18  Gene cloning using high copy transposable elements 681 Figure 4. Genomic Southern blot analysis of a class XV differential clone. Lane 1 contains DNA isolated rom the family 227 that was used to construct the library of dTph 1 lanking sequences and to make he mutant IPCR probe. Lane 2 contains DNA from the 171 family that was used o generate he wild-type PCR probe or differential screening. Lanes 3-10 contain genomic DNA of single F1 progeny plants obtained by selfing a 227 plant. The blot was hybridized with the insert of a differential clone from class XV (see Table 1). plants analysed. Further proof that the class V clone srcin- ated from the mutant locus, together with a detailed analysis of the structure and function of the locus will be presented elsewhere (Souer, van Houwelingen, Kloos, Mol and Koes, manuscript in preparation). Thus, in this way we were able to isolate dTphl flanking sequences unique to mutant plants. The same procedure as above was used to clone the regulatory an2 gene of the anthocyanin biosynthetic path- way (Quattrocchio, 1994). Since mutations in the an2 gene are not lethal, dTph I flanking sequences from one particu- lar mutant plant were screened against those from one homozygous revertant plant. This meansthat mutant probe and cloned fragments both srcinated from a single mutant plant. This screen yielded 74 differential clones which represented over 20 different classes. When we hybridized such differential clones to genomic Southern blots, in most cases a 300 bp band shift was detected in DNA from the mutant plant from which the IPCR products were generated, but not in any other genetically related plant. This indicated that these dTphl insertions were probably derived from somatic insertion events in the mutant plant. It seems unlikely that the same somatic insertion events occur in another mutant plant. Therefore, differential clones srcin- ating from somatic insertion events in one mutant plant are not expected to hybridize with IPCR products from a second mutant plant. Figure 5 shows that out of four differential clones, only plaque no. 1 hybridized to the IPCR probe from a second mutant plant (plant no. 2) and is therefore a candidate for the tagged locus. The other three clones may have srcinated from somatic insertion events in plant no. 1. In this particular experiment the number of differential clones decreased from the srcinal 74 to 27, representing only seven classes after cross-hybridization. Thus, by using IPCR fragments generated from one mutant plant for cloning and from another mutant plant for prob- ing, the number of unwanted differential dTphl flanking sequences identified can be reduced. Rgure 5. Screening of cloned PCR ragments rom one mutant plant with IPCR ragments rom another mutant plant. Plaques containing cloned IPCR products srcinating rom mutant plant no.1 were hybridized with IPCR products obtained rom different origins as ndicated on the left: he same mutant plant plant no.l), another mutant plant (plant no.2) and a wild-type plant. Note that plaque number 1 is detected by the probes rom both mutant plants and may correspond o the mutated gene. Plaques 2-4 hybridize o the probe rom mutant 1 only and probably correspond o somatic nsertion events n mutant plant 1. The use of the vectorette system to analyse differentials As differential dTphl flanking sequences can be obtained relatively easily, their characterization by Southern blot analysis becomes the most time-consuming and laborious step in the cloning of tagged genes. Genomic DNA needs to be isolated in large amounts and digested with various enzymes to show a band shift correlating with the pheno- type. To bypass this, we used the vectorette system (Riley et al., 1990) to amplify dTphl flanking sequences. In the vectorette system a special linker, which contains a non- complementarity region, is ligated to Sau3A- digested genomic DNA (Figure 6a). A primer identical to the non- complementary part of the linker will only anneal after an initial round of DNA synthesis starting from the out1 primer. Therefore, by using the dTph I primer out1 together with the universal vectorette primer in PCR, dTph I flanking sequences are specifically amplified. When performing such reactions on plants segregating for a particular muta- tion caused by a dTph insertion, part of the mutated gene
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks