Hormonal and Environmental Responsiveness of a ...

3 downloads 5287 Views 6MB Size Report
repair system for damaged proteins accumulating in aging and stressed seeds ... 1985; Aswad and Johnson, 1987; Barten and O'Dea, 1990;. Lowenson .... chlorofodisoamyl alcohol, to obtain a protein-free aqueous phase. To precipitate ... tionated by electrophoresis in a 1% agarose gel containing 17% formal- dehyde as ...
Vol. 269, No. 41, Issue of October 14, pp. 25605-25612, 1994 Printed in U.S.A.

CHEMISTRY THE JOURNAL OF BIOLOCICAI 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Hormonal and Environmental Responsiveness of a Developmentally Regulated Protein Repair L-Isoaspartyl Methyltransferasein Wheat* (Received for publication, May 19,

1994, and in revised form, August 10,

1994)

Mary Beth Mudgett and Steven Clarke From the Department of Chemistry and Biochemistry and The Molecular Biology Institute, University of California, Los Angeles, California 90024-1569

The L-isoaspartyl protein methyltransferase (EC 1985; Aswad and Johnson, 1987; Barten and O’Dea, 1990; 2.1.1.77) has been proposedto be involved inthe repair Lowenson and Clarke, 1992; Mudgett and Clarke, 1993). of spontaneously damaged proteins by facilitating the The generation of L-isoaspartyl and D-aspartyl residues in conversion of abnormal L-isoaspartylresidues to normal polypeptides arises from the spontaneousdeamidation, isomerL-aspartyl residues. Based on the abundance of this en- ization, and racemization of L-asparaginyl and L-aspartyl resizyme in the seeds of a variety of plants and its unique dues (Geiger and Clarke, 1987; Stephenson and Clarke,1989). substrate specificity, it has been hypothesized that it The presence of these abnormal aspartyl residues in modified functions to prevent the accumulation of abnormal asproteins can inhibit their catalytic and functional properties partyl residues in the proteins of aging seeds that can limit the viability of the embryo or its chances for ger- (Johnson et al., 198713; George-Nascimento et al., 1990). The mination. In this work, we show that the expression of enzymatic recognition and methylation of L-isoaspartyl resibeen shown in vitro to be the the L-isoaspartyl methyltransferase is under develop- dues by the methyltransferase has mental regulation in the winter wheat, Diticurn aesti- first step in a protein repair pathway that can lead to the vum. Methyltransferase mRNA and active enzyme are conversion of abnormal L-isoaspartyl residues to normal L-asfirst detected in seeds during the late stages (111-IV)of party1 residues (McFadden and Clarke, 1987; Johnson et al., caryopsisdevelopment. As mature seeds germinate, 1987a, 1987b; Galletti etal., 1988; Lowenson andClarke, methyltransferase mRNA levels decline and are nearly 1991). These results suggest that the repair of L-isoaspartyl undetectable by 72 h post-imbibition. Enzyme activity damage in proteins in vivo is possible. Genetic evidence to remains constant for 24 h post-imbibition and then desupport such a role for this enzyme comes from the analysisof creases rapidly following the reduction of its corresponding mRNA. Methyltransferase activity is very low the enzyme in Escherichia coli.Mutants of these bacterialcells or undetectable in wheat seedlings, including leaf and lacking the L-isoaspartyl methyltransferase ( p c m - ) survive root tissues. We show, however, that the L-isoaspartyl poorly in stationary phase or under thermal stress (Li and methyltransferase can be induced in vegetative tissues Clarke, 1992). These results suggest that the long term surin response to hormone treatment and environmental vival of E. coli cells requires that the methyltransferase prestress. Abscisic acid, a phytohormone involved in seed vent the accumulation of potentially labile or inactive proteins development and desiccation tolerance, induces both containing L-isoaspartyl residues. methyltransferase mRNA and enzyme activity in 4-dayIn plants, a protein L-isoaspartyl methyltransferase activity old wheat seedlings. Dehydration and salt stress also has been identified in both the monocots and thedicots as well induce its transcription and enzymatic activity in seed- as the green algae,Chlamydomonas reinhardtii (Mudgett and lings. The ability of a plant to regulate methyltrans- Clarke, 1993). Although some enzymatic activity is present in ferase activity in its seeds and vegetative tissues in re- most plant organs, the levels can vary considerably. Interestsponse to desiccation, aging, and environmental stress ingly, the highest level of L-isoaspartyl methyltransferase acmay allowthe plant to efficiently repair protein damage tivity is found in seeds (Mudgett and Clarke,1993). Moreover, associated with these physiological changes. the in vitro formation of carboxyl methylated proteins in the soluble fraction of seeds in the absence of exogenous peptide substrates suggests thatmethyl-accepting substrates exist for The protein-L-isoaspartate baspartate) O-methyltrans- the methyltransferasein vivo (Trivedi et al.,1982; Mudgett and ferase (EC 2.1.1.77) from the winter wheat !Piticum aestiuum Clarke, 1993). Based on the localization and unique substrate has been recently proposed t o participate in an intracellular specificity of the protein L-isoaspartyl methyltransferase, we repair systemfor damaged proteins accumulating in aging andhypothesize that one role of this enzyme in plants is to repair stressed seeds (Mudgett and Clarke, 1993). This cytosolic en- protein damage in seeds. We reasoned that the problem of zyme, abundant in plantseeds and many othereucaryotic and spontaneous protein damage may be particularly severe in procaryotic cells and tissues,catalyzes the transferof a methyl seeds that must survive for extended periods of time in a dorgroup from S-adenosylmethionine to the a-carboxyl group of mant state and then must rapidly develop into a functional L-isoaspartyl residues (as well as the P-carboxyl groupof plant. It is likely that, if unrepaired, seeds will accumulate t o the proteins containing racemized and isomerized aspartyl resio-aspartyl residues in mammalian cells) but not P-carboxyl group of normal L-aspartyl residues (Clarke, dues that may be inactive or more susceptible to inactivation. The viability of the embryo during dormancy maythus depend * This work was supported by National Institutes of Health Grant on the presence of the methyltransferase. GM 26020 and National Science Foundation Grant MCB9305405. The In this work, we show that the wheatL-isoaspartyl methylcosts of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- transferase is primarily a seed-specific protein whose mRNA expression and enzymatic activity is up-regulated in developtisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ing seeds and then down-regulated in germinating seeds. In

25605

25606

Methyltransferase Wheat Repair Protein in

addition, we show that this enzyme is hormonally induced in SDS, pH 8.0), the slurry was vortexed and ground further using a vegetative tissue inresponse to exogenous treatment with ab- Polytron homogenizer (Model PT 2000, 7-mm generator, Brinkmann scisic acid (ABA),l a phytohormone involved in caryopsis devel- Instruments) at 15,000-20,000 rpm for approximately 15 s. Chloroform/ opment and desiccation tolerance(Black, 1991; Bray, 1991; isoamyl alcohol(3 ml, 24:l (v/v))was added, vortexed, and pulsed again for 15 s with the Polytron homogenizer. The aqueous phase was reQuatrano et al., 1992). Finally, we provide clear evidence that moved following centrifugation for 5 min at 14,000 x g and extracted the levels of L-isoaspartyl methyltransferase mRNA and enzy- twice, once with 2 volumes of hot extraction buffer and then once with matic activity are induced in wheat seedlings exposed to dehy- one-half volume of extraction buffer and one-third volume of dration and salt stress. The regulation of this methyltranschlorofodisoamyl alcohol, to obtain a protein-free aqueous phase. To precipitate the RNA, the aqueous phase was mixed with an equal volferase in wheat is consistent with its proposed role in the maintenance of spontaneously damagedproteins in aging seedsume of 4 M LiCl and stored at -20 "C overnight. RNA was collected by centrifuging at 14,000 x g for 30 min at 4 "C. The resulting pellet was and environmentally stressed plants.

resuspended with 50-100 pl of water, and reprecipitated with 0.1 volume of 3 M sodium acetate, pH 5.2, and 2 volumes of ethanol at -20 "C MATERIALSANDMETHODS overnight. The RNA pellet was again collectedby centrifugation, Plant Material-Seeds from winter wheat (T aestiuum cultivar Au- washed with 70% ethanol, and dried in a Savant SpeedVac apparatus. Ten-pg samples of total RNA and RNA standards (Promega)were fracgusta) were provided by Dr. Robert Forsberg of the University of Wisconsin (Madison, WI). Developing wheat heads (T aestivum cultivar tionated by electrophoresis in a 1%agarose gel containing 17% formalStevens) grown at Spillman Agronomy Farm by Dr. R. E. Allan (USDA- dehyde as described by Lehrach et al. (1977).Following electrophoresis, ARS, Washington State University, Pullman, WA) were collectedby Dr. the RNA was transferred onto a Genescreen nylon membrane (NEN M.K. Walker-Simmons (USDA-ARS, Washington State University). Research Products) by the Southern procedure (Sambrook et al., 1989). The efficiency ofRNA transfer was determined as described by the The wheat heads were collected, temporarily placed on ice, and then manufacturer by wetting the membrane in 150 m~ NaCl, 10 mM sodium frozen in liquid nitrogen and stored at -20 "C. Immature and mature phosphate, pH7.4, and 1 mM sodium EDTAfollowedby staining in seeds were manually isolated from these wheat heads for analysis. Growth ofPlants for Developmental Analysis-Wheat seeds (cultivar 0.04% methylene blue, 0.5 M sodium acetate, pH5.2, until the RNA Augusta) were surface-sterilized with 10% sodium hypochlorite for 10 bands became visible. The membrane was then destained in 150 mM min at room temperature and were then rinsed with sterile water. For NaCl, 10 m~ sodium phosphate, pH 7.4, and 1 I"sodium EDTA with several changes of the solution to reduce the background staining. Preseedling samples collected between days 0 and 7, seeds were sown on 150-mm Petri dishes containing approximately 150 mlof 0.7% agar. hybridization of the membrane was performed for 3 h at 65 "C with Otherwise, seeds were sown in APS-6 pots (Accelerated Propagation Biodyne solution (1%bovine serum albumin, 0.5 M sodium phosphate, 1 System, Gardener's Supply Co., Burlington, VT) containing germinat- mM sodium EDTA, 7% SDS, pH 7; Church and Gilbert (1984)).Hybriding mixture (Gardener's Supply Co.). Wheat seedlings were grown in ization was then performed overnight a t 65 "C by incubating the memthe laboratory under 24 h of continuous light (-2OMOO pE/m2 s, Sun- brane with denatured probe and fresh solution. Membranes were probed with a 549-bppolymerase chain reaction product encoding light tubes, Gardener's Supply Co.) until day 7 of growth and then nucleotides 171-719 of the 952-bpL-isoaspartyl methyltransferase switched to 16 h of light and 8 h of darkness. Organs were harvested, cDNA in clone pMBMl (Mudgett and Clarke, 1993) and/or the PstI frozen in liquid nitrogen, and stored at -80 "C. insert from the cDNA clone p1015 containing the entire coding region Preparation of Plant Cytosol4rude cytosol was prepared from the wheat samples by homogenization using a mortar and pestle and a for the wheat Em polypeptide (Litts et al., 1987)provided by Dr. Ralph Polytron homogenizer (Brinkmann Instruments). In a liquid nitrogen- Quatrano of the University of North Carolina, Chapel Hill, NC. The probes were randomly labeled to a specific activity of 2 x lo9 c p d p g chilled mortar, frozen plant tissue (typically, 5-10 g of vegetative tissue with [ C X - ~ ~ P I ~(ICN C T PBiomedicals, Irvine, CA) using pd(N), random or 3 g of seeds) was crushed with a pestle, and then transferred to a deoxynucleosides(Pharmacia Biotech Inc.)and the Klenow fragment of 50-ml polypropylene conicaltube containing extraction buffer (10 ml of membranes were 100 mM sodium HEPES, pH 7.5, 1mM dithiothreitol, 1 p~ leupeptin, 1 DNA polymerase I (LifeTechnologies,Inc.).The mM phenylmethanesulfonylfluoride, 10 mM sodium hydrosulfite, and 10 washed twice in 75 mM NaC1,82.5 m~ sodium citrate, pH 7.0, and 0.1% mM sodium metabisulfite at 4 "C; Anderson and Rowan (1967)). The SDS at 55 "C for 30min, followed by twosimilar washes in 15 mM NaCl, 16.5 m~ sodium citrate, pH 7.0, and 0.1% SDS. Then the membranes slurry was kept on ice and then ground further using a Polytron homogenizer (ModelPT 2000, 7 mm generator) at 15,000-20,000 rpm for were autoradiographed by exposure to Kodak XAR-5 film at -80 "C with two30-s intervals. The resulting crude homogenate was filtered two intensifylng screens. Stress Deatment of Wheat Seedlings-Wheat seeds (cultivar Authrough four layers of cheesecloth and thencentrifuged at 2,200 x g for 30 min at 4 "C to remove cell debris. The supernatant was centrifuged gusta) were surface-sterilized and then sown on a piece of Whatman further at 172,200 x g for 60 min at 4 "C to remove membranous ma- filter paper (type 1)placed on a 150-mm Petri dish containing approxiterial and then filtered through one layer of Miracloth (Calbiochem)to mately 150 ml of 0.7% agar. Following germination under continuous remove the floating lipid layer.The resulting supernatant, identified as light (-200 pE/m2 s), plates were uncovered and the seedlings were the soluble fraction, was stored at -20 "C and utilized as the source of watered daily. After 4 days of growth, the seedlings were exposed to different 10-h treatments. Seedlings were exposed to water stress by methyltransferase. transferring them to a new Petri dish containing no agar or water unMethylation Assay-Methyltransferase assays were performedat pH 7.5 using 500 p~ Val-Tyr-Pro-L-isoksp-His-Ala as a methyl-acceptor der continuous light. Other seedlings were exposed to low and high exactly as described (Mudgett and Clarke, 1993). Endogenous cytosolic temperature stress by placing them in a cold room (4 "C) under continpolypeptides are also potential methyl-acceptors;therefore, parallel ex- uous light or in a 37 "C incubator without light, respectively. To subperiments were conducted in the absence of the peptide substrate. For ject seedlings to abscisic acid treatment and to salt stress, seedlings each soluble fraction assayed, L-isoaspartyl methyltransferase specific were transferred to a new dish and grown in solution (20 ml) containactivity was calculated by subtracting thepeptide-independent (endog- ing 50 p~ ABA ((+)-cis,trans-abscisic acid) (Sigma) in water, or 0.25 M enous) activity, typically less than 0.4 pmoYmidmg, from the peptide- NaCl in water, respectively, and received continuous light. Control seedlings were grown on moist agar plates or in a Petri dish containdependent activity. All assays were done in triplicate. Protein Determination-A modification of the Lowry procedure (Bai- ing water. Following treatment, the seedlings were frozen in liquid niley, 1967) was used to determine the concentration of protein after trogen and then assayed for methyltransferase mRNA and enzymatic activity. precipitation with 1ml of 10% (wh) trichloroacetic acid. RNA Isolation and Northern Analysis-Total RNA was isolated from the wheat samples as described byVenvoerdet al. (1989) with the RESULTS following modifications. Frozen wheat tissue (0.5 to 2.0 g) was ground L-Isoaspartyl Methyltransferase Activity in Wheat Seedlings with a pestle to a fine powder in a liquid nitrogen-cooled mortar and seedlings were harvested over a then transferred to a 50-ml polypropylene conicaltube. After the addi- (Cultivar Augusta)-Wheat tion of 5 mlof hot (80 "C) extraction buffer (molecular biology grade period of 35 days to obtain organs at various stages of develphenol equilibrated with 100 mM Tris-HC1, 10 m~ sodium EDTA, 1% opment. The soluble fractions from whole seedlings, leaves,and The abbreviations used are: ABA, (+)-cis, trans-abscisic acid; bp, base pair(s); Em, early methionine-labeled.

roots were assayed for methyltransferase activity using ValTyr-Pro-isoAsp-His-Ala as a methyl-accepting substrate. This peptide has been shownto be a substrate for the purified wheat

25607

Protein Repair Methyltransferasein Wheat A

0.030 A 0.025

t

0.020 0.01 5 0.01 0

+

0.005 seeds

4

seedlings

0

7

14

21

28

35

Days Post-Imbibition FIG.1. L-Isoaspartyl methyltransferase activity in developing wheat seedlings (cultivar Augusta). A, totalpeptide-dependent B , specific peptide-dependent methyltransferaseactivityperplant. methyltransferaseactivityinseeds (filled circle),seedlings (open circles), roots (open triangles), and leaves (filled triangles).

SeedlingLeaf

Root Seed

Specific Activity(pmoVmin/m@ 0.62 0.30 0.47 4.36

RNA Std DNA Probe -~ - 1898 L-lsoaspartyl + Methyltransferase cDNA - 872

- 872 4-

Em cDNA

FIG.2. Organ specificity of L-isoaspartylmethyltransferase mRNA and enzymatic activity levels in wheat in comparison with that ofthe Em polypeptide mRNA. The levels of methyltransferase specific activity, methyltransferase mRNA, and EmmRNA in wheat seeds (cultivar Augusta) and 5-day-old wheat seedlings, leaves, and roots were determined. Seedlings were grown as described under “Materials and Methods,” harvested, and then frozen in liquid nitrogen. Soluble fractions were isolatedfrom the tissues and assayed for methyltransferase activity and total protein a s described under “Materials and Methods” (A ). Methyltransferase-specific activity represents the average value obtained from two independent experiments. Northern analysis was performed a s described under “Materials and Methods” using a radiolabeled 549-bp polymerase chain reaction product coding for the central portion of the 952-bp wheat L-isoaspartyl methyltransferase cDNA (Mudgett and Clarke, 1993) (A). The membrane was then stripped and rehybridized with a radiolabeled PstI fragment from the clone p1015 coding for the Em cDNA (length of mature RNA is 780 bp, Litts etal. (1987))( B).RNAmarkers (9488,6225,3911,2800,1898,872, 562, and 363 nucleotides) synthesizedin uitro from specific templates (Promega) were used to standardize migration of the mRNAs. The autoradiographs showing the level of methyltransferase mRNA and Em mRNA were exposed for 10 and 2 days,respectively.

germ enzyme with a K,,, of 52 VM (Mudgett and Clarke,1993). We found that peptide-dependent L-isoaspartyl methyltransferase activity washighest in mature wheat seeds and that the activity was significantly reducedfollowing imbibition and germination (Fig. 1).After 7 days of growth, methyltransferase mRNA was in fact identical to that of the methyltransferase was undetectable in roots and was present at low levels in mRNA, i t was present in seeds and absent 5-day-old in seedling We found a significant tissues (Fig. 2B). leaves and whole seedlings (Fig. B). decrease in both the specific activity as well as the total activity Levels of L-Isoaspartyl Methyltransferase mRNA and Protein of methyltransferase per plant following imbibition (Fig. 1). Decline after Germination-To determine when methyltransThis latter result suggests that active methyltransferase pres- ferase expression is down-regulated in developing plants, the ent in a mature dry seed is inactivated or degraded once the steady-state levels of L-isoaspartyl methyltransferase mRNA seed begins to germinate. and active enzyme weremeasured in germinating wheat seeds. Organ-specific Expression of L-Isoaspartyl Methyltransferase Wheat seeds germinated under continuous illumination were mRNA in Wheat Seeds a n d Seedlings-We then compared the analyzed following the initialimbibition of the seeds. A Northlevel of methyltransferase mRNA and active protein in 5-day- ern blot of total RNA analyzed with the methyltransferase old wheat seedlings, leaves,and roots, and in mature dry seeds. probe as described above shows that methyltransferasemRNA, present as a single 1200-nucleotide species, decreases after 24 Total RNA was isolated and subjected to Northern analysis por- h of germination (Fig. 3 A ) . Under ourgrowth conditions, methusing a radiolabeled fragment corresponding to the central tion of the coding region of the wheat L-isoaspartyl methyl- yltransferase mRNA is nearlyundetectable in seedlings at 72 h transferase cDNA (Mudgett and Clarke, 1993). Northern anal- post-imbibition. Methyltransferase enzymatic activity remains ysis shows that methyltransferase mRNA was expressed as a relatively constant in the first 24 h of germination and then single 1200-nucleotide species only in seeds and not in whole decreases rapidly by 48 h post-imbibition (Fig. 3B ). The subseseedlings, leaves, or roots (Fig. 2 A ) . Moreover, a high level of quent decline of methyltransferase specific and totalenzymatic methyltransferase activity was detected only in the soluble activity follows the decline of its mRNA. When the sameNorthfraction of seeds expressing methyltransferasemRNA. One in- ern blot was reprobed with a radiolabeled fragment coding for terpretation of these results is that little, if any, synthesis of the Em cDNA, the mRNA was completely absent by 24 h postmethyltransferase occurs in 5-day-old seedlings and that the imbibition (Fig. 3A). The pattern of mRNA loss was similar to work published earlier (Cuming, 1984)except in that study the enzyme present at that stage was translated earlier in the development of the plant. Em mRNA was completely absent by 6 h post-imbibition. The The pattern of methyltransferase enzyme and mRNA expres- slower rate of Em mRNA disappearance in our studymay resion seen here in seeds and 5-day-old seedlings is similar to flect cultivar variationas well as thefact that we analyzed the that previously demonstrated for the Em (early methionine- germination of whole seeds andnot the germination of isolated labeled)protein in wheat. The Em polypeptide is the most embryos. abundant translationproduct in wheatembryos and is primar- Formation of L-Isoaspartyl Methyltransferase mRNA and ily seed-specific (Grzelczak et al.,1982). After germination, the Protein during Seed Development-The expression of seedlevels of the Emprotein and mRNA rapidly drop (Grzelczak et specific proteins is known to be regulated a t different stages of al., 1982; Cuming, 1984; Williamson et al., 1985). We thus seed maturation (Goldberg et al., 1989; Lane, 1991). To deterprobed the Northern blot with a radiolabeled fragment of the mine when the methyltransferase is expressed in wheat, we Em cDNA. We found that theexpression of the Em polypeptide measured enzyme and mRNA levels at different stages of car-

Protein Repair Methyltransferase Wheat in

25608

the level of methyltransferase activity corresponded with the rise in thelevel of mRNA present in stages11-IV (Fig. 4, A and B 1. We note that thelevel of methyltransferase mRNA in stage 26 S V seeds from the Stevens cultivar (Fig. 4A) was lower than the rRNA -2800 level detected in dry seedsfrom the Augusta cultivar (Fig. 2A ). The reason for this difference is not clear at present, although the methyltransferase activity in seeds of both cultivars was similar. The temporalinduction of methyltransferase mRNA and en872 zymatic activity in maturing seeds during lateembryogenesis demonstrates that the development of this methyltransferase 0 1 2 4 6 8 12 24 48 72 (hr) is similar to thatof late embryogenesis abundant proteins, or lea proteins (Dure et al., 1989; Dure, 1993). Lea proteins have been categorized according to the temporal appearance and B. disappearance of their mRNA and protein during seed matu2.0 ration and germination and according to their biased amino acid compositions and sequence homologies. The importantdistinction betweenthis methyltransferase andall of the otherlea 1.5 proteins described to date is that the protein sequence of the . 2I > .methyltransferase (Mudgett and Clarke, 1993) does not con.- C tain anyof the conserved amino acid sequence domainsshared 1.0 25 among the lea proteins in higher plants (Dure et al., 1989). mo Z E Thus, the L-isoaspartyl methyltransferase appears to be the first example of a seed protein with temporal expression char0.5 acteristics similar to lea proteins but that possesses a distinct enzymatic activity. 0.0 Hormonal and Environmental Stress Induction of L-Isoas1 2 4 6 a 12 24 4'8 72 party1 Methyltransferase in Seedlings-Because the phytohorTime Post-Imbibition(hr) mone ABA has been shown to induce the expression of a variety FIG.3. L-Isoaspartyl methyltransferase mRNA and enzymatic of lea proteins includingthe Empolypeptide (Galau et al., 1986; activity in germinating wheat plants. Wheat seeds (cultivar Au- Williamson et al., 1985; Berge et al., 1989), we were interested gusta) were germinated on 0.78 agar plates under continuous illumi- in determiningif the L-isoaspartyl methyltransferase isresponnation. At various times post-imbibition, tissue was frozen in liquid nitrogen and then analyzed for methyltransferase mRNA and enzy- sive to thisphytohormone as well. We noted that theinduction matic activity a s described in the legend to Fig. 2. A, the steady-state of methyltransferase mRNA expression and enzymatic activity level of methyltransferase mRNA (MT) and Em mRNA ( E m )in wheat in developing seeds shown in Fig. 4 coincides with an elevated plants post-imbibition (h). The relative amount of total RNA in each concentration of endogenous ABA expected during thisprocess lane is representedby the methylene blue staining of the 26 S ribosomal "Ma- (Quatrano, 1986; Black, 1991). Furthermore, the endogenous RNA band on the membraneperformed prior to hybridization (see terials and Methods"). The autoradiographs showing level the of meth- level ofABA is increased not only in seeds during caryopsis yltransferase mRNA and Em mRNA were exposed for 9 and 2 days, development but also in vegetative tissues in response to envirespectively. B , total methyltransferase activity per plant (dotted col- ronmental stress (Zeevaart and Creelman, 1988; Bray, 1991). umns) and specific methyltransferase activity (black columns). In some cases, the transcription andor translation of stressrelated proteins have been shown to be regulated by ABA(Bray, yopsis development. Developing wheat heads (cultivar 1991; Quatrano et al., 1992). To determine if the L-isoaspartyl methyltransferase is reStevens) were collected a t various stages of maturation and seeds were then dissected from them. The stage of develop- sponsive to ABA and possibly involved in a stress-response in ment was determined using the grain length, graincolor, and vegetative tissue, we exposed 4-day-old wheat seedlings to difthe endosperm characteristics of caryopsisdevelopment de- ferenttreatmentsandthen assayed for methyltransferase scribed by Rogers and Quatrano (1983) (Table I). Northern mRNA and enzymatic activity. Four-day-old untreated seedanalysis using a radiolabeled fragment coding the central por- lings contain no detectable methyltransferase mRNA and only tion of the wheat (cultivar Augusta) methyltransferasecDNA low levels of enzyme activity. Northern analysis shows that shows that no methyltransferase mRNA was detectable a t 10-h treatments of water deficit, exposure to 50 p~ ABA, and stage I1 of development while a low level of methyltransferase exposure to salt stress (0.25 M NaCl) dramatically induce the mRNA was found in maturing seedsa t stage 111-IV of develop- expression of methyltransferase mRNA in wheat seedlings ment (Fig. 4A). Stage 111-IV is a period when embryos are (Fig. 5A). Interestingly, in addition to the major 1200-nuclegrowing primarily by cell division, tissue differentiation is be- otide species, we also detected small amounts of a 3150-nucleing completed, and storage proteins are just starting toaccu- otide species that may represent an additional mRNA species mulate (Williamson et al., 1985). The highest level of methyl- or an inefficiently processed pre-mRNA. In contrast, methyltransferase mRNA was detected instage IV seeds whose transferase gene expression was not induced in seedlings exembryos have reached maximal size (Fig. 4A).A lower level of posed to low (4 "C) or high (37 "C) temperature stress. The methyltransferase mRNA was detected in stageV seeds which increase inmRNA levels was paralleled by increases in enzyme have started t u desiccate and enter developmental arrest. A activity. We found that methyltransferaseactivity increased87, similar pattern of mRNA expression during seed maturation 68, and 51% in response t o dehydration, ABA, and NaCl, rewas observed for the Em polypeptide mRNA when the North- spectively (Fig. 5B ). These results indicate that theinduction of ern blot was reprobed with the radiolabeled Em cDNA (Fig. L-isoaspartyl methyltransferase mRNA expression and enzy4A). We then measured methyltransferase enzymatic activity matic activity occurs not only in seed development and germiin the developing seeds and found that the temporal rise in nation, butcan alsobe up-regulated in seedlings during periods

A.

RNA

E

Std -3911

1-

"

+s

Methyltransferase Wheat Repair Protein in

25609

TABLEI Characteristics of the wheat (cultivar Stevens)seeds obtained during caryopsis development Date of wheat head collection 8/27/93

8/9/93

7130193

7/19/93

3-4 29.8 30.9 Whitelgreen Milky-dough I1

Grain length (mm) Average grain fresh w t (mg)" Average percent grain dryw t h Grain color Endosperm characteristics Stage of caryopsis developmentC a

6 7 86.2 42.2 Green Soft-dough 111-IV

7 19.8 93.1 Tan Hard V

8 58.3 57.3 Tan Tanlgreen Hard Hard-dough IV

6 7 19.2 95.1

V

Average grain fresh weight was determinedby averaging the weightsof five seeds.

* Average percent grain dry weight was determined after freeze drying thefive seeds used to determine average grain freshweight.

Stage of embryogenesis was determined using the grain length, grain color, and endosperm characteristicsof caryopsis development described by Rogers and Quatrano (1983).

A.

RNA

A.

RNA Std -

- 391 1

26 S

26 S

rRNA

rRNA

- 2800 - 2800

MT

C

Em

l5.

,

I

D

4"

,

,

37" C' A

,

,

N

I

- 1898 -

072

2.0,

I

I

II

Ill-IV IV

v v

B.

-

0 II Ill-IV IV

v v

8/27 4) 911 Stage of Caryopsis Development (Date Seeds Collected in 1993) (7119 7/30

819

FIG.4. Expression of L-isoaspartyl methyltransferase during wheat caryopsis development. Developing wheat heads (cultivar Stevens) were collected by Dr. M. K. Walker-Simmons from Spillman

I

I

C

D

4'

37'

C'

A

N

Stress Treatment

FIG.5. Induction of L-isoaspartylmethyltransferasemRNA ex-

pression and enzymatic activity in stressed wheat seedlings. Agronomy Farm from July 19 thru September 14 1993, in temporarily Four-day-old seedlings were exposed to different 10-h treatments as placed on ice, and then frozen in liquid nitrogen. Seeds dissected from described under "Materials andMethods:" C , control, no treatment. D, the wheat heads were analyzed for methyltransferase mRNA and en- dehydration stress.4 ', low temperature stress.37 ', high temperature zymaticactivity a s describedinthelegendtoFig. 2. The stage of stress. C', water control. A, ABA treatment. N , salt stress. Following treatment, the seedlings were frozen in liquid nitrogen and then ascaryopsis developmentfor each batchof grains was also determined (see a s described Table I). A, the level of methyltransferase mRNA ( M T )and Em mRNA sayed for methyltransferase mRNA and enzymatic activity ( E m )in stage11, stage 111-IV, and stageV seeds. The relative amount of in the legend to Fig. 2. A, induced expression of methyltransferase of total RNA in total RNAin each lane is represented by the methylene blue staining(seemRNA ( M T )in stressed seedlings. The relative amount by the methylene blue staining (see "Materials "Materials andMethods") of the 26 S ribosomal RNA band on the mem- each lane is represented brane performed prior to hybridization. The autoradiographs showing and Methods") of the 26 S ribosomal RNA band on the membrane exposed for 9 the level of methyltransferase mRNA and EmmRNA were exposed for performed prior to hybridization. The autoradiograph was days. B, top panel: methyltransferase-specificactivity in stressedseed9 and 3 days, respectively. B, specific methyltransferase activity in lings in two independent experiments (light and dark grey columns). seeds at thedifferent stagesof seed development. The date in 1993 that Bottom panel: average percent increase in methyltransferase-specific each batch of wheat seeds wascollected is shown inparentheses. activity (black columns)for the two experiments.

of water deficit and salt stress. Considering that endogenous ABA levels are elevated during dehydration stress (Zeevaart and Creelman, 1988; Bray, 1991) and lateembryogenesis (Quatrano, 1986; Black, 19911, and that exogenous ABA treatment mimics the alterationin methyltransferase mRNA expression and enzymatic activity in response to water deficit (Fig. 5), it is possible that ABA is an endogenous signal involved in the cel-

lular regulation of this protein in wheat. Time Course of L-Isoaspartyl Methyltransferase mRNA Induction in Dehydrated Seedlings-To determine how quickly wheat seedlings respond to dehydration stress in terms of methyltransferase induction, we dehydrated 5-day-old wheat seedlings for up to 10 h and then isolated the total RNA frac-

Protein Repair Methyltransferase in Wheat

25610

RNA Std

26 S rRNA

- 2800 - 1898

- 2800 - 1898 MT

Dehydration: 0 1 2 5 10 5 5 10 10 (hr) Rehydration: 0 0 0 0 0 2 10 2 10 (hr)

FIG.6. Expression of L-isoaspartyl methyltransferasemRNA in seedlings following dehydration and subsequent rehydration. Five-day-old seedlings (cultivar Augusta) were dehydratedfor various lengths of time (0, 1,2,5, and10 h) asdescribed in the legend to Fig. 5. The seedlings were then eitherfrozen immediately or rehydratedfor 2 and 10 h. Northern analysis was performed a s described in the legend to Fig. 2 to measure the level of methyltransferase mRNA (MT). The relative amount of total RNA in each lane is representedby the methylene blue staining (see ”Materials and Methods”) of the 26 S ribosomal RNA band on the membrane performed prior to hybridization. The autoradiograph was exposed for 8 days.

MT

-

872

C‘ A N A+N FIG.7. Enhanced L-isoaspartyl methyltransferase mRNA expression in seedlings treated with both ABA and NaCl. Five-dayold seedlings (cultivar Augusta) were treated with water ( C ’ )or a solution containing either 50p~ ABA (A), 0.25 M NaCl (N), or both ABA and NaCl (A+N).After 10 h, total RNA was isolated from the seedlings and subjected to Northern analysis as described in the legend Fig.to2 to measure the level of methyltransferase mRNA (MT). The relative amount of total RNA in each lane is represented by the methylene blue staining (see “Materials and Methods”) of the 26 S ribosomal RNA band on the membraneperformed prior to hybridization. The autoradiograph was exposed for 8 days.

tions. Northern analysis shows that methyltransferase mRNA the expression of comparable mammalian genes can be reguwas undetectable in thefully hydrated control seedlings (Fig. lated similarly, although the levels of enzyme activity and 6). However, methyltransferase mRNA was detected within 5 h mRNA increase inmouse testicular germ cells during developof the onset of dehydration and accumulated to high levels by ment (OConnoret al., 1989; Galus et al., 1994) and in the brain 10 h as a 1200-nucleotide species. When the plantswere dehy- and testis in aging rats(Mizobuchi et al., 1994). On the other drated for 5 h and then rehydrated, they recovered within 2 h hand, enzyme activity remains constant indeveloping Xenopus and continued t o grow. No methyltransferase mRNA was de- oocytes (O’Connor, 1987) and both enzyme and mRNA levels tected in the rehydrated plants a t 2 or 10 h after watering(Fig. are similar in HeLa cells during normal growth and during 6). Yet, when the plants were exposed to water deficit for 10 h heat shock (Ladino and O’Connor, 1992). In contrast t o plants, and then rehydrated, methyltransferase mRNA was detected methyltransferase activity is widely distributed in mouse (Rofor a t least 2 hafter rehydration but became undetectable by 10 manik et al., 1992) and rat (Diliberto and Axelrod, 1976) tish. By 10h of rehydration,thesestressedplantsappeared sues, althoughspecific activities can vary over a 10-fold range. healthy and also continued to grow. These results indicate thatIn general, the methyltransferase activity reflects the mRNA the expression of the methyltransferase gene is specifically concentration (Galus et al., 1994; Mizobuchi et al., 1994). Inup-regulated in plants in response to waterdeficiency and then terestingly, the activity of the comparable bacterial enzyme eventually down-regulated in response to hydration. from E. coli does not vary over more than a 2-fold range in Additive Effect of ABA and NaCl on L-Isoaspartyl Methyl- exponential and stationary phase growth (Li and Clarke, 1992). transferase mRNA Expression-Although it is clear that exog- Whether the dramatic type of “on-off regulation of mRNA levenous ABA can inducemethyltransferase mRNA in wheatseed- els inresponse to hormonal and environmental signals oblings, it is unclearif this phytohormone is theonly mediator of served in the wheat system occur in animal or bacterial systhis response. Bostock and Quatrano(1992) have shown that in tems remain to be seen. addition to itseffects via ABA, NaCl can independently mediate In thiswork, we show that theexpression of the L-isoaspartyl the induction of Em gene expression in rice. Since salt stress methyltransferaseinwheat depends on the developmental induces methyltransferase mRNA expression (see Fig. 5), we stage of the seed. During the initial stagesof caryopsis develinvestigated the possibility that methyltransferase gene ex- opment, L-isoaspartyl methyltransferase mRNA and enzymatic pression might also be mediated by NaCl through an ABA- activity are not detected. By the late stagesof caryopsis develindependent pathway. Five-day-old wheat seedlings were thus opment, both methyltransferase mRNA expression and enzywatered with solutions containing 50 V M ABA, 0.25 M NaCl, or matic activity are significantly induced. The seed maintains a a combination of 50 V M ABA and 0.25 M NaCl for 10 h.As relatively high level of methyltransferase activity until the onpreviously observed, no methyltransferase mRNA was detected set of germination. Once the seed beginsto germinate, there is in the control plants treated with water while a significant a concomitantreduction in total methyltransferase activity induction of methyltransferase mRNA expression was detected which coincides with the reduction in the steady statelevel of in seedlings treated with either AI3A or NaCl (Fig. 7). However, its corresponding mRNA. Only a very low level of methyltranswhen the seedlings were treated with both ABA and NaCl, ferase activity is detectable in thevegetative tissue of a develmethyltransferasegene expression was increasedapproxioping seedling growing under optimal conditions. mately 2-fold over the effect of either agentalone. The additive However, when the seedling is subjected to environmental effect of a combined ABA-NaCl treatment suggests that the stressessuch as dehydration or salinity, methyltransferase methyltransferase gene may bea salt-responsive gene inaddi- mRNA expression and enzymatic activity are coordinately intion to anABA-responsive gene. duced. In response to waterdeficit, the expression of the methyltransferase gene is up-regulated in a seedling within 5 h of DISCUSSION dehydration and down-regulated with a similar response time We present here thefirst example of a developmentally, hor- upon rehydration. However, if the seedling is exposed to a longmonally, and environmentally regulated protein L-isoaspartyl er period of dehydration, the expression of the methyltransmethyltransferase. No evidence has been obtained to date that ferase gene is greatly increased and sustainedlong after water

25611

Protein Repair Methyltransferase in Wheat has been replenished. This suggests that a plant can regulate the level of active L-isoaspartyl methyltransferase in cells t o protect its proteins from the types of spontaneous damage associated with dehydration stress. These regulatoryprocesses are also mimicked in seedlings in response to exogenous treatment with ABA. The characteristic increase in thelevel of endogenous ABA in plants during caryopsis development (Quatrano, 1986; Black, 1991) and in response to environmental stress (Zeevaart and Creelman, 1988; Bray, 1991) suggests that ABA may be a n endogenous signal involved in the regulationof the cellular levels of the L-isoasparty1 methyltransferase throughout the lifespan of a plant. Other late embryogenesis proteins that are ABA- and water stress-responsive have beenshown to be up-regulated at least in partby elevated levels of endogenous ABA. These includethe Em gene from wheat (Williamson and Quatrano, 1988; Morris et al., 1990), the pLE25 gene from tomato (Cohen and Bray, 1990), the dehydrin/RAB proteins from barley (Chandler et al., 1988) and rice (Mundy and Chua, 1988), and a glycine-rich protein from maize (Gomez et al., 1988). The responsiveness of these genes to ABA indicates that they are likely to be regulated ina tissue-independent, ABA-dependent manner. In fact, specific promoter elements (ABA-responsive elements)have been identified that are requiredfor AE3A-responsiveness in at least two lea genes, Em and rab 16A (Marcotte et al., 1988, 1989; Mundy et al., 1990). Transcription factors recognizing these ABA-responsive elements have been subsequently isolated (Guiltinan et al., 1990; Oeda et al., 1991). However, it is important to note that some seed protein genes (Pla et al., 1989) and some stress-induced proteins (Bray, 1991) are not regulated by elevated levels of ABA suggesting that other pathways or signals may involved be in theregulation of gene expression during embryogenesis and physiological stress. Given our preliminary resultson the dualeffect of NaCl and ABA on methyltransferase gene expression, we suspect that NaCl may alsobe involved in theregulation of the cellular levels of this enzyme via a ABA-independent NaCl pathway such as thatproposed by Bostock and Quatrano(1992). Hence, the activation and repression of the L-isoaspartyl methyltransferase gene in plantsmay involve a number of cellular participants and response pathways. The developmental regulation of the wheat methyltransferase shown in this work is consistent with the putative protein repair function of the L-isoaspartyl methyltransferase in limiting protein damage in aging seeds. Moreover, the methyltransferase can be induced in vegetative tissues subjected to dehydration and salt stress to repair t-isoaspartyl damage in proteins. Thisenzyme alongwith other lea proteins up-reguis lated in a seed just prior to metabolic arrest (dormancy) and desiccation. Although the relationshipbetween water loss and L-isoaspartyl residue formation in proteins is notclear, the altered environment indesiccated seeds or in drought-affected mature plants may result in more rapid spontaneous protein degradation. It has beenshown that succinimideformation from L-asparaginyl residues inpeptides, the reaction that leads directly to L-isoaspartyl residue formation, increases with the dielectric strength of the medium (Brennan and Clarke,1993). Thus, the increase in intracellularsalt concentration brought acceleration of on by desiccation or salt stress may result in an L-isoaspartyl residue formation. The abilityof the methyltransferase to repairL-isoaspartyl damage incurred during thedesiccation period of seed maturation and the solvation period of germination, along with periods of salt stress and drought stress, may be crucial to plant survival.Ultimately, the viability of an aging plant will depend on how wellthe plantprotects

its functional and structural proteins from environmentally and developmentally imposed stresses. Acknowledgments-We thank Dr. M. K. Walker-Simmonsfor collecting the wheat heads at Spillman Agronomy Farm (Washington State University) and valuable discussions. We are also grateful to Dr. Robert Forsberg for providing the wheat seeds and Dr. Ralph Quatrano forthe gift of the wheat Em cDNAclone. We thank Drs. Elaine Tobin, Sharlene Weatherwax, and Judy Brusslan (all at the University of Californiaat Los Angeles) for their helpful advice.

REFERENCES Anderson, J. W., and Rowan, K. S. (1967) Phytochemistry ( O r f )6, 1047-1056 Aswad, D. W., and Johnson, B. A. (1987) 'Ilends Biochem. Sci. 12,155-159 Bailey, J. L. (1967) Techniques in Protein Chemistry, pp. 34lL346, Elsevier Publishing Co., New York Barten, D. M., and O'Dea, R. F. (1990) Life Sci. 47,181-194 Berge, S. K., Bartholomew, D. M., and Quatrano,R. S. (1989) in Molecular Basisof Plant Development (Goldberg, R. B.,ed) pp. 193-201, Alan R. Liss, New York Black, M. (1991) in Abscisic Acid Physiology and Biochemistry (Davies, W. J., ed) pp. 99-124, BIOS Scientific Publishers Ltd., Oxford, UK Bostock, R. M., and Quatrano, R. S. (1992) Plant Physiol. 98, 1356-1363 Bray, E. A. (1991) in Abscisic Acid Physiology and Biochemistry(Davies, W. J., ed) pp. 81-98, BIOS Scientific Publishers Ltd., Oxford, UK Brennan, T. V., and Clarke, S. (1993) Protein Sci. 2, 331-338 Chandler, P. M., Walker-Simmons, M., King, R.W., Crouch, M., and Close, T. J. (1988) J. Cell Biol. 12C, (suppl.) 143 Church, G.M., and Gilbert, W. (1984)Proc. Natl. Acad. Sci.U. S. A . 81,1991-1995 Clarke, S. (1985) Annu. Reu. Biochem. 54,479-506 Cohen, A,, and Bray, E.A. (1990) Planta (Basel)182,27-33 Cuming, A.C. (19841 Eu,: J. Biochem. 145,351-357 Diliberto, E. J., Jr.. andAxelrod, J . (1976) J. Neurochem. 26, 1159-1165 Dure, L., I11 (1993) in Control of Plant Gene Expression (Verma, D. P. S., ed) pp. 325-335, CRC Press, Boca Raton, FL Dure, L., 111, Crouch, M., Harada, J., Ho, T-H.D., Mundy, J., Quatrano, R. S., Thomas, T.,and Sung, Z. R. (19891 Plant Mol. B i d . 12,475-486 Galletti, P., Ciardiello, A,, Ingrosso, D., Di Donato, A., and DAlessio, G. (1988) Biochemistry 27, 1752-1757 Galau, G . A., Hughes, D. W., and Dure, L.,111 (1986)Plant Mol. Biol. 7, 155-170 Galus, A,, Lagos, A,, Romanik, E. A,, and O'Connor, C. M. (1994)Arch. Biochem. Biophys. 312,524-533 Geiger, T., and Clarke, S. (1987) J. B i d . Chem. 262, 785-794 George-Nascimento, C., Lowenson, J., Borissenko, M., Calderon, M., Medina-Selby, A,, Kuo, J., Clarke, S., and Randolph, A. (1990) Biochemistry 29,9584-9591 Goldberg, R. B., Barker, S. J., and Perez-Grau, L. (1989) Cell 56, 149-160 Gomez, J., Sanchez-Martinez, D., Stiefel, V., Rigau, J., Puigdomenech, P., and Pages, M. (1988)Nature 334,262-264 Grzelczak, Z. F., Sattolo, M. H., Hanley-Bowdoin, L. K., Kennedy, T.D., and Lane, B. G. (1982) Can. J. Biochem. BO, 389-397 Guiltinan, M. J., Marcotte, W. R., Jr., and Quatrano, R. S. (1990) Science 260, 267-27120131-20139 Johnson, B.A., Murray, E. D., Jr., Clarke, S., Glass, D. B., andaswad, D. W. (1987a) J . Biol. Chem. 262,5622-5629 Johnson, B. A., Langmack, E. L., and Aswad, D. W. (1987b) J . Biol. Chem. 262, 12283-12287 Ladino, C. A,, and O'Connor, C. M. (1992) J. Cell. Physiol. 153, 297-304 Lane, B. G. (1991) FASEB J. 5, 2893-2901 Lehrach, H., Diamond, D., Wozney, J. M., and Boedtker,H. (1977)Biochemistry 16, 4743A751 Li, C., and Clarke, S. (1992) Proc. Natl. Acad. Sci. U.S. A. 89, 9885-9889 Litts, J. C., Colwell, G. W., Chakerian, R. L., and Quatrano, R. S. (1987)Nucleic Acids Res. 15, 3607-3618 Lowenson, J., and Clarke, S. (1991) Gerontology 37, 128-151 Lowenson, J., and Clarke, S.(1992) J . Biol. Chem. 267, 5985-5995 Marcotte, W. R., Jr., Bayley, C. C., and Quatrano,R. S. (1988)Nature 336,454457 Marcotte, W. R., Jr.,Russell, S. H., and Quatrano, R. S. (1989) Plant Cell 1, 969-976 McFadden, P. N., and Clarke. S.(1987) Proc. Natl. Sei. Acad. U. S. A. 84, 25952599 Mizobuchi, M., Murao, K.. Takeda, R., and Kakimoto, Y.(1994) J. Neurochem. 62, 322-328 Moms, P. C., Kumar, A., Bowles, D. J., andCuming, A. C. (1990)Eu,: J. Biochem. 190,625-630 Mudgett, M. B., and Clarke. S.(1993) Biochemistry 32, 1110&11111 Mundy, J., and Chua, N.-H. (1988) EMBO J. 7,2279-2286 Mundy, J.,Yamaguchi-Shinozaki, K., and Chua,N.-H. (199O)Proc.Natl. Acad. Sci. U. S. A . 87, 1406-1410 OConnor, C. M. (1987) J. Biol. Chem. 262, 10398-10403 OConnor, C. M.,Germain, B. J., Guthrie, K. M., Aswad, D. W., and Millette, C. F. (1989) Gamete Res. 22, 307-319 Oeda, K., Salinas, J., and Chua, N.-H. (1991) EMBO J. 10, 1793-1802 Pla, M., Goday, A., Vilardell, J., Gomez, J., and Pages, M. (1989)Plant Mol. Biol. 13,385-394 Quatrano, R. S. (1986) Oxt: Sura Plant Mol. Cell B i d . 3, 467476 Quatrano, R. S., Marcotte, W. R., and Guiltinan,M. (1992)in Control ofplant Gene

25612

Protein Repair Methyltransferase in Wheat

Expression (Verma, D.P. S., ed) pp. 69-87, CRC Press, Boca Raton, FL Rogers, S. O.,and Quatrano, R. S. (1983)Am. J. Bot. 70,308-311 bm-k, E. A,, Ladino, c. A., filloy, L. c., whdenne,s. c., and o’comor, c.M. (1992)Gene (Amst.) 118,217-222 Sambrook, J , , kitsch, E, F,, and “atis, T,(1989)Molecular Cloning:A LaboratoryManual, 2nd Ed., Cold Spring Harbor Laboratory, Cold Spring Harbor,

NY Stephenson, C., R.

an$ Clarke, S. (1989) J. Biol. Chem. 284,61644170 39,439-473

Trivedi, L., Gupta, A,, Paik, W. K, and Kim, S. (1982) Eur. J. Biochem. 128, 349-354 Verwoerd, T.C., Dekker, B. M. M., and Hoekema, A. (1989)Nucleic Acids Res. 17, 2362 Williamson, J. D., and Quatrano, R. S. (1988) Plant physiol. 86,208-215 Williamson, J. D., Quatrano, R. S., and Cuming, A. C. (1985)Eur: J. Biochem. 162, 5 ~ 1 4 ~ 7 Zeevaart, J. A. D., and Creelman, R. A. (1988)Annu.Physiol. Rev. Plant Mol. Biol.