The Use of Antisense mRNA to lnhibit the Tonoplast H+ ATPase in Carrot

The Plant Cell, Vol. 4, 851-864, July 1992 O 1992 American Society of Plant Physiologists

The Use of Antisense mRNA to lnhibit the Tonoplast H+ ATPase in Carrot

J. Peter Gogarten,’ Jenny Fichmann,* Yael B r a ~ n , ~ Louis Morgan, Paul Styles, Saundra Lee Taiz, Karen DeLapp, and Lincoln Taiz4 Sinsheimer Laboratories, Biology Department, University of California, Santa Cruz, California 95064

Carrot root cells were transformed with the coding or 5’ noncoding regions of the carrot vacuolar H+ ATPase A subunit cDNA cloned in the antisense orientation behind the cauliflower mosaic virus 35s promoter. Bafilomycin-sensitive ATP- ase, H+-pumping, and 14C-O-methyl-glucose uptake activities were specifically inhibited in the tonoplast fractions of mutant cell lines. Protein gel blotting confirmed that the expression of the A subunit was inhibited in the tonoplast fraction, but not in the Golgi fraction. ’RHo-dimensional protein gel blots of total microsomes of wild-type and control transformant cell tines revealed two major immunoreactive polypeptides in the acidlc pl range. In contrast, highly purified tonoplast membranes contained only the less acidic polypeptide. Because the less acidic polypeptide was preferentially diminished in the two antisense cell lines, we infer that the antisense constructs specifically blocked expression of a tonoplast-specific isoform of the V-ATPase A subunit in carbt. Regenerated plants containing the antisense constructs exhibited altered leaf morphologies and reduced cell expansion. The altered phenotype was correlated with the presence of the antisense construct.

INTRODUCTION

Vacuolar H+-pumping ATPases (V-ATPases) are large, multi- meric enzymes composed of a hydrophilic, catalytic V1 complex and a hydrophobic, membrane-spanning Vo complex. V1 is composed of two major subunits (A and B) and three minor subunits (C, D, and E), with a stoichiometry of A3B3CDE. The A and B subunits are related to the p and a subunits, respectively, of the FoF1-type ATPases (Gogarten et al., 1989), with the catalytic site located on the A subunit (Nelson and Taiz, 1989). VO contains six copies of a 16-kD proteolipid (c subunit) and one copy each of minor subunits of about 100, 38, and 19 kD (Nelson, 1991). Recent electron microscope im- ages of negatively stained vacuolar preparations have shown that the V1 complex forms a “ball-and-stalk” structure similar to the F, complex of the FoF1-type ATPases (Bowman et al., 1989; Klink and Lüttge, 1991; Taiz and Taiz, 1991).

The primary function of V-ATPases is to provide the proton motive force for solute transport across various endomembrane compartments of eukaryotic cells, including vacuoles, lysosomes, Golgi bodies, endosomes, coated vesicles, and chromaffin granules (Forgac, 1989). In plant cells, V-ATPases

Current address: Department of Molecular and Cell Biology, The University of Connecticut, Storrs, CT 06269-3044. * Current address: Parasitology Laboratory, San Francisco General Hospital, San Francisco, CA 94110. Current address: Department of Biological Chemistry, The Hebrew

University of Jerusalem, Jerusalem, Israel, 91904. To whom correspondence should be addressed.

are present on the tonoplast, Golgi, and coated vesicle membranes (Chanson and Taiz, 1985; Fichmann et al., 1989) and are likely to be present on other related structures and organelles, including multivesicular bodies and the trans Golgi network. In addition to transport, V-ATPases in animal cells have been implicated in the regulation of receptor-ligand in- teractions in endosomes and post-translational modifications in the Golgi (Forgac, 1989).

The multiple locations and diverse functions of the V-ATPases suggest that the enzyme plays an essential role in eukaryotic cells. In yeast, which contains single copies of the vacuolar H+ ATPase genes, it has been possible to directly assess the metabolic role of the vacuolar ATPase by generating null mutants. Yeast null mutants obtained by disrupting the genes for the A, B, C, E, or c subunit are incapable of growth at neutra1 pH, are strongly inhibited by externa1 calcium ions, fail to carry out endocytosis of lucifer yellow, and exhibit protein mistar- geting (Rothman et al., 1989; Foury, 1990; Hirata et al., 1990; Nelson and Nelson, 1990; Umemoto et al., 1990; Beltran et al., 1992). Because the cells grow normally at pH 5.5, the V-ATPase-minus phenotype is conditionally lethal in yeast (Nelson and Nelson, 1990).

It has not been possible to carry out similar gene replacement experiments in higher eukaryotes; hence, the metabolic and developmental role of the V-ATPase in plant cells has not been directly tested. Plant cells have prominent central vacuoles that occupy more than 90% of the cell volume. One would predict that any perturbation of the osmotic machinery of the

Dow

nloaded from https://academ

ic.oup.com/plcell/article/4/7/851/5984362 by guest on 22 January 2022

852 The Plant Cell

tonoplast would interfere with normal water uptake, therebyinhibiting plant cell expansion and growth. However, plant cellsalso have a second proton pump on the tonoplast, the H+

pyrophosphatase, which potentially could serve as a backupsystem if the V-ATPase were impaired (Rea and Sanders, 1987).

Antisense mRNA has proven to be a useful tool to inhibitthe expression of specific genes in plants (Ecker and Davis,1986; Rothstein et al., 1987; Smith et al., 1988; Van der Krolet al., 1988; Hamilton et al., 1990; Oeller et al., 1991). To probethe function and possible isoforms of the V-ATPase in plants,we transformed carrot root tissue with two antisense constructsof the carrot A subunit cDNA, one corresponding to the entirecoding region (cr) and the other consisting of 150 bp of the5' noncoding sequence (ncr). We expected that the coding re-gion antisense construct would block the expression of allpossible A subunit isoforms, whereas the noncoding antisenseconstruct would be isoform specific. As will be presented, how-ever, both constructs had similar effects, specifically inhibitingthe V-ATPase of the tonoplast. Morphological and anatomicalobservations on the phenotype of the regenerated antisensemutant plants are also described. The biochemical character-ization of the antisense mutants were described by Fichmann(1991).

RESULTS

1 2 3 4 5

i.okb-

-350bp

Figure 1. PCR Amplification of the Two Antisense Constructs fromCarrot Genomic DNA.

Lane 1, DNA standards; lane 2, DNA isolated from leaves of a T1regenerated plant transformed with the cr antisense construct; lane3, DNA from the wild type; lane 4, DNA isolated from leaves of a T2plant transformed with the ncr antisense construct; lane 5, DNA fromthe wild type. Note the presence of the constructs in the transformants(lanes 2 and 4) and their absence from the wild type (lanes 3 and 5).

Detection of Antisense DNA and mRNA by thePolymerase Chain Reaction

To determine whether the genomic DNA of the putative trans-formants contained the antisense DNA, polymerase chainreaction (PCR) analyses were carried out using a sequencefrom the T-DNA within the cauliflower mosaic virus (CaMV)35S promoter as the left primer and the appropriate sequencefrom the coding or 5' noncoding regions of the A subunit asthe right primer (see Methods). As shown in Figure 1, DNAfragments of the expected size were detected. Bands at ~1.0and ~350 bp were specifically amplified using DNA from thecr and ncr cells, respectively. These results confirm that thetransformants contain at least one copy of the antisense-containing T-DNA.

PCR was also used to detect the sense and antisense tran-scripts in the two transformed cell lines, as shown in Figure2A. Lanes 2 and 3 of Figure 2A represent the PCR productswhen the left primer was used in the reverse transcriptase stepto amplify the antisense transcript in the cr and wild-type cells,respectively. A band of the expected length, 350 bp, is pres-ent in lane 2, but is absent from lane 3. Lanes 4 and 5 representthe reciprocal experiment, using the right primer to amplifythe sense transcripts of the two cell lines. Wild-type cellsshowed a prominent band at 350 bp (lane 5), whereas onlytrace amounts of the band were detectable in the cr cells (lane4). These results suggest that the level of sense transcript maybe reduced in the antisense transformant cells.

Similar experiments were carried out with the ncr transfor-mants. As shown in lanes 2 and 3 of Figure 2B, the sensetranscript (150 bp) was readily detected in both the wild-typeand the ncr antisense cells. When probed for the antisensetranscript, only the ncr-transformed cells showed a clear bandat 150 bp (Figure 2B, lane 5); the wild-type DNA yielded multi-ple nonspecific bands (lane 4).

Biochemical Characterization of the Cell Lines

To determine whether the presence of antisense mRNA affectedV-ATPase and proton-pumping activities in the transformants,cells from the four lines were homogenized, and the micro-somal supernatants were centrifuged on step gradients toseparate the tonoplast-enriched (fraction 1) and Golgi-enriched(fraction 3) membranes. As shown in Figure 3, both bafilomycin-sensitive ATPase activity and proton-pumping activity wereinhibited in the tonoplast-enriched fractions of the two antisensetransformants, relative to the wild-type cells. In contrast, theGolgi-enriched fractions of the two antisense mutants hadnormal levels of ATPase activity and elevated levels of proton-pumping activity. The Golgi marker (latent UDPase) was notsignificantly affected by the antisense constructs, indicatingthat the transformed cells are normal with respect to the Golgimarker (data not shown).

Because the ATP-driven H+ pump is known to be involvedin driving secondary transport processes across the tonoplast,

Dow



V-ATPase Antisense Mutants of Carrot 853

we measured the ATP-dependent uptake of 3-O-methyl-D-(U-14C)glucose (14C-OMG) into tonoplast vesicles (fraction 1)isolated from the four cell lines. Figure 4 shows the kineticsof uptake in wild-type, cr, ncr, and control cells. ATP-dependent14C-OMG uptake by wild-type and (3-glucuronidase (GUS)control cells was complete by 40 min, after which a decreasewas observed, probably due to vesicle leakage. The rates ofATP-driven 14C-OMG uptake are similar to those observedpreviously in other systems (Rausch et al., 1987). Uptake by

1 2

A Tonoplast-Enriched Fraction

350bp-

B

-isobp

Figure 2. Detection of Sense and Antisense Transcripts by PCR.(A) PCR amplification of RNA transcripts from carrot wild type andcr calli. Lane 1, 1-kb ladder (Bethesda Research Laboratories); lane2, cr RNA probed for the antisense transcript; lane 3, wild-type RNAprobed for the antisense transcript; lane 4, cr RNA probed for the sensetranscript; lane 5, wild-type RNA probed for the sense transcript. Bandsappearing at 350 bp in lanes 2 and 5 indicate the transcript is present.(B) PCR amplification of RNA transcripts from wild-type and ncr calli.Lane 1,1-kb ladder; lane 2, wild-type RNA probed for the sense tran-script; lane 3, ncr RNA probed for the sense transcript; lane 4, wild-typeRNA probed for the antisense transcript; lane 5, ncr RNA probed forthe antisense transcript. The prominent band at 150 bp (lanes 2, 3,and 5) represents the amplified fragment. Multiple nonspecific bandswere obtained with wild-type DNA (lane 4).

Wl cr ncr Wt cr ncr

• Bafilomycin-sensitive • ..,ATPase activity • H+-P"-"pmg «ctmty

B Golgi-Enriched Fraction400,

300 -

200-

100-

Wt

B Bafilomvcin-sensitive •H+-pumping activity• ATPase activity •

Figure 3. Bafilomycin-Sensitive ATPase and ATP-Dependent H+-Pumping Activities of Membranes Isolated from Cell Lines TransformedEither with the cr or the ncr Antisense Constructs.The data were calculated on a per protein basis and expressed asthe percent of the wild-type (Wt) levels.(A) Tonoplast-enriched fraction.(B) Golgi-enriched fraction.

tonoplast vesicles isolated from the two antisense cell lineswas negligible, in agreement with the inhibition of proton pump-ing activity shown in Figure 3. The slightly higher levels of14C-OMG uptake by the ncr cell line is consistent with theslightly higher levels of H+-pumping activity measured inthese membranes.

Protein Gel Blots

The level of expression of the 69-kD A subunit in the controland antisense cell lines was determined by protein gel blot-ting, using a monospecific antibody raised against a syntheticpolypeptide. In preliminary experiments using total microsomalmembranes, there was no apparent reduction in the level ofA subunit in the antisense transformants relative to wild-typeor control cell lines. To determine whether the tonoplast wasspecifically affected, protein gel blotting was subsequently car-ried out on tonoplast- and Golgi-enriched fractions obtained

Dow



854 The Plant Cell

Figure 4. ATP-Dependent 14C-OMG Uptake in the Tonoplast-Enriched Fractions of the Wild-Type, cr, ncr, and GUS Control Cell Lines.

The control cell lines that were transformed with the GUS gene shownearly wild-type levels of uptake, whereas uptake activity is negligiblein the two cell lines transformed with antisense DNA to the V-ATPaseA subunit.

from step gradients. As shown in Figure 5, the amount of Asubunit in the tonoplast-enriched fractions was much lowerin the two antisense transformants (lanes 4 and 7), as com-pared with the wild type (lane 1). However, A subunit expressionin the Golgi-enriched fraction was either stimulated, in the caseof the cr mutant (lane 6), or unaffected, in the case of the ncrmutant (lane 9), relative to the wild type (lane 3). These resultswere thus consistent with the biochemical assays, in whichit was shown that the tonoplast ATPase and proton pumpingactivities were specifically inhibited by the antisense DNA.

ure 6B), two putative isoforms were observed at ~70 kD, bothof which were in the acidic pi range. The more acidic of thetwo may consist of multiple components. In the wild-type blot,a third protein is visible at a lower molecular mass that isprobably a degradation product because it was not alwaysobserved. Similar gel blots of microsomal membranes of thecr (Figure 6C) and ncr (Figure 6D) transformants were ana-lyzed. Diminished levels of the less acidic isoform wereobserved in both cell lines. In addition, the more acidic iso-form was often lower in the cr cells compared to the ncr cells.

To determine the identity of the inhibited isoform, highly pu-rified tonoplast membranes were purified from carrot rootprotoplasts and analyzed by two-dimensional immunoblotting(see Methods). Figure 6E shows the pattern observed with pu-rified tonoplast membranes. The monospecific antibodyrecognized only a single spot, which aligned with the less acidicof the two A subunit isoforms. Thus, both antisense constructsappear to preferentially inhibit expression of the tonoplast (i.e.,less acidic) isoform.

Phenotypes of the Regenerated Plants

Calli from seven cr cell lines, representing at least seven in-dependent transformation events, were regenerated into wholeplants. Ten T1 plants, one or two from each cell line, were ob-tained, and all exhibited the same basic phenotype. The shootswere characterized by shortened petioles and finely dissected,fernlike leaves. The taproots developed abnormally due to ex-tensive fusion with adjacent plantlets during tissue culture.Attempts to regenerate the ncr cell lines proved more difficult,and only a single plant was obtained. The leaflets of the ncr

1 2 3 4 5 7 8 9

kD70-

Two-Dimensional Gels

To further characterize the inhibition of the A subunit in thetonoplast-enriched fraction, immunoblots of two-dimensionalgels of total microsomal membranes and purified vacuoles werecarried out. Immunoblots of wild-type, GUS control, cr, andncr cells are shown in Figures 6A to 6E. All of the blots aredirectly comparable except for wild type (Figure 6A), in whichthe isoelectric focusing (IEF) separation was performed usingdifferent ampholytes ("Pharmalytes"). The other IEF separa-tions (Figures 6B to 6E) were performed using "Servalytes"(see Methods) over a narrower pH range, which increased thedistances between the protein spots.

In both the wild-type (Figure 6A) and GUS control cells (Fig-

Figure 5. Protein Gel Blot Analysis of Carrot Tonoplast-Enriched andGolgi-Enriched Membranes Fractionated on a Sucrose Step Gradientand Probed with the Monospecific Antibody to the V-ATPase A Subunit.

Lanes 1 to 3, wild type; lanes 4 to 6, coding region antisense transfor-mants; lanes 7 to 9, noncoding region antisense transformants. Lanes1,4, and 7 are the tonoplast-enriched fractions; lanes 2, 5, and 8 con-tain a mixture of tonoplast and Golgi membranes; and lanes 3, 6, and9 are the Golgi-enriched fractions. Fifteen micrograms of protein wasloaded onto each lane.

Dow




IEF

70-

Mr

B

70-

70-

70-

70-

6.5 PH 4.5

Figure 6. Protein Gel Blots of Two-Dimensional Protein Gels of Car-rot Microsomal Membranes Probed with the Monospecific Antibodyto the V-ATPase A Subunit.

(A) Wild-type microsomes. The antibody recognized two putative iso-forms at ~70 kD, which differed slightly in their isoelectric points. Athird polypeptide with a smaller molecular mass was sometimes ob-served below the less acidic isoform and is believed to represent adegradation product.(B) GUS control microsomes. The same two putative isoforms are pres-ent as were observed in the wild type. Note that the apparent differencesin isoelectric points between (A) and (B) through (E) are an artifactof using different ampholytes.(C) Coding region transformant microsomes. The less acidic isoform

T1 plant were smaller, darker, and more lobed than those ofthe cr plants (photographic data not shown).

The T1 plants all flowered and produced viable seeds. Seedswere collected from all T1 plants (in the case of the cr antisenseconstruct), but only the seeds from the plant with the mostextreme phenotype were used for the T2 generation. The prog-eny of the T1 plant transformed with the cr construct displayeda continuous range of phenotypes. Only one of 60 plants waswild type in appearance, whereas the others ranged frommoderately dissected to parental in phenotype, suggesting mul-tiple transformation events. Examples of the opposite extremesin the phenotypes of the T2 generation are shown in Figure7. At the intermediate stage of development shown in Figure7, there is a large difference in the size of the young leavesand taproot. However, the fibrous root system seems to be un-affected in the stunted phenotype. In older plants, the leavesand taproots of the stunted individuals continue to enlarge,approaching the normal plants in size and shape. About 15ncr T2 plants were obtained, all exhibiting the parental mutantphenotype. A comparison of the leaf morphologies of the wild-type, cr, and ncr T2 plants is shown in Figure 8.

It was important to establish that the variation in phenotypeamong the T2 generation of the cr antisense transformantscorrelated with the presence of the antisense cDNA. PCR wasused to amplify the same 1-kb T-DNA fragment from the crconstruct (Figure 1) using DMA isolated from leaves of T2 plants.As shown in Figure 9, the cr construct was undetectable inthe wild-type DMA (Figure 9, lane 2) and in the DNA from theT2 plant with a wild-type phenotype (Figure 9, lane 3). A traceamount was amplified when DNA from an intermediate typewas used (Figure 9, lane 4), whereas prominent bands wereobtained with DNA from two individuals exhibiting the mostextreme phenotype (Figure 9, lanes 5 and 6). The variabilityin phenotype of the T2 plants may be the result of gene dos-age effects arising from multiple transformation events.

Finally, to determine whether the reduction in leaflet sizeof the T2 transformants was caused by an inhibition of cell ex-pansion or cell division, light microscopy was carried out. Asshown in Figure 10, cell expansion in the T2 leaves of the plantstransformed with the cr antisense construct was inhibited byabout 50% at the early stages of development. The outerepidermal cell walls were also thinner and less cuticularized

(arrowhead) is reduced in amount, whereas the more acidic isoformis nearly normal. The more acidic isoform often appeared to separateinto smaller spots, but this effect was not always observed.(D) Noncoding region transformant microsomes. The less acidic puta-tive A subunit isoform (arrowhead) is barely visible, whereas the moreacidic isoform is prominent.(E) Highly purified wild-type tonoplasts. A single polypeptide corre-sponding to the less acidic isoform is present.The proteins in (A) were separated in the IEF dimension using Phar-macia ampholytes, whereas the proteins in (B) through (E) were focusedusing Serva ampholytes (see Methods).

Dow



856 The Plant Cell

Figure 7. Two T2 Plants, Representing Extreme Phenotypes, Obtainedfrom Seeds of a T1 Plant Transformed with the cr Antisense Construct.

The plant on the left resembles wild type with normal-sized leavesand a large taproot (curved because it was grown in a small pot). Theplant on the right was germinated at the same time and grown underidentical conditions. It is characterized by stunted leaves with morefinely dissected leaflets and a shorter taproot.

in the cr transformants. As the leaves approached maturity,cell expansion in the transformants approached that of the wildtype (Figures 10C and 10E), although the blades of the mu-tants retained their dissected appearance. The latter suggeststhat cell division may also be affected.

DISCUSSION

Despite much indirect evidence supporting the function of thetonoplast H+ ATPase in vacuolar solute accumulation, os-motic water uptake, and cell expansion, direct in vivo evidencehas been difficult to obtain. In yeast it has been possible togenerate V-ATPase null mutants by gene replacement, but suchapproaches are not yet feasible in higher eukaryotes. As analternative approach, antisense mRNA has been shown to in-hibit gene expression in higher plants (Ecker and Davis, 1986;

Smith et al., 1988; Van der Krol et al., 1988; Hamilton et al.,1990; Oeller et al., 1991). Although the exact mechanism bywhich antisense RNA inhibits gene expression is poorly un-derstood, it is believed that the antisense transcript hybridizeswith the sense transcript, blocking processing and translation,and/or accelerating turnover (Rothstein et al., 1987; Munroe,1988). Inhibition is rarely 100%, and the degree of inhibitioncan vary in different tissues (Van der Krol et al., 1988).Nevertheless, antisense RNA remains the best tool availablefor inhibiting the expression of specific genes in plants.

Transformation of carrot cells with antisense DMA to the crand 5' ncr of the V-ATPase A subunit cDNA inhibited thebafilomycin-sensitive ATPase, ATP-dependent H+-pumpingand ATP-driven 14C-OMG uptake activities in tonoplast-enriched vesicles. The cr antisense construct inhibited morestrongly (Figures 3 and 4), which is expected because it shouldhybridize with the entire cr of the sense transcript, not just the5' leader sequence. The activity results were consistent withimmunoblots, which showed that the level of the A subunit waslower in the tonoplast fractions of the two antisense mutants.These results indicate that the two antisense cell lines are defi-cient in the tonoplast V-ATPase.

Surprisingly, the Golgi V-ATPase was either unaffected oreven stimulated by the two antisense constructs. This conclu-sion was based on the ATPase and proton-pumping assays,as well as on immunoblot experiments using a monospecificantibody to the A subunit. Because the results were difficultto reconcile with a model based on inhibition of the expres-sion of a single gene, we hypothesized that the tonoplast andGolgi might contain different isoforms. Support for the two-isoform model was obtained by analyzing immunoblots of two-dimensional gels. Microsomal membranes of wild-type andGUS control transformants contained two main isoforms of theA subunit, both in the acidic pi range. Using barley microsomalmembranes, DuPont et al. (1988) also detected two putativeA subunit isoforms in immunoblots of two-dimensional gels.However, we found that highly purified carrot vacuolar mem-branes contained only the less acidic isoform. The persistenceof the more acidic isoform in the antisense transformants andthe high levels of Golgi ATPase activity suggest that the acidicisoform is associated with the Golgi complex.

The finding that both cr and ncr antisense constructs pref-erentially inhibit the more alkaline (tonoplast) isoform is difficultto explain. We predicted that the ncr construct would be iso-form specific because it corresponds to the more variable 5'untranslated region. Because the A subunit is highly conserved(Gogarten et al., 1989), the cr construct would be expectedto inhibit all isoforms. The frequently observed elevation in theactivity of the Golgi ATPase in the antisense transformants sug-gests that it may be up regulated in response to the antisenseinhibition. An up regulation of mRNA levels was not detectedduring the PCR amplification experiments, although it is pos-sible that the PCR primers failed to recognize the Golgi isoformsequence. Further studies are needed to clarify this issue.

Although two-dimensional gels of membrane proteins aresubject to numerous artifacts (Hurkman and Tanaka, 1986),

Dow




Figure 8. Leaf Morphologies of T2 ncr, T2 cr, and Wild-Type Plants at Intermediate and Mature Stages.Left pair, ncr; middle pair, cr; right pair, wild type. The ncr leaves are smaller than those of the cr plants, with lobed rather than dissected leaflets.The leaves of both transformants failed to expand to the wild-type proportions.

3 4 5 6

i.okb-

Figure 9. Detection of the Coding Region Antisense Construct in theGenomic DMA of the T2 Generation.

A 1-kb fragment of the T-DNA containing the coding region antisenseconstruct (see Figure 1, lane 2) was amplified by PCR. Lane 1, DMA

we recently obtained independent evidence for the existenceof multiple A subunit isoforms in carrot. Genomic DNA gel blotssuggest the presence of several putative genes, and genomicfragments representing three different genes for the A subunithave now been cloned and sequenced (Y. Braun and L. Taiz,unpublished data). Although the exons of the three cloned frag-ments are nearly identical, the introns exhibit major differences,suggesting that they are isoforms rather than alleles. Oncecomplete sequences are available, it may be possible to de-sign isoform-specific probes to verify the expression of the threegenes. However, there is already evidence that at least someof the V-ATPase subunits are encoded by multigene familiesin higher eukaryotes. Starke et al. (1991) used PCR amplifica-tion to identify two separate genes encoding the A subunit inthe primitive vascular plants Psilotum and Equisetum, and asmany as four genes encoding the A subunit have been de-tected by DNA gel blots in tobacco (Narasimhan et al., 1991).

standards; lane 2, wild type; lane 3, T2 plant showing wild-type pheno-type; lane 4, T2 plant with intermediate phenotype; lanes 5 and 6, T2plants with the extreme mutant phenotype.

Dow



858 The Plant Cell

o

Figure 10. Light Micrographs of the Leaflets of Wild-Type and T2 Coding Region Antisense Transformants at Comparable Stages of Development.

(A) Wild-type leaflet at an early stage of development. The cross-section shows the midrib area flanked by the two lamina. The epidermal cellshave undergone considerable expansion, and prominent central vacuoles are present.(B) Wild-type leaflet at an intermediate stage of development. Only the edge of the blade, which has continued to expand laterally, is shown.(C) Wild-type leaflet at the mature stage. Bulging around the veins is visible.(D) T2 cr antisense leaflet at an early stage of development. The cr leaflet is only half the size of the wild-type leaflet at a comparable stageof development because cell expansion has been inhibited by 50%. Note that central vacuoles still form in the antisense mutant.(E) T2 cr antisense leaflet at an intermediate stage of development. Lateral proliferation of the lamina has occurred, accompanied by some cellexpansion, but the blade remains narrower than wild type and more compact due to fewer intercellular spaces.(F) T2 antisense leaflet at the mature stage. The cells are fully expanded, and intercellular spaces are evident.Bar = 0.5 mm.

Dow




In contrast, a single gene seems to encode the A subunit of the bovine vacuolar H+ ATPase, which may be a difference between plants and animals (Puopolo et al., 1991). In animals, the B subunit appears to be encoded by a small gene family (Südhof et al., 1989; Bernasconi et al., 1990; Nelson et al., 1992; Puopolo et al., 1992).

It is not yet known whether vacuolar H+ ATPase isoforms are tissue specific or organelle specific. Support for the latter possibility comes from work with Chinese hamster ovary cells selected for resistance to diphtheria toxin (Robbins et al., 1984; Timchaket al., 1986; Stone et al., 1987). The mutation, apparently caused by a single gene, results in defective acidification in endosomes, but not in lysosomes. It is also conditionally lethal when the cells are grown at 39.5OC. Kinetic analyses of the ATPase activity of purified endosomes demonstrated that the ATPase activity and 32P-ATP exchange activity of the mutant enzyme were strongly inhibited by preincubation at 44OC, suggesting that the mutation was in a structural gene for the ATPase (Stone et al., 1987). The fact that the lysosomal proton pump was unaffected suggests that the subunits of the two organelles may be encoded by different genes. In kidney, Wang and Gluck (1990) have shown that the V-ATPase of brush border membranes differs in its kinetic properties from the V-ATPase of kidney microsomes and may contain different B subunit isoforms. More recently, Nelson et al. (1992) reported that a kidney isoform of the B subunit is specifically localized on the plasma membrane of the intercalated cells, suggesting that the B isoform may be both membrane and tissue specific.

Whereas it is often assumed that the tonoplast H+ ATPase drives the osmotic uptake of water into the central vacuole, facilitating cell expansion, the phenotypes of the tonoplast ATPase-deficient antisense mutants presented in this study offer the first direct evidence that this is, in fact, the case. Cell expansion in young leaves and taproots was strongly inhibited in the T2 plants transformed with antisense DNA, suggesting that the tonoplast H+ ATPase may play an important role in these tissues at early stages of development. Although it is tempting to attribute the reduced rates of cell expansion in the antisense transformants to a slower rate of osmotic adjustment, other explanations are possible. For example, cytosolic calcium levels may be elevated because of the reduced uptake by the vacuole. In yeast, the cytosolic calcium concentration is sixfold higher in the V-ATPase null mutants and is believed to be responsible for a variety of pleiotropic effects (Ohya et al., 1991).

The fact that the fibrous root system is apparently unaffected in the T2 plants and that the older leaves and taproots continue to expand at a slow rate and gradually recover to nearly normal size may be due either to residual tonoplast H+ ATPase activity or to the activity of the H+ pyrophosphatase. The difference in the leaf morphologies of the cr versus ncr antisense mutants suggests that caution should be exercised before inferring phenotypic effects of inhibiting genes with antisense DNA. Given the existence of at least three isoforms, it is possible that the ncr construct is more specific for the

tonoplast isoform than the cr construct, which should inhibit all isoforms more dr less equally. The results could indicate that slight alterations in the ratios of the isoforms of key metabolic enzymes may have dramatic effects on morphology.

METHODS

Construction of Binary Ti Plasmids Containing Antisense cDNAs

The plasmids containing the coding (cr) and noncoding (ncr) regions of the carrot A subunit were constructed as shown in Figure 11. For the cr antisense construct, the starting material was the hgtll clone 74 containing the complete cr and the 3' ncr of the cDNA encoding the catalytic 69-kD subunit of the carrot V-ATPase (Zimniak et al., 1988) and only nine nucleotides of the 5' ncr. This short mRNA species is probably not a cloning artifact but is due to a shorter transcript initi- ated at the second TATA box (Zimniak et al., 1988; Struve et al., 1990). After partia1 digestion with EcoRI, the complete cDNA was subcloned into the EcoRl cloning site of pBluescript SK+ plasmid (Stratagene). The Escherichia colistrain used for propagation of plasmids was XL1- Blue (Stratagene). The orientation was such that the noncoding strand was rescued after M13 helper phage infection. The interna1 EcoRl site was removed by oligonucleotide site-directed mutagenesis using the method described by Kunkel (1985).

The major part of the 3' noncoding region, including the poly(A) attachment signal and part of the pBluescript SK+ polylinker, was deleted by Smal digestion and subsequent religation. The cr, including nine nucleotides of the 3' ncr upstream of the Smal site, the 5' ncr, and part of the pBluescript SK+ polylinker were removed using EcoRV to produce a blunt end and Xbal.

This EcoRV-Xbal fragment was then subcloned in the reverse orientation into pGA643 (An et al., 1988), which had been digested with Xbal and Hpal to produce a blunt end.

For the ncr construct, the starting material was the hgtll clone 71 that includes 150 nucleotides from the 5' ncr (Zimniak et al., 1988). After EcoRl digestion, the small fragment containing the 5' untranslated region and the 5'cr was subcloned into pBluescript SK+. This plasmid was then digested with Ddel and blunt ended with the Klenow fragment of DNA polymerase I (there is one Ddel site just before the translation start site). The 800-bp fragment containing the 5' untranslated region and about 650 nucleotides of the pBluescript SK+ plasmid, including the left side of the polylinker, was isolated and digested with Sstl, releasing about 150 bp of the 5' untranslated region plus a por- tion of the pBluescript SK+ polylinker. This fragment was then subcloned into pBluescript SK+ cut with EcoRV and Sstl. Utilizing the Hindlll and Xbal site of the polylinker, the untranslated region was then subcloned in the antisense orientation into pGA643 that was cut with Hindlll and Xbal.

The plasmid pB1121 (Clonetech, Palo Alto, CA) containing the same antibiotic resistance gene and the P-glucuronidase (GUS) gene driven by the cauliflower mosaic virus (CaMV) 35s promoter was used as the control plasmid.

Transformation of Agrobacterium tumefaciens

The pGA643 plasmids containing the cr and the ncr of the V-ATPase 69-kD subunit in the antisense orientation were amplified in E. coli

Dow



860 The Plant Cell

EmRl 1 .

I I I I EcoR SmalEmRl

I 1 I 1

and transformed into various strains of Agrobacterium using the freeze- thaw method as described by An et al. (1988). Agrobacterium strains were used as follows: A281 carrying pTiBo542 (generously provided by G. An of Washington State University, Pullmqn), LBA4404, with the disarmed helper,plasmid pAL4404 (Clonetech), and Agi0 (a kanamycin- sensitive derivative of EHAlOl [Hood, 19861) containing pTiBo542A T-DNA as the helper plasmid (obtained from R. Ludwig, University of California, Santa Cruz).

Carrot Transtormation and Cell Suspenslon Culture

Carrot (Daucus carota) disks were transformed according to a modifi- cation of the procedure described by van Sluys et al. (1987). Carrot root slices, 2-mm thick, were placed on solid Murashige-Skoog medium (Murashige and Skoog, 1962) supplemented with 1% agar, 3% sucmse, 2 mglL I-naphthaleneacetic acid, and 0.3 mglL benzylaminopurine (BAP). A suspension of Agrobacterium cells was pipetted to the cut surface and cocultivated for 3 days in the dark at 25OC. The explants were then transferred to fresh selective medium containing O5 mg/mL

carbenicillin, 0.25 mglmL cefotaxime, and 0.5 mglmL kanamycin. Kanamycin-resistant calli were subcultured in liquid Murashige-Skoog medium supplemented with 3% sucrose, 2 pglmL 2,4-dichlorophenoxy- acetic acid, 0.2 pglmL BAR and 250 pglmL kanamycin and transferred every week to fresh medium. Although the presence of kanamycin did notseem toaffecttheATPaseactivitiesortheamountoftheAsubunit expressed in the transformants, for biochemical analyses the cells were grown in the absence of kanamycin and in the presence of 500 pglmL ampicillin to prevent contamination. The wild-type cells were also grown in the presence of ampicillin.

Control Cell Lines

Two controls were utilized in the biochemical studies: a nontransformed cell line and a cell line derived from root tissue transformed with a disarmed Ti plasmid (pB1121, Clonetech) containing the GUS gene driven by the CaMV 35s promoter, as described by Struve et al. (1990).

Dow




Plant Regeneration and Microscopy

For shoot initiation, calli were transferred to Murashige-Skoog agar medium supplemented as given above with 3% sucrose, 250 pglmL kanamycin, and 2.0 mglL BA!? To induce roots, calli with shoots were transferred to Murashige-Skoog medium supplemented as described above with 2.0 mglL naphthaleneacetic acid. When the young T1 plantlets had regenerated green shoots and roots, they were transferred to sterile soil mix in a Magenta sterile box (Magenta Corp., Chicago). When they reached a height of about 3 to 4 cm, the containers were gradually opened to the ambient air over a 5-day period, during which time they were watered with sterile water and maintained under continuous light (85 pmol m-2 sec-I). After approximately 3 weeks, the young plants were transferred to pots containing soil mix and grown to maturity in the greenhouse. Seeds (the T2 generation) obtained from a single T1 plant, one showing the extreme phenotype, were germinated directly in soil and grown in the greenhouse in pots.

For microscopy, tissue pieces were fixed in 4% glutaraldehyde in 0.1 M sodium cacodylate, pH 7.2. They were embedded in Spurr's resin, sectioned, stained with toluidine blue and basic fuchsin, and pha- tographed using a Leitz Diaplan microscope.

Detection of the Antisense DNA and mRNA by PCR

Genomic DNA was isolated by the miniprep method of Dellaporta et al. (1983), followed by cesium gradient purification according to Maniatis et al. (1982).

The presence of the antisense transcript was detected using the polymerase chain reaction (PCR). Total RNA was isolated from the carrot calli as described by Logemann et al. (1987), resuspended in diethyl pyrocarbonate-treated distilled water, and either digested with RNase-free DNase or purified as follows. After the ethanol precipitation step, the RNA was resuspended in 28 mL of diethyl pyrocarbonate-treated distilled water and was loaded on 9 mL of 5.7 M CsCl and centrifuged at 120,000gfor 24 hr at 22OC. The pellet was washed with 70% ethanol, resuspended in Tris-EDTA, pH 8, containing O.l0/0

SDS, extracted with phenollchloroform, and then precipitated with ethanol. The RNA was analyzed by denaturation with formamide, followed by electrophoresis on a 1% agarose gel containing 5 pg of ethidium bromide (Logemann et al., 1987). cDNA synthesis was carried out using reverse transcriptase and the appropriate PCR primer to produce a cDNAlRNA hybrid as follows: 1 pg of total RNA was added to 2 units of RNase Block II placenta1 RNase inhibitor (Stratagene), 2 pL of a mixture of deoxyribonucleotide triphosphates (10 mM), 2 pL of 10 x reverse transcriptase buffer (500 mM Tris-HCI, 500 mM KCI, 60 mM MgCl2, 1 mglmL BSA, and 10 mM DTT, pH 8.3), and 1 pg of primer. For detection of the antisense transcript, the left primer was used (see below); the right primer was used to detect the sense transcript. The mixture was heated to 90°C for 5 min, and 2 pL of Moloney murine leukemia virus reverse transcriptase (20 units per microliter) (Stratagene) was added, bringing the volume to 20 pL. The tube was placed in a temperature cycling unit (PTC-100; MJ Research Inc., Water- town, MA) programmed for 30 min at 23OC, 1 hr at 42OC (to allow the reverse transcriptase reaction to occur), 10 min at 95OC, and 10 min at 68OC. One microliter of RNase (10 mglmL) was added after the 95OC step so that only the newly synthesized single-stranded cDNA was present during PCR.

PCR was carried out by adding the second primer (1 m), 8 pL of 10 x PCR buffer (500 mM KCI, 13 mM MgCI2, 1 mg/mL BSA, pH 8.3), 2.5 units of Taq polymerase (Cetus Corp., Norwalk, CT), and 70.5 pL

of water directly to the reverse transcriptase reaction mixture. After overlaying with mineral oil, PCR was run at 95OC for 5 min, followed by 30 cycles consisting of a 1-min denaturation step at 95OC, anneal- ing at 37% for 1 min, and extension at 72°C for 3 min. The final extension was at 72% for 20 min. The reaction products were analyzed either on a 1% agarose gel containing 5 pg of ethidium bromide (coding region antisense construct) or on a 16% polyacrylamide gel (noncoding region antisense construct), followed by staining with ethidium bromide.

Primer Design

Genomic DNA

The left primer used to detect the presence of the antisense constructs in genomic DNA was the same for both the cr and ncr mutants. The sequence was taken from the CaMV 35s promoter (5LGCTATCGTT- CAAGATGCCTCTG-37, which should be present only in the transformants. The right primer for the cr construct corresponded to the sense sequence (5'-AACATGCCGGTTGCAGCCAGAGAGGCT- TCAATTTATACAGGAATC-39, which should generate a fragment of -1.0 kb when used with the left primer. The right primer for the ncr (5'- GGAATTCGGGCACTTCTTTAC-39, which corresponds to the sense sequence in positions 1 to 21 of the cDNA (Zimniak et al., 1988), should generate a fragment of -350 bp.

Antisense mRNA

The left primer for detecting the coding region antisense mRNA in the reverse transcriptase step was identical to the region between positions 976 and 996 (GTTTATcrrTGGTTGCGGGGA) of the cDNA (Zimniak et al., 1988). The left primer for detecting the noncoding region antisense transcript corresponded to the sequence between positions 1 and 21 in the 5' noncoding region (5'-GGAATTCGGGCA- CTTCTTTAC-3') (Struve et al., 1990). l h e right primers for detecting the sense transcripts were the complementary sequences of the region between positions 1309 and 1330 (5'-GCACGCTCATAGAAGGACGC-39 and of the region between positions 129 and 150 (B'TCAGATTCAGAT- AAAATCGAT-3') (Zimniak et al., 1988) for the cr and ncr transformants, respectively. The expected lengths of the amplified sequences for the cr and ncr cells were 356 and 150 bp, respectively.

lsolation of Carrot Membranes

Microsomal Pellets

Five hundred milligrams fresh weight of calli or 5 mL of cell suspension (both gave identical results) was homogenized in 1 mL of cold homogenization buffer (250 mM sucrose, 2 mM EDTA, 1 mM DTT, 0.5 mM phenylmethylsulfonyl fluoride [PMSq, 0.1% BSA, 50 mM Tris/Mes, pH 5.8) and pelleted in an Eppendorf microcentrifuge for 1 min at 5000 rpm. The supernatants were then centrifuged in a TL100.2 rotor in a Beckman TL 100 ultracentrifuge at 55 K for 10 min at 4OC. The pellets were resuspended in 300 pL of resuspension buffer (1 mM EDTA, 250 mM sorbitol, 1 mM DTT, 0.5 mM PMSF, 0.25 mM chymostatin, 10 mM TrislMes, pH 7.5). Profein concentration was determined by the Lowry method as previously described (Mandala and Taiz, 1986).

Dow



862 The Plant, Cell

Linear and Step Sucrose Gradients

Cell suspensions (8 g fresh weight of tissue) were homogenized twice with 7.4 mL of homogenization buffer and pelleted (10009 for 5 min), and. the supernatant was layered onto a linear gradient (12 to 43% sucrose in gradient buffer, consisting of 0.5 mM EDTA, 20 mM KCI, 1 mM DTT, 0.5 mM PMSF, in 2.5 mM TridMes, pH 7.5) or onto a step gradient (10 to 18 to 25 to 35% sucrose in gradient buffer). The gradients were centrifuged for 3 hr at 26000 rpm in a rotor (model No. SW 28.1; Beckman). One-milliliter fractions were collected from the linear sucrose gradient, or 1.5-mL fractions were collected from the 10 to 18% (fraction l), 18 to 25% (fraction 2), and 25 to 35% (fraction 3) interfaces of the step gradient. Fraction 1 was the tonoplast-enriched fraction, fraction 3was the Golgi-enriched fraction, and fraction 2 was’ a mixture of tonoplast- and Golgi-enriched membranes.

, .

Purified lonoplasts

Highly purified vacuoles were isolated from carrot taproot phloem by the method of Keller (1988). The purified vacuoles were lysed and pelleted according to the procedure described previously (Taiz and Taiz, 1991).

Monospeciflc Antibody to the A Subunit

Monospecific antibodies raised against a synthetic polypeptide based on a sequence that occurs near the amino terminus (FEDSEKESE- YGYVR) were used to detect the A subunit on immunoblots. The polypeptide was synthesized (Biomolecular Resource Center, University of California, San Francisco) directly onto a branching lysine core as described by Posnett et al. (1988). The resulting branched product can be used directly for antibody production without further coupling to a protein. Rabbit polyclonal antisera to the synthetic peptide was pre- pared as described by Mandala and Taiz (1986), except that subcutaneous rather than intradermal injections were given.

Protein Gel Blots

SDS-PAGE and immunoblots of the membranes from the microsomal pellet or from the sucrose gradient fractions were performed as previously described by Fichmann et al. (1989) using the monospecific antibody described above. For twdimensional gels, microsomal pellets were solubilized using the urea-Nonidet P-40 method according to Hurkman and Tanaka (1986). Sixty microliters of membranes (2 mglmL protein) was solubilized with 150 pL of solubilization buffer consisting of 9.5 M urea, 2% Nonidet P-40,5% P-mercaptoethanol, and 2% ampholytes. lnitial experiments were carried out with ampholytes from Pharmacia (“Pharmalytes”) (Uppsala, Sweden) (1.6% pH 4 to 6.5 and 0.4% pH 3.5 to 10). We subsequently switched to “Servalytes” (Serva, Heidelberg, Germany) (1.6% pH 3 to 10,0.4% pH 5 to 7). In some cases, the protein was first extracted with phenol, followed by Nonidet P-40 solubilization (Hurkman and Tanaka, 1986). Electrophoresis was carried out in a Hoefer SE 250 “Mighty Small” unit. The first dimension was run as follows. After 30 min of prefocusing at 100 V, the samples (12 to 20 pL) were separated by isoelectric focusing (IEF) in a tube gel at 500 V for 3 hr as described by Hurkman and Tanaka (1986). For the second dimension, the tube gel was first equilibrated in sam- ple preparation buffer (2% SDS, 10% glycerol, 125 mM Tris, 1 mg

pyronin Y, pH 6.8) for 5 min and then transferred onto a 10% SDS-polyacrylamide slab gel and sealed with sealing solution (24 mM Tris, 0.2 M glycine, 0.1% SDS, and 05% agarose, pH 8.6). Following electrophc- resis, the protein was electrotransferred onto nitrocellulose paper and probed with the monospecific antibody to the A subunit as described by Fichmann et al. (1989). The pl values were estimated from the positions of the standards (3 to 10 pH range) after silvei staining the two-dimensional gels.

Enzyme and Traansport Assays

ATP-dependent H+ pumping, ATPase, and latent-UDPase activities were measured according to Chanson and Taiz (1985). Vacuolar ATPase (V-ATPase) activity was measured as bafilomycin-sensitive activity (Bowman et al., 1988). ATP-dependent 3-Omethyl-D-(U-14C)glucose (12.06 MBqlmmol, Amersham) uptake in tonoplast vesicles from a step gradient was determined by the method of Rausch et al. (1987).

ACKNOWLEDGMENTS

We thank Dr. Paul Bernasconi for initiating the PCR analysis of antisense mRNA. The cloning vector pGA643 and Agrobacterium A281 carrying pTiBo542 were generously provided by Gynheung An from Washington State University. This research was funded by grants to LI . from the Department of Energy and from the National Science Foundation. J.P.G. was the recipient of a postdoctoral fellowship from the Deutsche Forschungsgemeinschaft.

Received December 19, 1991; accepted May 20, 1992.

REFERENCES

An, G., Ebert, P.R., Mitra, A., and Ha, S.A. (1988). Binary vectors. In Plant Molecular Biology Manual, A3, S.B. Gelvin, R.A. Schilperoort, and D.P.S. Verma, eds (Dordrecht, Belgium: Kluwer Academic Pub- lishers), pp. 1-19.

Beltran, C., Kopecky, J., Pan, Y.-C.E., Nelson, H., and Nelson, N. (1992). Cloning and mutational analysis of the gene encoding subunit C of yeast vacuolar H+-ATPase. J. Biol. Chem. 267, 774-779.

Bernasconi, I?, Rausch, T., Struve, I., Morgan, L., and Taiz, L. (1990). An mRNA from human brain encodes an isoform of the B subunit of the vacuolar H+-ATPase. J. Biol. Chem. 265, 17428-17431.

Bowman, E.J., Siebers, A., and Altendorf, K. (1988). Bafilomycins: A class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proc. Natl. Acad. Sci. USA 85,

Bowman, B.J., Dschida, W.J., Harrls, T., and Bowman, E.J. (1989). The vacuolar ATPase of Neurospora crassa contains F1-like structure. J. Biol. Chem. 264, 15606-15612.

Chanson, A., and Taiz, L. (1985). Evidence for an ATP-dependent proton pump on the Golgi of corn coleoptiles. Plant Physiol. 78,

Dellaporta, S.L., Wood, J., and Hicks, J.B. (1983). A plant DNA

7972-7976.

232-240.

minipreparation. Version II. Plant MOI. Biol. Rep. 1, 19-21. .

Dow




DuPont, F.M., Tanaka, C.K., and Hurkman, W.J. (1988). Separation and immunological characterization of membrane fractions from barley roots. Plant Physiol. 86, 717-724.

Ecker, J.R., and Davis, R.W. (1986). lnhibition of gene expression in plant cells by expression of antisense RNA. Proc. Natl. Acad. Sci.

Fichmann, J. (1991). A comparative study of the vacuolar H+-ATPase of higher plant cells. Ph.D. Dissertation (Santa Cruz, CA: University of California, Santa Cruz).

Fichmann, J., Taiz, L., Gallagher, S., Leonard, R.T., Depta, H., and Robinson, D.G. (1989). lmmunological comparison of the coated vesicles H+-ATPases of plants and animals. Protoplasma 153,

Forgac, M. (1989). Structure and function of vacuolar class of ATP-

Foury, F. (1990). The 31-kDa polypeptide is an essential subunit of 1 the vacuolar ATPase in Saccharomyces cerevisiae. J. Biol. Chem.

Gogarten, J.P., Kibak, H., Dittrich, P., Taiz, L., Bowman, E.J., Bowman, B.J., Manolson, M., Poole, R.J., Date, T., Oshima, T., Konishi, J., Denda, K., and Yoshida, M. (1989). The evolution of the vacuolar H+-ATPase: lmplications for the origin of eukaryotes. Proc. Natl. Acad. Sci. USA 86, 6661-6665.

Hamilton, A.J., Lycett, G.W., and Grierson, D. (1990). Antisense gene that inhibits synthesis of the hormone ethylene in transgenic plants. Nature 346, 284-287.

Hirata, R., Ohsumi, Y., Nakano, A., Kawasaki, H., Suzuki, K., and Anraku, Y. (1990). Molecular structure of a gene, VMAl, encoding the catalytic subunit of the H+-translocating adenosine triphospha- tase from vacuolar membranes of Saccharomyces cerevisiae. J. Biol. Chem. 265, 6726-6733.

Hood, T. (1986). The hypervirulence of Agmbacterium tumefaciens A281 is encoded in a region of pTiBo542 outside of T-DNA J. Bacteriol.

Hurkman, W.J., and Tanaka, C.K. (1986). Solubilization of plant membrane protein for analysis by hvo-dimensional gel electrophoresis. Plant Physiol. 81, 802-806.

Keller, F. (1988). A large scale isolation of vacuoles from protoplasts of mature carrot taproots. J. Plant Physiol. 132, 199-203.

Klink, R., and Lüttge, U. (1990). Electron microscope demonstration of a “head and stalk” structure of the leaf vacuolar ATPase in Mesem- bryanthemum crystallinum L. Bot. Acta 104, 122-131.

Kunkel, T.A. (1985). Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA 82, 488-492.

Logemann, J., Schell, J., and Willmitzer, L. (1987). lmproved method for the isolation of RNA from plant tissue. Anal. Biochem. 163,16-20.

Mandala, S., and Taiz, L. (1986). Characterization of the subunit structure of the maize tonoplast ATPase. lmmunological and inhibitor binding studies. J. Biol. Chem. 261, 12850-12855.

Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982). Molecular Clon- ing: A Laboratory Manual (Cold Spring Harbor, NY Cold Spring Harbor Laboratory).

Munroe, S.H. (1988). Antisense RNA inhibits splicing of pre-mRNA in vitro. EMBO J. 7, 2523-2532.

Murashige, T., and Skoog, F. (1962). A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473-497.

USA 83, 5372-5376.

117-125.

driven proton pumps. Physiol. Rev. 69, 765-796.

265, 18554-18560.

168, 1298-1304.

Narasimhan, M.L., Binzel, M.L., Perez-Prat, E., Chen, Z., Nelson, D.E., Singh, N.K., Bressan, R.A., and Hasegawa, P.M. (1991). NaCl regulation of tonoplast ATPase 70-kilodalton subunit mRNA in tobacco cells. Plant Physiol. 97, 562-568.

Nelson, N. (1991). Structure and pharmacology of the proton-ATPases. Trends Pharmacol. Sci. 12, 71-75.

Nelson, N., and Taiz, L. (1989). The evolution of H+-ATPases. Trends Biochem. Sci. 14, 113-116.

Nelson, H., and Nelson, N. (1990). Disruption of genes encoding subunits of yeast vacuolar H+-ATPase causes conditional lethality. Proc. Natl. Acad. Sci. USA 87, 3503-3507.

Nelson, R.D., Guo, X.-L., Masood, K., Brown, D., Kalkbrenner, M., and Gluck, S. (1992). Selectively amplified expression of an isoform of the vacuolar H+-ATPase 56-kilodalton subunit in renal intercalated cells. Proc. Natl. Acad. Sci. USA 89, 3541-3545.

Oeller, P.W., Min-Wong, L., Taylor, L.P., Pike, D.A., and Theologis, A. (1991). Reversible inhibition of tomato fruit senescence by antisense RNA. Science 254, 437-439.

Ohya, Y., Umemoto, N., Tanida, I., Ohta, A., lida, H., and Anraku, Y. (1991). Calcium-sensitive cls mutants of Saccharomyces cerevisiae showing Pet- phenotype are ascribable to defects of vacuolar membrane H+-ATPase activity. J. Biol. Chem. 266, 13971-13977.

Posnett, D.N., McGrath, H., and Tarn, J.P. (1988). A nove1 method for producing antipeptide antibodies. Production of site-specific antibodies to the T cell antigen receptor P-chain. J. Biol. Chem. 263,

Puopolo, K., Kumamoto, C., Adachi, I., and Forgac, M. (1991). A single gene encodes the catalytic ‘A” subunit of the bovine vacuolar H+-ATPase. J. Biol. Chem. 266, 24564-24572.

Puopolo, K., Kumamoto, C., Adachi, I., Magner, R., and Forgac, M. (1992). Differential expression of the “ B subunit of the vaccolar H+-ATPase in bovine tissues. J. Biol. Chem. 267, 3696-3706.

Rausch, T., Butcher, D.N., and Taiz, L. (1987). Active glucose transport and proton pumping in tonoplast membranes of Zea mays L. coleoptiles are inhibited by anti-H+-ATPase antibodies. Plant Phys- iol. 85, 996-999.

Rea, P.A., and Sanders, D. (1987). Tonoplast energization: Two H+ pumps, one membrane. Physiol. Plant. 7l, 131-141.

Robbins, AR., Oliver, C., Bateman, J.L., Krag, S.S., Galloway, C.J., and Melman, 1. (1984). Asingle mutation in Chinese hamster ovary cells impairs both Golgi and endosomal functions. J. Biol. Chem.

Rothman, J.H., Yamashiro, C.T., Raymond, C.K., Kane, P.M., and Stevens, T.H. (1989). Acidification of the lysosome-like vacuole and the vacuolar H+-ATPase are deficient in two yeast mutants that fail to sort vacuolar proteins. J. Cell. Biol. 109, 93-100.

Rothstein, S.J., DiMaio, J., Strand, M., and Rice, D. (1987). Stable and heritable inhibition of the expression of nopaline synthase in tobacco expressing antisense RNA. Proc. Natl. Acad. Sci. USA 84, 8439-8443.

van Sluys, M.A., Tempe, J., and Federoff, N. (1987). Studies on the introduction and mobility of the maize Activator element in Arabidopsis thaliana and Daucus carota. EMBO J. 6, 3881-3889.

Smith, C.J.S., Watson, C.F., Ray, J., Blrd, C.R., Morris, P.C., Schuch, W., and Grierson, D. (1988). Antisense RNA inhibition of polygalac- turonase gene expression in transgenic tomatoes. Nature 334,

1719-1725.

99, 1296-1308.

724-726.

Dow



864 The Plant Cell

Starke, T., Linkkila, T.P., and Gogarten, J.P. (1991). Twoseparate genes encode the catalytic 70 kDa V-ATPase subunit in PMotum and Equisi- rum. Z. Naturforsch. 46c, 613-620.

Stone, D.K., Marnell, M., Yang, Y., and Draper, R.K. (1987). Ther- molabile proton translocating ATPase and pump activities in a clathrin-coated vesicle fraction from an acidification defective Chi- nese hamster cell lhe. J. Biol. Chem. 262, 9883-9886.

Struve, I., Rausch, T., Bernasconi, P., and Taiz, L. (1990). Structure and function of the promoter of the carrot V-type H+-ATPase catalytic subunit gene. J. Biol. Chem. 265, 7927-7932.

Siidhof, T.C., Fried, V.A., Stone, D.K., Johnston, P.A., andXie, X.S. (1989). Human endomembrane H+ pump strongly resembles the ATP-synthase of archaebacteria. Proc. Natl. Acad. Sci. USA 86, 6067-6071.

Taiz, S.L., and Taiz, L. (1991). Ultrastructural comparison of the vacuolar and mitochondrial H+-ATPases of Daucus carota. Botanica Acta. 104, 117-121.

Timchak, L.M., Kruse, F., Marnell, M.H., and Draper, R.K. (1986). A thermosensitive lesion in a Chinese hamster cell mutant causing differential effects on the acidification of endosomes and lysosomes. J. Biol. Chem. 261, 14154-14159.

Umemoto, N., Yoshihisa, T., Hirata, R., and Anraku, Y. (1990). Roles of the VMA3 gene product, subunit c of the vacuolar membrane ATPase on vacuolar acidification and protein transport. J. Biol. Chem.

Van der Krol, A.R., Lenting, P.E., Veenstra, J., Van der Meer, I.M., Koes, R.E., Gerats, A.G.M., MOI, J.N.M., and Stuitje, A.R. (1988). An anti-sense chalcone synthase gene in transgenic plants inhibits flower pigmentation. Nature 333, 866-869.

Wang, Z.-Q., and Gluck, S. (1990). lsolation and properties of bovine kidney brush border vacuolar H+-ATPase. A proton pump with enzymatic and structural differences from kidney microsomal H+- ATPase. J. Biol. Chem. 265, 21957-21965.

Zimniak, L., Dittrich, P., Gogarten, J.P., Kibak, H., and Taiz, L. (1988). The cDNA sequence of the 69-kDa subunit of the carrot vacuolar H+-ATPase. Homology to the 0-chain of FoF1-ATPases. J. Biol. Chem. 263, 9102-9112.

265, 18447-18453.

Dow



Documents

The Use of Antisense mRNA to lnhibit the Tonoplast H+ ATPase in Carrot