The Plant Cell, Vol. 5, 371-378, April 1993 O 1993 American Society of Plant Physiologists

RESEARCH ARTICLE

Tagging and Cloning of a Petunia Flower Color Gene with the Maize Transposable Element Activator

George Chuck,’ Tim R o b b i q 2 Charanjit Nijjar, E d Ralston, Neal Courtney-Gutterson, and Hugo K. Dooner3 DNA Plant Technology Corporation, 6701 San Pablo Avenue, Oakland, California 94608

We report here the use of the maize transposable element Activator (Ac) to isolate a dicot gene. Ac was introduced into petunia, where it transposed into Ph6, one of several genes that modify anthocyanin pigmentation in flowers by affecting the pH of the corolla. Like other Ac-mutable alleles, the new mutation is unstable and reverts to a functional form in somatic and germinal tissues. The mutant gene was cloned using Ac as a probe, demonstrating the feasibility of heterol- ogous transposon tagging in higher plants. Confirmation that the cloned DNA fragment corresponded to the mutated gene was obtained from an analysis of revertants. In every case examined, reversion to the wild-type phenotype was correlated with restoration of a wild-type-sized DNA fragment. New transposed Acs were detected in many of the rever- tants. As in maize, the frequency of somatic and germinal excision of Ac from the mutable allele appears to be dependent on genetic background.

INTRODUCTION

Transposon tagging is a powerful means of gene isolation. It enables researchers to isolate genes on the basis of mutant phenotypes, rather than relying on prior knowledge of the gene’s primary function. In higher plants, this approach has been limited to those species, such as maize and snapdragon, that possess well-characterized endogenous transposable ele- ment systems. The transposable element Acfivator (Ac) has been used extensively as a tag to isolate genes in maize, its natural host (Walbot, 1992). Achas been introduced by trans- formation into a variety of heterologous hosts, such as tobacco, tomato, Arabidopsis, and petunia, where it has also been shown to transpose (Haring et al., 1991). To date, however, no instances of gene isolation by Ac tagging have been reported outside of maize.

Many of the genes that have been isolated by transposon tagging in maize and snapdragon affect anthocyanin pigmen- tation (Dooner et al., 1991a). This is readily understandable because the conspicuous nature of these pigments greatly facilitates the detection of the unstable phenotypes produced by transposable element alleles. Although endogenous trans- posons have been described in petunia (Gerats et al., 1990), transposon tagging has not been used as a gene isolation tool

Current address: Department of Plant Biology, University of Califor- nia, Berkeley, CA 94720. * Current address: Cambridge Laboratory. The John lnnes Centre, Colney Lane, Norwich NR4 7UJ, England. To whom correspondence should be addressed.

in this species. A great diversity in flower color exists in petu- nia, and many genes are known that modify anthocyanin expression by affecting, for example, the synthesis of flavonol copigments (n), or the pH of the corolla @h7 to ph5) (Wiering and de Vlaming, 1984). With the objective of tagging some of these modifier genes, we introduced the maize transposon Ac into petunia, one of the several dicot species in which Ac transposition has been demonstrated (Haring et al., 1989).

In this report, we document the isolation of a striking varie- gated flower color mutation in petunia, which arose by insertion of Ac into the Ph6 gene, and the cloning of the mutant gene using Ac as a transposon tag. This constitutes a demonstra- tion of heterologous transposon tagging in higher plants, i.e., the isolation of a gene from one plant species using a trans- posable element from a different plant species as a probe.

RESULTS

lsolation of a Variegated Flower Color Mutation in a Petunia Line Carrying Ac

Severa1 independent transformants carrying Ac have been generated in V26, a highly inbred, purple-flowered genetic line that was obtained from the collection at the Free University of Amsterdam (Robbins et al., 1991). The binary vector used in transformation, pJJ4411, has been described previously

372 The Plant Cell

(Keller et al., 1993a). In addition to a hygromycin resistancetransformation marker, this vector contains the streptomycinphosphotransferase (SPJ)::Ac excision marker between theright and left T-DNA borders. The maize element Ac interruptsthe SPfgene and prevents its expression (Jones et al., 1989).In several plants, such as tobacco and Arabidopsis, this markeris a useful visual indicator of somatic and germinal Ac activity(Jones et al., 1989; Dean et al., 1992; Keller et al., 1992). Inpetunia, however, the streptomycin germination screen is notas reliable. It can be used to enrich for plants carrying trans-posed Ac elements (trAcs), but such plants need to beconfirmed by DMA gel blot analysis. This procedure resultsin a greater than 10-fold enrichment for trAcs in petunia (T.Bobbins and N. Courtney-Gutterson, unpublished results).

One of the plants analyzed (3057.12) carried two Ac elementsin heterozygous condition: one still in its resident site in a T-DNA(Jones et al., 1989) and the other at an unlinked chromosomallocation, into which it had integrated following a secondarytransposition event from a different T-DNA. Plant 3057.12 wasselfed to make homozygous the transposed Ac (trAc) elementand, therefore, any mutation caused by the trAc insertion. Whenthe self-progeny were planted, a new, variegated flower colorphenotype was found to segregate as a simple recessiveMendelian trait. As can be seen in Figure 1A, the variegatedflower phenotype is striking; darkly colored (revertant) sectors,

outlined by white rims, stand out sharply against the palecolored (mutant) background of the corolla. The color of thebackground and the revertant sectors varies depending on theresidual genotype. In segregants from outcrosses to othergenetic lines, the background color was blue and the rever-tant color red; however, in all flowers, a white rim separatesthe revertant sector from the mutant background. Borders ofdifferent colors are rarely seen in examples of anthocyaninvariegation. When they are seen, the rims tend to be more,not less, pigmented than the areas they delimit. The forma-tion of new anthocyanin pigments in the border cells has beenattributed to the diffusion of accumulated intermediates fromadjacent cells (McClintock, 1951; Rhoades, 1952). The pres-ence of white rims in the variegated flowers also suggests thatcompounds can diffuse into the border cells from adjacent cells,i.e., the effect of the mutated gene on anthocyanin pigmenta-tion is not strictly cell autonomous.

Evidence That the New Mutation Is Tagged by Ac

The following evidence indicated that the new variegated mu-tation had arisen as a consequence of an Ac transpositionevent, and was, therefore, tagged by Ac.

Figure 1. Flower Phenotypes.

(A) Variegated petunia corolla produced by a plant carrying the new unstable allele.(B) Plant with a variegated mutant flower (bottom) and solidly colored revertant flower (top).(C) Branch showing an older variegated flower (left) with a faded background color relative to that of the recently opened variegated flower (right).

Heterologous Transposon Tagging in Petunia 373

kb T- CM m •* mo- > > > > > (D T- CM to "t m co i^ootn> W C O W W W W W W C 5

9.4-

6.6-

4.4-

Bkb o.

9.4-

6.6- *»

4.4-

:*»

1.0kbFigure 2. DMA Gel Blot Analysis of Variegated and Solidly ColoredProgeny of Plant 3057.12, Carrying a trAc Element.

Genomic DMA (6 ng) was digested with EcoRI, separated by electropho-resis on a 1% agarose gel, and transferred to a nylon membrane. Pindicates the 3057.12 parent; V1 to V6, variegated progeny; S1 to S8,solidly colored progeny; GB, nontransgenic plant of V26 genetic back-ground. Molecular length markers are given at left in kilobases.(A) Hybridization to a probe made from the EcoRI-Hindlll fragment of Ac.(B) After removal of the Ac signal, hybridization to a probe made fromthe BstXI-EcoRI fragment flanking the Ac insertion.(C) Restriction map of the gene mutated by insertion of Ac. The posi-tions of the probes corresponding to Ac and to the flanking DMA areindicated by bars above the map.

Cosegregation with Ac-Hybridizing Band

The mutation cosegregated with a new Ac-hybridizing bandin DNA gel blots of the self-progeny of plant 3057.12. DMAfrom the variegated (mutant) and solidly colored (parental)progeny was analyzed by digestion with different enzymes andhybridization with an Ac probe. In an EcoRI digest, a newAc-hybridizing band was found that cosegregated with the newvariegated phenotype. As seen in the genomic DNA gel blotshown in Figure 2A, a 4.7-kb band is present in every varie-gated plant (V1 to V6) but only in some solidly colored siblings(S1 to S3).

The larger (~5-kb) band represents the second unlinked Acelement. This band can be seen to segregate in both the varie-gated and the solidly colored siblings. Among those variegatedplants that received both Ac fragments, the intensity of the 4.7-kb band relative to the 5-kb band is either the same (V5) ordouble (V2 to V4), suggesting that the new, 4.7-kb band ishomozygous but that the 5-kb band can be either homozy-gous (V5) or heterozygous (V2 to V4). Conversely, some solidlycolored progeny (S3 and S7) appear to be homozygous forthe 5-kb but not the 4.7-kb band.

In all, 26 variegated siblings were analyzed and all showedthe new, Ac-hybridizing band at approximately the same rela-tive intensity. Of 25 solidly colored siblings analyzed, 16 hadthe 4.7-kb band, a result in agreement with the proportion ofheterozygotes for the new mutation (two-thirds) expected withinthe solidly colored class. These data indicate that the new mu-tation is linked to the trAc band (x2 = 10.2, P < 0.01). Althoughno recombinants were found in the self-progeny, the resolu-tion of this type of F2 linkage data (repulsion phase, completedominance) is limited, so the 95% confidence interval for theestimate of p, the recombination fraction, is large (p = 0;Cl = 0-0.34).

Homozygosity for the Ac-Tagged Fragment

All of the mutant plants were homozygous for the Ac-taggedDNA fragment. The Ac-homologous, 4.7-kb EcoRI fragment,containing part of Ac and DNA adjacent to the Ac insertion,was cloned into the vector XZapll (Stratagene). A restrictionmap of this fragment (and of the adjacent 6.8-kb EcoRI frag-ment subsequently isolated) is shown in Figure 2C. TheBstXI-EcoRI fragment flanking Ac in the 4.7-kb Ac fragmentwas labeled and used to reprobe the blot shown in Figure 2A.If the variegated plants are indeed homozygous for the new,4.7-kb Ac band, they should lack the allelic wild-type fragment,which, conversely, should be present in all of the solidly coloredsiblings. The DNA gel blot presented in Figure 2B confirmsthis expectation. The solidly colored progeny now show a 7-kbband of identical mobility to the band seen in the wild-typeV26 inbred parent. In contrast, the variegated siblings eitherlack that band or show it at a very reduced intensity. The weakband present in the variegated progeny can be attributed tooccasional somatic excisions of Ac.

374 The Plant Cell

As anticipated, those solidly colored progeny that did notshow a 4.7-kb /Ac-homologous band in Figure 2A are homozy-gous for the V26 wild-type fragment. Solidly colored progenythat did show a 4.7-kb /Ac-homologous band are heterozygousfor the V26 fragment. The overall segregation obtained fromhaving scored the progeny of plant 3057.12 with the two differ-ent probes (Ac and its flanking sequence) is as follows. Amongthe 25 solidly colored progeny analyzed, 16 were Acl+, 9 were+/+, and none was Ac/Ac. Among the 26 variegated progeny,all were Ac/Ac. This more complete genotypic classificationof the F2 progeny significantly reduces the size of the 95°/oconfidence interval for p (p = 0; Cl = 0-0.05) and demon-strates that the new mutation is, in fact, closely linked to Ac.

Reversion to Wild-Type Phenotype

Reversion of the mutation to the wild-type phenotype was cor-related with restoration of a wild-type-sized DMA fragment.Confirmation that a mutation is in fact tagged by Ac can besought from an analysis of revertants because an excision ofAc that restores the wild-type phenotype should also producea DNA fragment of the original wild-type size. Progeny fromvariegated plants were grown and screened for somatic andgerminal reversion events. Branches with solidly coloredflowers, representing large somatic revertant sectors, were oc-casionally seen on plants producing mostly variegated flowers(Figure 1B). DNA from the solidly colored and variegated

Akb

9.4-

6.6-

4.4-

1 2 3 4 5Bkb

9.4-

1 2 3 4 5 6

6.6-1* '

4.4-

Flgure 3. DNA Gel Blot Analysis of Revertants.Genomic DNA (6 ng) was digested with EcoRI, separated by electropho-resis on 1% agarose gels, transferred to nylon membranes, and probedwith the BstXI-EcoRI fragment from DNA flanking the Ac insertion (barin Figure 2C). Molecular length markers are given at left in kilobases.(A) Somatic sector shown in Figure 1B. Lane 1, 3057.12 parent plant;lanes 2 and 3, leaf and flower, respectively, borne on a variegatedbranch; lanes 4 and 5, leaf and flower, respectively, borne on a solidlycolored branch.(B) Germinal revertants. Lane 1, V26 inbred line; lanes 2, 5, and 6,solidly colored progeny of a variegated plant; lanes 3 and 4, varie-gated progeny of same plant.

Table 1. Frequency of Germinal RevertantsNumber

Number Solidly FrequencyVariegated Colored Germinal

Family Pedigree Plants Plants Revertants8

342634283434346634293468

V26 x M59V26xM59V26xM59V26xM59V26V26

7310335512291

92513926

0.110.190.270.150.080.06

a Number of solidly colored plants to total number of plants.

branches of one such plant was prepared and analyzed byDNA gel blotting. The blot shown in Figure 3A was probedwith the BstXI-EcoRI DNA fragment flanking Ac (Figure 2C).Two bands of roughly equal intensity can be seen in the lanescontaining DNA from a flower and a leaf that were borne ona revertant branch (lanes 4 and 5). One is a 7-kb, wild-type-sized band and the other, a 4.7-kb band, which also hy-bridizes to Ac (data not shown). This observation indicates thatthe revertant sectors are heterozygous for the original Ac-induced mutation and a revertant allele produced by excisionof Ac during development of the chimeric plant. Capsules borneon the revertant branches produced solidly colored and varie-gated individuals in a 3:1 ratio, confirming that the reversionevent was heritable. The lanes containing DNA from a flowerand leaf that were borne on a variegated branch (Figure 3A,lanes 2 and 3) show, in contrast, a strong 4.7-kb Ac band anda faint 7-kb band. The latter band probably represents emptysites generated by somatic excisions of Ac during the forma-tion of the variegated flower. Capsules borne on variegatedbranches produced, as expected, mostly variegated progeny.

Plants with only solidly colored flowers have been obtainedamong the progeny of variegated plants at frequencies rang-ing from 6 to 27%, indicating that the new mutation is alsogerminally unstable and reverts frequently to the wild-type state.Representative reversion data are shown in Table 1. To date,10 independent germinal revertants have been analyzed byDNA gel blots, confirming the observations made earlier forthe somatic revertant sectors. Figure 3B illustrates the analy-sis of three such germinal revertants. All the revertants (lanes2,5, and 6) showed the 7-kb, wild-type-sized band in additionto the 4.7-kb band. Therefore, they are heterozygous for arevertant allele and the original Ac-induced mutation. Thesegregating variegated siblings (Figure 3B, lanes 3 and 4), onthe other hand, showed only the 4.7-kb band; they are homozy-gous for the Ac mutation, as expected from their phenotype.In addition, six of eight revertants analyzed had new Ac bands,indicating that the excised Ac elements continue to be capa-ble of reinsertion (data not shown).

We concluded from the above evidence that the new varie-gated petunia mutant arose from the transposition of the maizeelement Ac into a gene affecting flower color.

Heterologous Transposon Tagging in Petunia 375

The Gene Tagged by Ac Affects the Acidity ofthe Corolla

Several considerations suggested that the new/to-tagged mu-tant was apft-type mutation, i.e., a mutation in one of the severalgenes known to affect anthocyanin pigmentation in petuniaby changing the acidity of the corolla (de Vlaming et al., 1983;Wiering and de Vlaming, 1984). First, in certain genetic back-grounds, the revertant sectors in the variegated flowers appearred, whereas the mutant background has a distinct bluish hue.This color change is reminiscent of that brought about by theph mutations in petunia. These mutations cause a bluingof the corolla by increasing the vacuolar pH in the antho-cyanin-accumulating cells (Wiering and de Vlaming, 1984).Anthocyanins in solution undergo a similar shift from red toblue as the acidity decreases.

Second, the new mutation affects the pH of the corolla ina manner similar to the known ph mutant phi. This was estab-lished by comparing the corolla pH of mutant and revertantplants that arose in the self-progeny of a variegated plant, whichwas also Hf1 Ph1/hf1 phi. Hf1 and Ph1 are closely linked geneson chromosome 1 that affect pigmentation of the corolla. Hflcontrols hydroxylation at the 5' position of the anthocyanin Bring and causes a bluing of the corolla. Ph1 increases acidityin the vacuoles of the corolla and produces a more reddishhue. The Hf1 Ph1/hf1 phi heterozygote was obtained from anoutcross of a variegated V26 plant (Hf1 Ph1)\o line M59 (hf1phi). Among Hf1 segregants, the pH of the corolla was higherin the mutant plants (5.89 ± 0.02) than in the revertant plants(5.57 ± 0.03), a result that suggested that the new mutationaltered the acidity of the corolla. This increase in pH can be

BHtfJK

Figure 4. Seeds Produced by the ph6-m1(Ac) Mutation.

The variegated seeds show normal, pigmented sectors on a mutant,unpigmented background. The pigmented seed indicated by the ar-row looks normal and probably arose as a consequence of an Acexcision event early in seed coat development.

compared to that caused by the phi mutation in the samesegregating family. Because phi and hf1 are only 1 centimor-gan apart, hf1 segregants should also be ph1/ph1 and,therefore, were used to compare the effects on corolla pH ofphi and of the new mutation. Two revertant progeny of the hf1class (and, presumably, pM/phl) were recovered; both hadhigh corolla pH values (5.90 ± 0.05), similar to those mea-sured in the variegated flowers of the Hf1 class.

Third, in some variegated plants the color of the flower fadeswith aging (Figure 1C), a phenomenon that has also been ob-served in petunia lines carrying certain ph mutations inconjunction with the Fa allele (Wiering and de Vlaming, 1984).The pigments present in the faded corollas of the variegatedline were compared with those present in the faded corollasof aph4 mutant line. In both cases, the faded flowers accumu-lated a phenolic compound fluorescing blue under UV light.This compound was absent in extracts of recently opened, non-faded flowers.

Fourth, the new mutation has a pleiotropic effect on seeddevelopment, an effect that has also been associated withsome ph mutations (Wiering and de Vlaming, 1984). When ex-amined under 30x magnification, seeds borne on mutantplants appear abnormal. Some are shriveled or irregularlyshaped and the vast majority are variegated. The seed coatof normal petunia seeds is uniformly pigmented and reticu-lated. As shown in Figure 4, the mutant seed coat is largelyunpigmented and lacks the honeycomb network of normalseeds, except for areas of varying size where the pigmentedand reticulated peripheral structure is restored. The variegatedseed phenotype of the .Ac-induced mutation can be readily ex-plained as another manifestation of somatic instability: thenormal sectors on the mutant seed coats would form as a re-sult of Ac excisions that occurred during seed development.

The observations given above cumulatively suggested thatAc had become inserted in one of the Ph genes. We conductedallelism tests with existing ph mutants and established that thevariegated mutant was capable of complementing all of themutants tested, except forp/76, as illustrated in Figure 5. There-fore, we have designated the new, Ac-tagged ph6 mutationph6-m1(Ac). The petunia line carrying the standard ph6 muta-tion used in the complementation test is W160. The flowerphenotype conditioned by the ph6-m1(Ac) allele in outcrossesto W160 is clearly different from that produced in the pure V26line or in a mixed V26/M59 genetic background. Only smallsectors can be seen (Figure 5) due to reversion events thatoccur late in flower development. The seed phenotype is simi-larly affected: a few of the seeds borne on the outcross plantsshow traces of pigmentation, but the majority are unpigmented(data not shown).

The ph6-m1(Ac) Mutation Encodes an Altered Form ofa Flower-Specific Transcript

To detect a Ph6 transcript, total RNA was prepared from Ph6and ph6-m1(Ac) flower buds and leaves. The RNAs were sepa-rated on a 1.1% agarose gel, blotted onto a nylon membrane,

376 The Plant Cell

1 2 3 4

Figure 5. Flower Phenotypes from the Allelism Test to ph6.

(A) Ph6/ph6 (top) and ph6-m1(Ac)lph6 (bottom), both in a V26/W160genetic background.(B) Magnification of bottom flower in (A) showing the small revertantsectors on a mutant background.

B

I

Figure 6. RNA Gel Blot of Wild-Type and Mutant Flower Buds andLeaves.

Total RNA (10 ng) was separated on a 1.1% agarose gel, transferredto a nylon membrane, and hybridized sequentially to three probes.(A) Hybridization to the EcoRI-BamHI DNA fragment on the left sideof the Ac insertion (Figure 2C).(B) Hybridization to the flower-specific CHS-A probe (Koes et al., 1989).(C) Hybridization to a wheat rDNA probe.Lane 1, mutant leaf; lane 2, 2-cm mutant flower bud; lane 3, wild-typeleaf; lane 4, 2-cm wild-type flower bud. Top arrow indicates the posi-tion of 28S rRNA; bottom arrow, position of 18S rRNA.

and probed with the EcoRI-BamHI fragment that extends fromthe BamHI site in Ac to the left of the insertion site (Figure2C). The corresponding RNA gel blot is shown in Figure 6A.An ~2.8-kb transcript is detected in wild-type flower buds (lane4). In the mutant flower bud (lane 2), only a trace of the 2.8-kbtranscript can be seen; the major signal is given, instead, bya less than 2-kb transcript. Possibly, alternate splicing causedby the Ac insertion accounts for the multiple transcripts seenin the mutant (Wessler, 1988). No transcript can be detectedin either mutant or wild-type leaves (lanes 1 and 3), suggest-ing that the Ph6 gene is expressed preferentially in flowers.

The RNA gel blot was rehybridized with the petunia flower-specific chalcone synthase probe CHS-A (Koes et al., 1989)after washing away the first probe (Figure 6B). As expected,transcripts of the same size and intensity were detected in mu-tant and wild-type flower buds (lanes 2 and 4, respectively)but not in mutant and wild-type leaves (lanes 1 and 3, respec-tively). This result shows that the mutant RNA sample was notdegraded and confirms the flower-specific nature of the Ph6transcript. Finally, Figure 6C shows the comparable 28S rRNAsignal given by the four RNA samples when the RNA gel blotwas rehybridized with a wheat rDNA probe, confirming thatthe four lanes were loaded with approximately the sameamount of RNA.

Heterologous Transposon Tagging in Petunia 377

DlSCUSSlON

The new petunia unstable allele described here represents a validation of the heterologous transposon tagging strategy. The ph6-m7(Ac) mutation arose in a random or nondirected Ac-tagging experiment in which a petunia plant carrying a trAc in heterozygous condition was selfed to reveal the mutation. We do not know the linkage relationship between the Ph6gene and the donor locus from which Ac transposed because the trAc element was not initially inserted in a readily scored exci- sion marker. The random transposon tagging approach resembles T-DNA tagging (Feldman, 1991) in that new, reces- sive mutations at nontargeted loci are revealed after the insertion is made homozygous. Given Ads propensity to trans- pose to linked sites in most systems (Greenblatt, 1984; Dooner and Belachew, 1989; Jones et al., 1990; Dooner et al., 1991b; Keller et al., 1993b), a directed tagging approach using an Ac element linked in cis to the target locus is preferable in experi- ments that aim to tag specific genes. However, the process of identifying a close linkage between an introduced T-DNA and the gene of interest is time consuming, so the first few mutations tagged with heterologous transposons in dicots can be expected to arise from random tagging experiments such as this one.

The frequency of revertant sectors in the variegated flowers and of germinal revertants among the progeny of the mutant indicate that Ac excision from the ph6-ml(Ac) allele is not a raie event. Ac appears to excise as frequently from the petu- nia Ph6 gene as it does from maize genes, such as Bronze and W a x ~ In addition, Ac transposition, i.e., excision and rein- sertion, was demonstrated in six of eight germinal revertants. This ratio of recovery of trAc elements relative to excisions is also similar to what has been reported in maize (McClintock, 1956; Greenblatt, 1984; Dooner and Belachew, 1989) and other dicots (Jones et al., 1990; Dooner et al., 1991b; Dean et al., 1992; Keller et al., 1992).

The residual genetic background has been shown to affect both somatic and germinal excision of Ac in maize (Brink and Nilan, 1952) and may also do so in petunia. The developmen- tal timing of Ac excision from ph6-m7(Ac), as evidenced by the size of the revertant sectors, is considerably delayed in a V26IW160 background relative to a V26 or a V26lM59 back- ground (Figure 5). Similarly, the frequency of germinal reversion appears to be higher in a V261M59 hybrid background than in a V26 inbred background (Table 1). The reduced variega- tion observed in the V26xW160 outcross plants may also be partly due to lower Ac dosage because these plants are het- erozygous, rather than homozygous, for the ph6-m7(Ac) allele. A positive dosage effect of Ac on somatic variegation has been previously documented for the engineered Sm:Ac gene in both tobacco and Arabidopsis (Jones et al., 1989; Dean et al., 1992; Keller et al., 1992).

The Ph6 gene tagged by Ac modifies anthocyanin pig- mentation in the corolla. A Ph6-specific transcript was de- tected in flowers, but not leaves, indicating that the gene is

preferentially expressed in corolla cells. However, because the ph6 mutation also affects the phenotype of the seed coat (Fig- ure 4), the Ph6 gene must also be expressed during seed development. At this time, we do not know the gene’s primary function. We have isolated severa1 clones from a petunia flower bud cDNA library and are in the process of sequencing them. Comparison with other gene sequences may provide a clue to the function of Ph6.

The white rims around the revertant sectors in the variegated flowers remain a puzzle. Certain forms of anthocyanins have been shown to appear red at low pH, colorless at higher pH values, and blue at pH values near neutrality (Timberlake and Bridle, 1975). If the white rim is due to an intermediate pH value at the border between mutant and revertant tissue, we thought it would be possible to recreate this condition in vitro by prepar- ing anthocyanin extracts of the variegated flowers and then examining color over a pH range from 2 to 6. When this was done, the expected shift from red to blue was observed as the pH was increased; however, no pH was found that gave a color- less solution. It appears that the particular anthocyanin form found in these extracts of petunia flowers is sufficiently stable not to become colorless at any acidic pH value. We are left to speculate that a diffusible compound essential for the sta- bility of anthocyanins is deficient in mutant tissues, possibly as a consequence of a defective transport system. Normal func- tioning cells at the borders of the revertant sectors would deplete that compound from the adjacent mutant cells, thus preventing the accumulation of anthocyanins in the latter. In agreement with this model, we have observed that the white rim’s border is sharp on the side of the revertant sector but diffuse on the side of the mutant background. This is what would be expected if the cells in the white rim were genotypi- cally mutant and were depleted of the pigment-stabilizing compound by the adjacent normal cells.

METHODS

Constructs

The binary vector plasmid pJJ4411 is described in detail elsewhere (Keller et al., 1993a).

Plant Genotypes

V26 petunia plants were transformed with Agmbacterium tumefaciens LBA4404 carrying the binary vector plasmid 4411. Other petunia lines used in this work were M59 and W160, all obtained from the collection maintained at the VUA (Free University, Amsterdam).

Nucleic Acid Extraction and Analysis

DNA was extracted from 4- to 6-cm flower buds by a CTAB procedure (Dooner et al., 1991b). Approximately 6 pg of DNA was digested with the appropriate restriction enzymes, separated on 1% agarose gels,

378 The Plant Cell

and blotted to Duralon-UV membranes (Stratagene). The DNA gel blots were hybridized to random primer-labeled probes. RNA was extracted from ground tissue resuspended in 5 M guanidinium thiocyanate (Schmidt et al., 1981). Ten micrograms of total RNA was separated on 1.1% agarose gels and blotted to Duralon membranes. The RNA gel blots were hybridized to random primer-labeled probes.

Genomic Cloning

DNA was extracted from pooled leaf nuclei of variegated plants V1 and V6 (Figure 2). Approximately 40 pg of DNA was digested with EcoRl and size fractionated by centrifugation through a 20 to 40% glycerol gradient. Fractions containing the 5-kb- and 6-kb-hybridizing bands corresponding to the 3’and 5’halves of Ac, respectively, were ligated to Napll (Stratagene). Positive clones were subcloned into pBluescript SK- (Stratagene).

ACKNOWLEDGMENTS

We thank Eda Lim for technical assistance, Rob Narberes and Joyce Hayashi for figure preparation, Cara Velton and Sara Glenn for plant care, Tom Gerats for line M59, Ronald Koes for line W160, and Janis Keller and William Tucker for valuable discussions. The petunia lines carrying Ac were generated originally for a projecf funded by Florigene, Rijnsburg, The Netherlands. This is DNA Plant Technology paper num- ber 5-24.

Received January 19, 1993; accepted February 8, 1993.

REFERENCES

Brink, R.A., and Nilan, R.A. (1952). The relation between light varie- gated and medium variegated pericarp in maize. Genetics 37,

Dean, C., Sjodin, C., Page, T., Jones, J., and Lister, C. (1992). Be- haviour of the maize transposable element Ac in Arabidopsis maiana. Plant J. 2, 69-81.

de Vlaming, P., Schram, A.W., and Wiering, H. (1983). Genes af- fecting flower colour and pH of flower limb homogenates in Petunia hybrida. Theor. Appl. Genet. 66, 271-278.

Dooner, H.K., and Belachew, A. (1989). Transposition pattern of the maize element Ac from the bz-m2(Ac) allele. Genetics 122,447-457.

Dooner, H.K., Robbins, T., and Jorgensen, R.A. (1991a). Genetic and developmental control of anthocyanin biosynthesis. Annu. Rev. Genet. 25, 173-199.

Daoner, H.K., Keller, J., Harper, E., and Ralston, E. (1991b). Vari- able patterns of transposition of the maize element Activator in tobacco. Plant Cell 3, 473-482.

Feldman, K.A. (1991). T-DNA insertion mutagenesis in Arabidopsis: Mutational spectrum. Plant J. 1, 71-82,

Gerats, A.G.M., Huits, H., Vrijlandt, E., Marafia, C., Souer, E., and Beld, M. (1990). Molecular characterization of a nonautonomous transposable element (dTph7) of petunia. Plant Cell2, 1121-1128.

519-544.

Greenblatt, I.M. (1984). A chromosome replication pattern deduced from pericarp phenotypes resulting from movements of the trans- posable element Modulator in maize. Genetics 108, 471-485.

Haring, M.A., Gao, J., Volbeda, T., Rommens, C.M.T., Nijkamp, H.J.J., and Hille, J. (1989). A comparative study of Tam3 and Ac transposi- tion in transgenic tobacco and petunia. Plant MOI. Biol. 13,189-201.

Haring, M., Rommens, C., Nijkamp, J., and Hille, J. (1991). The use of transgenic plants to understand transposition mechanisms and to develop transposon tagging strategies. Plant MOI. Biol. 16,

Jones, J.D.G., Carland, F., Maliga, P., and Dooner, H.K. (1989). Vi- sual detection of transposition of the maize element Activator in tobacco seedlings. Science 244, 204-207.

Jones, J.D.G., Carland, F., Lim, E., Ralston, E., and Dooner, H.K. (1990). Preferential transposition of the maize element Activator to linked chromosomal locations in tobacco. Plant Cell 2, 701-707.

449-461.

Keller, J., Lim, E., James, D.W., and Dooner, H.K. (1992). Germina1 and somatic activity of the maize element Activator(Ac) in Arabidop- sis. Genetics 131, 449-459.

Keller, J., Jones, J.D.G., Harper, E., Lim, E., Carland, F., Ralston, E., and Dooner, H.K. (1993a). Effect of gene dosage and sequence modification on the frequency and timing of transposition of the maize element Activator in tobacco. Plant MOI. Biol. 21, 157-170.

Keller, J., Llm, E., and Dooner, H.K. (1993b). Preferential transposi- tion of Ac to linked sites in Arabidopsis. Theor. Appl. Genet., in press.

Koes, E., Spelt, C.E., and MOI, J. (1989). The chalcone synthase mul- tigene family of Petunia hybrida (V30): Differential, light-regulated expression during flower development and UV light induction. Plant MOI. Biol. 12, 213-225.

McClintock, B. (1951). Chromosome organization and gene expres- sion. Cold Spring Harbor Symp. Quant. Biol. 16, 13-47.

McClintock, 8. (1956). Controlling elements and the gene. Cold Spring Harbor Symp. Quant. Biol. 21, 197-216.

Rhoades, M.M. (1952). The effect of the bronze locus on anthocyanin formation in maize. Am. Nat. 86, 105-108.

Robbins, T., Jenkin, M., and Courtney-Gutterson, N. (1991). A high frequency of secondary transposition of Ac in petunia. Third Inter- national Congress of the lnternational Society for Plant Molecular Biology, Oct. 6-11, Tucson, AZ, 1737 (abstr).

Schmidt, G.W., Bartlett, S.G., Grossman, A.R., Cashmore, A., and Chua, N.-H. (1981). Biosynthetic pathways of two polypeptide subunits of the light-hawesting chlorophyll a/b protein complex. J. Cell Biol.

Timberlake, C.F., and Bridle, R (1975). The anthocyanins. In The Flave noids, J.B. Harborne, T.J. Mabry, and H. Mabry, eds (New York: Academic Press), pp. 214-266.

Walbot, V. (1992). Strategies for mutagenesis and gene cloning using transposon tagging and T-DNA insertional mutagenesis. Annu. Rev. Plant Physiol. Plant MOI. Biol. 43, 49-82.

Wessler, S.R. (1988). Phenotypic diversity mediated by the maize trans- posable elements Ac and Spm. Science 242, 399-405.

Wiering, H., and de Vlamlng, P. (1984). Genetics of pollen and flower color. In Monographs on Theoretical and Applied Genetics 9: Petu- nia, K.C. Sink, ed (Berlin: Springer-Verlag), pp, 49-67.

91, 468-478.

DOI 10.1105/tpc.5.4.371 1993;5;371-378Plant Cell

G. Chuck, T. Robbins, C. Nijjar, E. Ralston, N. Courtney-Gutterson and H. K. DoonerActivator.

Tagging and Cloning of a Petunia Flower Color Gene with the Maize Transposable Element

 This information is current as of May 17, 2018

 

Permissions X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists