Plant Molecular Biology 11:495-506 © Kluwer Academic Publishers, Dordrecht - Printed in the Netherlands

495

Cloning and characterization of a cDNA encoding a mRNA rapidlyinduced by ABA in barley aleurone layers Bimei Hong, Scott J. Uknes and Tuan-hua David Ho*

Plant Biology Program, Department of Biology, Division of Biology and Biomedical Sciences, Washington University, St. Louis, MO 63130, USA (*author for correspondence) Received 7 March 1988; accepted in revised form 19 July 1988

Key words: abscisic acid, barley aleurone layers, cDNA cloning, gene expression Abstract

Abscisic acid (ABA) inhibits the gibberellic acid induced synthesis o f a-amylase in barley aleurone layers, yet ABA itself induces more than a dozen polypeptides (Lin & Ho, Plant Physiol 82: 289-297, 1986). As part o f our effort to elucidate the molecular action o f ABA in barley aleurone layers, we have isolated and characterized an ABA-induced cDNA clone, pHV A1. This cDNA clone hybridizes to an RNA species of approximately 1.1 kb from ABA-treated barley aleurone layers. The level of this m R N A is tripled within 40 minutes after ABA treatment, reaches a peak at 8-12 h, and is present up to 48 h. The induction of this mRNA responds to concentrations of ABA as low as 10-9 M, but higher ABA concentrations induce higher expression of this mRNA. The products of hybrid-select translation and in vitro transcription/translation with pHV A1 comigrate on SDS gel as a 27 kDa polypeptide. However, the sequence of pHV A1 indicates that it has an open reading frame encoding a 22 kDa protein. This size discrepancy is probably due to the high content o f the basic amino acid, lysine. This notion has been confirmed by two-dimensional gel electrophoresis showing that this polypeptide is one o f the most basic proteins in ABA-treated barley aleurone layers. The deduced amino acid sequence of pHV A1 contains nine imperfect repeats 11 amino acids long which share homology with cotton Lea 7 protein (Baker, Steele & Dure, Plant Mol Biol, in press). The identity and function of the encoded product of pHV A1 is under investigation.

Introduction

The plant hormone abscisic acid (ABA) exerts regulatory roles in several important physiological and developmental processes such as seed development and germination, stomatal function and the plant's response to water and salinity stress [28, 35]. In barley aleurone layers, ABA antagonizes all of the effects of gibberellic acid (GA 3) so far studied [34]. Based on transcriptional and translational inhibitor studies, Ho and Varner [14] have suggested that the effect o f ABA in barley aleurone layers may depend

on synthesis of new proteins. Induction o f new proteins in barley aleurone layers by ABA has been investigated by several groups [13, 20]. More recently, Lin and H o [17] have cataloged the ABA-induced proteins by gel electrophoretic analyses of in vivo and in vitro synthesized proteins. Nine ABAinduced polypeptides were observed by onedimensional SDS-polyacrylamide gel electrophoresis and at least 16 polypeptide spots were seen on a two-dimensional gel. Although most of the ABAinduced proteins remain unidentified, a few of them are known to have some interesting features. A

496 29 kDa acid soluble protein is the most abundant species among all the ABA-induced proteins [17]. A 36 kDa protein has some common antigenic determinants with a barley lectin specific for glucosamine, mannosamine and galactosamine [23]. A 21 kDa protein is an a-amylase inhibitor that has been reported by Mundy et al. [21] and by Weselake et al. [37]. Mundy et al. [21] have shown that this aamylase inhibitor is synthesized in the endosperm during seed development, probably to prevent the premature hydrolysis of stored starch. It has been observed that ABA enhanced its own conversion to a metabolite, phaseic acid, which was as effective as ABA in inhibiting oL-amylase synthesis in aleurone layers [33]. Uknes and Ho [33] have proposed that some of the ABA-induced proteins may be involved in the metabolic conversion of ABA to phaseic acid. The action of ABA on gene expression is also being studied in other plant systems. The effect of ABA on the regulation of seed storage proteins was first observed in Phaseolus vulgaris [31] and Brassica napus [8]. Abscisic acid is required for the continuous expression of storage protein genes in embryo culture of these plants [9]. The synthesis of several seed storage proteins, ~subunit of ~3-conglycinin of soybean [5], albumin Em [38] and wheat germ agglutinin (WGA) [24] of wheat, have been shown to be induced by ABA. In developing cotton seeds, although the synthesis of storage proteins is not regulated by ABA, it has been observed that a set of late embryogenesis-abundant mRNA (Lea mRNA) is induced by ABA [11]. It has been suggested that the function of this set of proteins is to inhibit premature germination before seed development is complete or to compensate for water loss during seed desiccation [1, 11]. The ABA-induced proteins also include a 26 kDa protein in salt-stressed tobacco cells which is believed to be involved in the adaptation to salt stress [28]. As part of our effort toward the elucidation of the molecular mechanism underlying the ABA induction of new proteins, we have isolated cDNA clones corresponding to ABA-induced proteins in barley aleurone layers. We report here the isolation and characterization of a cDNA clone which is rapidly induced by ABA in barley aleurone layers.

Materials

and methods

Chemicals (+)-cis, trans-Abscisic acid, chloramphenicol, and Fraction V BSA were obtained from Sigma Chemical Co., St. Louis, MO. Guanidine HC1 and nick translation enzymes were obtained from Bethesda Research Laboratories, Bethesda, MD. [35S] Methionine (> 1000 Ci/mmol) was obtained from New England Nuclear, Boston, MA. Deoxycytidine ot-[32p] triphosphate (>400 Ci/mmol), adenosine ,y_[32p] triphosphate (3000 Ci/mmol) and [o~-[35S] thio]dATP (> 1000 Ci/mmol) were purchased from Amersham, Arlington Heights, IL. Restriction enzymes were obtained from Bethesda Research Laboratories and New England Biolabs, Beverly, MA. Rabbit reticulocyte lysate in vitro translation . system was obtained from Promega Biotec, Madison, WI. Ampholines (pH 3.5-10 and 5-7) were purchased from LKB, Bromma, Sweden as 40°7o solutions. DNA sequencing kit with Sequenase (modified T7 polymerase) was obtained from United States Biochemical Corporation, Cleveland, OH.

Plant materials and incubation conditions

Barley seeds (Hordeum vulgare L. cv. Himalaya) from the 1981 and 1985 harvests at Washington State University, Pullman, WA were used. The embryoless half seeds were imbibed for 4 days as described by Belanger et al. [2]. Aleurone layers were prepared from imbibed half seeds by peeling away the starchy endosperm and were incubated at 25 °C in 20 mM sodium succinate pH 5.0, 20 mM CaCI2, and 10 ~tg/ml chloramphenicol in a shaking water bath at 120 strokes per minute.

Cloning and screening o f p U C cDNA library

For cDNA cloning, total RNA was isolated from layers incubated with 2×10 -5 M ABA for 24 h, according to the procedure described by Rogers and Milliman [26]. Poly(A) ÷ mRNA was isolated by oligo(dT)-cellulose chromatography. Cloning of

497 cDNA generally followed the procedure of Heidecker and Messing [12], with modifications described by Rogers [25]. Double-stranded cDNA was synthesized, inserted into the Pst I site of pUC18 via dG:dC tailing, and transformed into Escherichia coli strain JM101. Ampicillin-resistant colonies were streaked on nitrocellulose filters, replicated, and lysed [18]. After baking for 2 h at 80°C under vacuum, the membranes were washed to remove cellular debris, and prehybridized [18]. One set of replica filters was hybridized overnight with end-labeled poly(A) ÷ RNA isolated from aleurone layers treated with 2 × 10-5 M ABA for 4 h, and the other set was hybridized with similarly labeled RNA from tissue not treated with ABA. Clones that preferentially hybridized to poly(A) + RNA from ABA-treated tissue but not to RNA from the control tissue were isolated. For preliminary characterization, plasmids were isolated from overnight cultures by the alkali denaturation-rapid isolation procedure [4]. Largescale plasmid preparation was purified by CsCl/ethidium bromide centrifugation [18]. The plasmid DNA was digested with the restriction enzymes Eco RI and Hind III, separated on an agarose gel, and blotted onto a nitrocellulose filter by the method o f Southern [29]. Duplicate filters were made and probed with end-labeled mRNA. The filters were then washed, and exposed to Kodak XAR-5 film at - 8 0 ° C with a DuPont Cronex "Lightning plus" intensifying screen.

Screening o f kgt 10 library A kgt 10 library was constructed according to the method of Watson and Jackson [36], from poly(A) ÷ RNA isolated from ABA-treated aleurone layers (S. J. Uknes, unpublished). The plaques were plated, lifted onto nitrocellulose, and then hybridized with a [32p]-labeled cDNA isolated from the pUC 18 library.

Hybrid-select translation Twenty/~g of plasmid DNA was digested with Eco RI and Hind III to cut out the cDNA insert, electrophoresed in an agarose gel and blotted onto

nitrocellulose by the method of Southern [29]. The insert band was cut from the filter and baked at 80 °C under vacuum. Prehybridization and hybridization were performed as described by Miller et al. [19]. The filters were hybridized overnight at 37 °C in 100 #1 o f solution containing 50% formamide, 10 mM PIPES pH 6.4, and 0.4 M NaCl with 5/~g o f poly(A) ÷ RNA from aleurone layers treated with 2 × 10-5 M ABA for 4 h. The filters were washed 10 times (5 min each) at 60 °C with 0.1 × SSC (1 × SSC is 0.15 M NaCI, 0.015 M trisodium citrate) containing 0.5O7o SDS followed by three washes in 2 mM EDTA at pH 7.9. The bound RNA was eluted from the filters by boiling them in 300/~1 1 mM EDTA containing 10/~g of wheat germ tRNA as carrier, followed by quick freezing in an ethanol-dry ice bath. The supernatant was removed after thawing at room temperature and precipitated with ethanol. The RNA was pelleted, dried and dissolved in 10/~1 H20. As a control, a filter with the pHV A1 cDNA insert was hybridized to mRNA from layers incubated without ABA. Five ~tl of the hybrid-released RNA was in vitro translated in rabbit reticulocyte lysate with [35S] methionine as the label, and analyzed by SDS-PAGE as described by Laemmli [16]. The hybrid-select translation products were also analyzed by two-dimensional PAGE as described by O'Farrell [22] with modifications in isoelectric focusing (IEF) as described by Cooper et al. [7]. The ampholines in the IEF gel were in a 3:2 ratio of pH 3.5 to 10 species to pH 5 to 7 species. All SDS polyacrylamide gels were 11.5°/0 (w/v). Following electrophoresis, the gels were stained for the molecular weight markers. Gels were prepared for fluorography following the procedure of Jen and Thach [15]. The gels were then dried and exposed to Kodak XAR-5 film at -80°C.

Northern analysis For the time course and dosage response studies, the total RNA was isolated from 50-75 aleurone layers as described by Belanger et al. [2]. Twenty or thirty /zg total RNA samples were used in formaldehydeagarose gel electrophoresis as described by Maniatis et al. [18]. After electrophoresis, the RNA was blotted onto GeneScreen membranes (New England Nu-

498 clear, Boston, MA) using 10× SSC as the transfer medium. The filters were air-dried and baked under vacuum for 2 h at 80°C. Before hybridization, the filters were wetted with H20 and the R N A was cross-linked to the membrane by irradiation with a UV lamp for 5 min [6]. Hybridization and washing were performed according to the method of Church and Gilbert [6].

GeneScreen. The hybridization was performed as described by Church and Gilbert [6], with [32p]_ . labeled cDNA as the probe. The filter was exposed as described above.

Results Isolation o f an A B A - i n d u c e d c D N A

Nucleotide sequencing The c D N A insert was subcloned into MI3 mpl8 and mpl9 vectors. D N A was sequenced by the method of Sanger [27] using the Sequenase Kit from United States Biochemical Corporation. The products of the Sequenase reactions were analyzed on 0.4 m m wedged urea-acrylamide gels. After electrophoresis, the gel was soaked in 5070 acetic acid and 5070methanol for 30 min, and then dried in a gel drier at 70 °C for 1 h. Autoradiography was performed at room temperature.

A pUC 18 c D N A library was prepared with poly(A) + R N A from aleurone layers treated with 2 × 10-5 M ABA for 24 h. About 2500 recombinants were obtained in this library. One bacterial colony, pHV A1 hybridized selectively to poly(A) + R N A from ABA-treated tissue. The plasmid D N A of clone pHV A1 was isolated, digested with Eco RI and Hind tII, and analyzed by agarose gel elec-

Transcription in vitro The c D N A clone was subcloned into the Eco RI site of the Bluescript KS(+) plasmid (Stratagene, San Diego, CA). The subclone with the 5' end proximal to the T7 promoter was transcribed and capped in vitro using T7 R N A polymerase. As the controls, the subclone with c D N A in negative orientation, Bluescript KS(+) vector alone, or a deletion mutant which lacks the speculated translation initiation codon were also transcribed in vitro. The transcripts were translated in vitro in nuclease treated rabbit reticulocyte lysate with [35S] methionine as the label, and the products were run on 11.5070 SDSacrylamide gel.

Genomic southern Genomic D N A was isolated from 5-day old etiolated barley seedlings by the method o f Blin and Stafford [3]. Ten #g of D N A were digested with restriction enzymes indicated, and electrophoresed on an 0.8°7o agarose gel which was subsequently blotted onto

Fig. 1. Northern blot analysiswith clonepHV A1 plasmid DNA. Total RNA was isolated from layers treated without ABA (lanes 1, 2), or with 20/~M ABA (lanes 3, 4) for 4 h. Twenty/~g(lanes 1 and 3) or 40 ~g (lanes 2 and 4) of total RNA was subjected to electrophoresis through a 1% agarose gel in the presence of formaldehyde, transferred to GeneScreenmembraneand hybridized to [32p]_labeled cDNAinsert from clone pHV A1 as described in Materials and methods. The molecular weights indicated at the left wereestimated accordingto the position of ribosomal RNAs.

499

Fig. 2. Analysis of the in vitro translation product of mRNA selected by clone pHV A1. The cDNA insert of pHV A1 was transferred

to nitrocellulose, and hybridized with poly(A) ÷ RNA isolated from control and 20/~M ABA treated barley aleurone layers. RNA was translated in vitro using rabbit reticulocyte lysate system with [35S] methionine as the label. A. One-dimensional gel electrophoretic analysis of [35S]-labeled translation product. Brome mosaic virus (lane 1) and no RNA (lane 2) control were included. Lanes 3 and 4 are in vitro translation products from total RNA isolated from tissue treated with 20/~M ABA for 4 h, or without ABA treatment, respectively. Lane 5 represents the translated product of mRNA selected from ABA treated RNA by the insert from clone pHV A1. Lane 6 represents the translated product of mRNA selected from tissue incubated without ABA. Per lane 100000 cpm were loaded. Molecular weight markers in kDa are shown on the left edge of the gel. The arrow indicates polypeptide selected by the hybrid-select translation. B and C. Two dimensional gel electrophoretic analysis of [35S]-labeled translation product of mRNA (B, from ABA treated aleurone layers; C, from tissue incubated without ABA) selected by hybridization to clone pHV A1 insert. To the IEF -gel 200000 cpm of in vitro translated products were aplied first, followed by SDS PAGE. The arrow indicates the polypeptide translated from the selected mRNA.

500 trophoresis. A short insert of about 300 bp was observed on the gel. As described below duplicate filter blots containing the insert DNA were prepared, hybridized with [32p] end-labeled poly(A) ÷ RNA from 4 h ABA-treated (+ ABA) and from control tissue respectively. The insert hybridized preferentially with + A B A poly(A) ÷ RNA, indicating that clone pHV A1 is ABA induced.

m R N A species encoded by p H V A1 Northern blot of aleurone layer RNA probed with clone pHV A1 DNA is shown in Fig. 1. Clone pHV A1 hybridized preferentially to an RNA species from the sample treated for 4 h with ABA (lanes 3, 4). In the control sample (lanes 1, 2), the level of this mRNA was barely detectable. The size of the RNA species was estimated to be about 1.1 kb, based on comparison of its mobility to a ladder of RNA size standards (BRL) in another experiment (data not shown).

Polypeptide encoded by clone p H V A1 Five #g of poly(A) + RNA from ABA-treated barley aleurone layers was hybridized with immobilized clone pHV A1 insert DNA and the selected mRNA was then eluted and in vitro translated in rabbit reticulocyte lysate system. The in vitro translation product was analyzed by both one- and twodimensional gel electrophoresis. As shown in Fig. 2A, a polypeptide of 27 kDa (arrow) was translated from mRNA selected from ABA poly(A) ÷ RNA (lane 5), but not from the control (-ABA) sample (lane 6). In Fig. 2B, the hybrid-select translation product was analyzed by two-dimensional gel electrophoresis. Besides the background proteins from reticulocyte lysate, there was only one protein spot (indicated by the arrow) observed at the basic end of the gel. The pI of this 27 kDa protein is estimated to be higher than 8.5. This protein spot was not seen when the RNA from control tissue was used in the hybrid-select translation (Fig. 2C).

Fig. 3. Dosage response of ABA induction of clone pHV A1 mRNA. A. Total RNA was isolated from aleurone layers incubated for 24 h with no ABA (lane 7), or with different concentrations of ABA, ranging from 10-9 to 10-4 M (lanes 1 to 6). Twenty #g of total RNA was applied to each lane in a formaldehyde gel. After electrophoresis, the RNA was transferred onto GeneScreen, and the membrane was hybridized with clone pHV A1 plasmid DNA as described in Materials and methods. B. Estimated level of RNA as determined by scanning the autoradiogram shown in Fig. 3A with a Joyce Loebl Chromoscan 3 densitometer. Relative density units are given.

501

Dosage response of ABA induction of clone p H V A1 m R N A

Time course of the induction of m R N A encoded by p H V A1

Figure 3A shows the dosage-response relationship o f p H V A1 R N A expression and ABA concentrations ranging from 10 -4 tO 10 -9 M after the 24 h treatment. The level of this m R N A increases in response to 10-9 M ABA (lane 1), but higher concentrations of ABA induce higher amounts o f this m R N A . AT 10 -4 M ABA (lane 6), the highest concentration used in this experiment, the highest expression of this m R N A was induced. The induction at this concentration is about 7-fold over the control RNA, as estimated by densitometric scanning shown in Fig. 3B.

As shown in Fig. 4A, the level o f m R N A encoded by p H V A1 reached a peak at 8-12 h treatment of 2 × 10 -5 M ABA, then decreased afterwards to an intermediate level. This m R N A is present only at very low level in the control tissue (lanes 1 to 4), and its level increased slightly upon prolonged incubation. The enhancement on the level of this m R N A at 12 h of ABA treatment was about 13-fold over the con-

B

D

80

/

200 60'

C

80'

~ 40.

uJ

w "1" ~ o.

-r

20' r

0, 0

-

o.

ABA

4o 20

,

i

i

i

10

20

30

40

T

6O

•

~

0

50

0

(hours)

20

,i . A B A

,

,

40

60

T

II

720

(minutes)

Fig. 4. Time course of expression of clone p H V A1 m R N A in barley aleurone layers. A. Long-term time course of clone p H C A1 m R N A induction. Total R N A was isolated from aleurone layers treated with or without 20/~M A B A for the indicated lengths of time. Twenty #g total R N A was applied to each gel lane. The Northern blot procedure was identical to that described under Fig. 3. B. Estimated level o f R N A determined by scanning the autoradiogram o f Fig. 4A by densitometer. Relative density units are given. C. Early time course of clone p H V AI m R N A induction. Total R N A was isolated from aleurone layers treated with 200 #M A B A for the indicated time (lanes 2 to 7), or without A B A for 1 h (lane 1). Thirty ttg total R N A was applied to each lane except for lane 8, which contains 20 #g total R N A from layers treated with 20 #M A B A for 12 h. D. Estimated level of R N A determined by densitometric scanning. Relative units are given.

502 -119

GTGCCC-GTAGTAAATCATGAGCATCTCI-rGC

-88

GACTCGAAAC,GTAGTACAGCAACAGCCTAAAGCGAGTCC,GAGTGGTG ABCCAGRCGTGn-rGTTTGAGCTAGATCGTGAGACC~G

1

ATG GCC TCC AAC CAG AAC CAG GGG AG~C TAC CAC GCC GGC GAG ACC hAG C-.-.-.-.-.-.-.-.-~CCGC ACC GAG Met Ala Ser Asn Gin Asn Gin Gly Set Tyr His Ala Gly Glu Thr Lys Ala Arg Thr Glu

61

GAG AAG ACC GGG CAG ATG ATG GGC C.-:.-:.-:.-:.-:.-~ACC AAG CAG AAG Glu Lys Thr Gly Gin Met Met GIy Ala Thr L~/s Gin L~ts I AOC AAG CAG AAG GCC GGC GAG ACG GCC GAG GCC ACC AAG Thr L~'s Gin L~/s Ala Gly Glu Thr Ala Glu Ala Thr L}~s 2 GAG GCC GCC AAG CAG AAG GCC GC,C GAG ~ ,SAG GAC AAG Glu Ala Ala L~'s Gin Lys Ala Ala Glu Ala Lys Asp Lys 4 GCC AAG GAC AAG ACG TAC GAG ACG GCG CAG GCG GCC AAG Ala L~/s Asp Lys Thr T~r Glu Thr Ala Gin Ala Ala L~/s

121 181 241

5

301 361

GC,G GGG GAG ACC ACC GAG GCC Ala Gly Gin Thr Thr Glu Ala CAG AAG ACC GGC GAG AC.,G C-:-:-:-:-:-:-:-~ Gin L~ts Thr GI~' Glu Thr Al..~a 3

ACG GC.G CAG ACG GCG CAG GCG Thr Ala Gin Thr Ala Gin Ala GAG (Z3C GCC GCC CAG GGC AAG Glu Arg Ala Ala Gin Gly Lys 6

GAC CAG ACC GGC AGC GCC CTC GGC GAG AAG ACG GAG GCG Asp Gin Thr Gly Ser Ala Leu Gly Glu Lys Thr Glu Ala Ala

AAG CAG AAG GCC GCC Lys Gin Lys Ala Ala 7 AC,;G ACG GAG GCG GC,C AAG CAG AAG C-CC GCC GAG GCA ACC GAG GCG GCC hAG CAG AAG Th~=r Thr Glu Ala Ala Lys Gin Lys Ala Ala Glu Ala Thr Glu Ala Ala Lys Gin Lys 8

GAG Glu GCG Ala

9

421

TCC GAC ACG GCG CAG TAC ACC AAG GAG TCC GC,G GTG GCC GGC AAG GAC ,SAG ACC GGC AGC Ser Asp Thr Ala Gin Tyr Thr Lys Glu Ser Ala Val Ala Gly Lys Asp Lys Thr Gly Ser

481

GTC CTC CAG CAG C-CC GGC GAG ACG GTG GTG AAC GCC GTG GTG GGC GCC AAG GAC GCC GTG Val Leu Gin Gin Ala Gly Glu Thr Val Val Asn Ala Val Val Gly Ala Lys Asp Ala Val

541

GCA AAC AC;G CTG GGC ATG GGA GGG GAC AAC ACC AGC GCC ACC AAG GAC GCC ACC ACC GGC Ala Asn Thr Leu Gly Met Gly Gly Asp Asn Thr Ser Ala Thr Lys Asp Ala Thr Thr Gly

601

~ Ala

669

VFCC3-rTAGTCGTGTI-FGGTCGTrCGAGGGCCFICTACATAfTTCATAVI-IGTATGITrCCACTCR-FCATGAITFCCGCTCATFFAGTGT~

ACC GTC AAG GAC ACC ACC ACC ACC ACC AGG AAT CAC TAG ACGCATGC;GVFCGCGGI-rAATI-rCCG Thr Val Lys Asp Thr Thr Thr Thr mhr Arg Ash His End

762

GTI-FGCCTCCGA]TTGATGTA CTCGTCTCTGGI-FCTGTAATGAGnATAATCCATC-GGCTI-rGGTGTAAATGGATAACGAGGACACTC, GA

852

AGGCGGC,AA.TAj~,.GTTGTATGTGATCGAAAAAAAAAAAAAA

Fig. 5. DNA sequence and deduced amino acid sequence of clone pHV AI-1. The eDNA was sequenced as described under Materials and methods. Nucleotide + 1 was assigned to the predicted translation initiation codon, and the negative number refers to the 5' noncoding region. The nine imperfect l l amino acid repeats are underlined and numbered. The arrow indicates the deletion site of a deletion mutant used in in vitro transcription/translation experiment. The polyadenylation signal, AATAAA, is underdotted.

trol, as estimated by densitometric scanning shown in Fig. 4B. The decline in level of pHV A1 m R N A after 12 h of ABA treatment was probably due to the depletion of ABA in the incubation medium. For the shorter time course, total R N A was isolated from aleurone layers treated with 2 × 1 0 -4 M ABA. As shown in Fig. 4C, the level of m R N A encoded by clone pHV A1 was already 3-fold over the control after 40 min of ABA treatment as observed in densitometric scanning in Fig. 4D.

DNA sequence of clone p H V AI-1 and its predicted amino acid sequence A Xgt 10 library was constructed from poly(A) ÷ R N A isolated from ABA-treated aleurone layers. It was screened with clone p H V A1 isolated from the p U C e D N A library to search for clones with longer

insert. Phage D N A isolated from plaques which hybridized to [32p]-labeled pHV A1 was digested with Eco RI, and separated by agarose gel electrophoresis. The nucleotide and deduced amino acid sequence of one positive clone designated pHV A1-1 are shown in Fig. 5. The first AUG (at position - 103) gives a very short open reading flame because of a stop codon 42 nucleotides downstream. The second methionine codon (+ 1) gives a long open reading frame of 642 nucleotides, and the deduced polypeptide contains 213 amino acid residues. The calculated molecular mass of this polypeptide is about 22 kDa in discrepancy with the molecular weight of 27 kDa on SDS-polyacrylamide gel (see Discussion). The polypeptide has compositional bias for Lys (13.6%0), Thr (16.4%) and Ala (22.1°/0). These three amino acids compose 52% of the polypeptide sequence. Other residues such as Cys, Phe, Pro and Trp are lacking. Nine imperfect repeats of the 11 ami-

503 vector and transcribed in vitro using T7 R N A polymerase. The transcripts were translated in rabbit reticulocyte system. As shown in Fig. 6, a prominent b a n d o f 27 k D a in mass (lane 3) comigrates with the polypeptide obtained by hybrid-select translation (lane 1). The 27 k D a b a n d was not observed in Bluescript vector only control (lane 2) or a subclone with the c D N A in reverse orientation (lane 4). In order to demonstrate that the A U G at + 1 is the initiation c o d o n o f translation, the c D N A was cleaved at nucleotide + 26 (indicated by arrow in Fig. 5) by restriction enzyme A l u I, and the sequence f r o m + 27 to the 3 ' end was cloned into Bluescript at S m a I and Eco RI. This deletion was also transcribed and translated in vitro. The 27 k D a polypeptide was not p r o d u c e d in the deletion m u t a n t (lane 5), indicating A U G at + 1 is the initiation c o d o n for the 27 k D a protein. The discrepancy between the predicted and apparent molecular mass o f the 27 k D a protein could be due to an artifact o f gel electrophoresis as reported for some heat shock proteins [31].

Genomic southern analysis with p H V A1 Fig. 6. Comparison between the hybrid-select translation product of clone pHV AI and the translation product of mRNA generated by in vitro transcription of pHV Al-l. The cDNA insert of pHV AI- 1 was subcloned into Bluescript KS(+) at the Eco RI site. Bluescript vector only and the Bluescript with positive or negative orientation of cDNA to T7 promoter were transcribed and translated in vitro, and the products were loaded on lanes 2, 3 and 4 respectively. Lane 5 represents the product of a deletion mutant (beginning at nucleotide +27 as indicated by an arrow in Fig. 5) that lacks + 1 AUG which is believed to be the translation initiation codon of the 27 kDa polypeptide. Molecular weight markers in kDa are shown on the right edge of the gel.

Ten/~g o f genomic D N A was digested with restriction enzymes indicated in Fig. 7. The restriction fragments were separated by agarose gel electrophoresis, blotted and hybridized with nick-translated p H V A1 c D N A . A 3.2 kb b a n d was very intensified in Barn HI-digested barley (lane 3). A b a n d o f 2.7 k D a was observed with Eco RI-digested D N A (lane 2), and at least 2 bands are present when the barley D N A was digested with Hind III.

Discussion no acids T h r - G l u - A l a - A l a - L y s - G l n - L y s - A l a - A l a G l u - T h r are present in this polypeptide.

Polypeptide obtained by in vitro transcription~translation In order to c o n f i r m that p H V A1 contains the initiation c o d o n o f translation and the entire coding sequence, the c D N A was subcloned into the Bluescript

It has been shown that A B A induces several polypeptides in barley aleurone layers [17]. To understand the m e c h a n i s m o f A B A action, it is essential to isolate and characterize these polypeptides and their genes. By differential screening using [32p]-labeled m R N A prepared f r o m b o t h ABA-treated and control aleurone layers, we have isolated an ABA-induced clone, p H V A1. In N o r t h e r n analysis p H V A1 hybridizes to a 1.1 kb m R N A species. The dosage re-

504

Fig. 7. Southern analysis of barley genomic DNA probed with clone pHV A1. Ten #g o f etiolated barley seedling DNA was digested with Hind III (lane 1), Barn HI (lane 2), or Eco RI (lane 3), separated on an agarose gel, transferred onto a GeneScreen membrane and hybridized with the [32p]-labeled insert o f clone pHV A1. The size markers indicated on the right are )xphage DNA digested with H i n d III.

sponse experiment shows that pHV A1 m R N A can be induced with an ABA concentration as low as 10-9 M. Hybrid-select translation results indicate that a basic protein with an apparent molecular mass of 27 kDa on SDS gel is translated from the selected mRNA. From two-dimensional gel analysis of in vitro translated total RNA, the pHV A1 encoded protein does not appear to be the 29 kDa protein, the most abundant protein in ABA-treated layers, or the "lectin-like" 36 kDa protein [17]. Likewise, the pHV

A1 encoded protein did not cross-react with antibody against barley a-amylase inhibitor (L. S. Lin, unpublished). Another protein that bears similarity in molecular weight to the polypeptide-encoded by pHV A1 is the salt stress-induced protein in tobaccocultured cells [28]. However, pHV Al-encoded protein did not precipitate with the antibody against the tobacco salt stress protein (data not shown). In vitro transcription and subsequent translation demonstrate that the whole coding sequence of the 27 kDa polypeptide (although the reading frame actually encodes only a 22 kDa polypeptide, this protein is still referred to as the "27 kDa protein" to maintain the consistency in literature) was contained in pHV AI-1. The AUG at + 1 position functions as the initiating codon for translation. The nucleotide and amino acid sequences have been compared in a computer data base (Genetics Computer Group, University o f Wisconsin, Madison, WI), and no significant homology to other known sequences has been found in this database. This polypeptide is comparatively rich in lysine (13.6°70). Another feature is that it contains nine imperfect repeats of 11 amino acids Thr-Glu-Ala-Ala-Lys-Gln-Lys-AlaAla-Glu-Thr. This amino acid sequence shares homology to the 11 amino acids in the direct, imperfect repeats of the cotton Lea 7 protein (1). Such amino acid sequence similarities found in both monocots and dicots might imply some common function of these proteins in seed development and some other physiological processes. In plants, systems with rapid response upon application of plant hormones have attracted great attention because they are likely to reveal the primary actions of hormones at the molecular level. The auxin responding system has been studied mostly intensively, cDNA clones of rapid auxin-induced mRNA (within 10-30 min) have been obtained from soybean hypocotyl and pea epicotyl tissue [32]. Our data indicated that three-fold enhancement of mRNA encoded by pHV A1 was found within 40 min, and the level is 13-fold enhanced at 12 h. Thus, pHV A1 mRNA is rapidly induced by ABA, as compared with other ABA-induced clones reported. It was reported that in excised cotton embryos, some of the ABA-regulated mRNAs exhibit seven- to seventyfold increases within 38 h [11]. The induction of

505 clone pHV A1 mRNA by ABA is also relatively high compared to that of some storage protein genes, for example, about 3-5-fold enhancement was found for the ~-subunit of/3-conglycinin in soybean [5]. We expect to obtain antibody for this 27 kDa ABA-induced protein, and the immunohistochemistry analysis is expected to provide information about the cellular localization and the potential physiological role of this protein.

Acknowledgements This work was supported by NSF grant DCB 8326319. We would like to express our thanks to A. Eggenberger for help in DNA sequencing, to G. Heck for graphic work, to L. S. Lin for encouragement, and to J. Rogers for helpful advice.

References 1. Baker J, Steele C, Dure L III: Sequence and characterization of 6 Lea protein and their genes from cotton. Plant Mol Biol 11:277-291 (1988). 2. Belanger FC, Brodl MR, Ho THD: Heat shock causes destabilization of specific mRNAs and destruction of endoplasmic reticulum in barley aleurone cells. Proc Natl Acad Sci USA 83:1354-1358 (1986). 3. Blin N, Stafford DW: A general method for isolation of high molecular weight DNA for eukaryotes. Nucl Acids Res 3: 2303-2308 (1976). 4. Birnboim HC, Doly J: A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl Acids Res 7: 1513-1523 (1979). 5. Bray EA, Beachy RN: Regulation by ABA of/3-conglycinin expression in cultured developing soybean cotyledons. Plant Physiol 79:746-750 (1985). 6. Church GM, Gilbert W: Genomic sequencing. Proc Natl Acad Sci USA 81:1991-1995 (1984). 7. Cooper P, Ho THD, Hauptman RM: Development and tissue specificity of the heat-shock response in maize. Plant Physiol 75:431-441 (1984). 8. Crouch ML, Sussex IM: Development and storage-protein synthesis in Brassica napus L. embryos in vivo and in vitro. Planta 153:64-74 (1981). 9. Crouch ML, Tenbarge K, Simon R, Finkelstein R, Scofield S, Solberg L: Storage protein mRNA levels can be regulated by abscisic acid in Brassica embryos. In: Van Vloten-Doting L, Groot GSP, Hall TC (eds) Molecular Form and Function of the Plant Genome, pp. 555-565. Plenum Press, New York (1985).

10. Donnis-Keller H, Maxam AM, Gilbert W: Mapping of adenines, guanines and pyrimidines in RNA. Nucl Acid Res 4: 2527-2538 (1977). 11. Galau GA, Hughes DW, Dure L III: Abscisic acid induction of cotton late embryogenesis-abundant (Lea) mRNAs. Plant Mol Biol 7:155-170 (1986). 12. Heidecker G, Messing J: Sequence analysis of zein cDNAs obtained by an efficient mRNA cloning method. Nucl Acids Res 11:4891-4906 (1983). 13. Higgins TJV, Jacobsen TV, Zwar JA: Gibberellic acid and abscisic acid modulated protein synthesis and mRNA level in barley aleurone layers. Plant Mol Biol 1:191-215 (1982). 14. Ho THD, Varner JE: Response of barley aleurone layers to abscisic acid. Plant Physiol 57:175-178 (1976). 15. Jen G, Thach RE: Inhibition of host translation in Encephalomyocarditisvirus-infected L cell: a novel mechanism. J Virol 43:250-261 (1982). 16. Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 (1970). 17. Lin LS, Ho THD: Mode of action of abscisic acid in barley aleurone layers induction of new proteins by abscisic acid. Plant Physiol 82:289-297 (1986). 18. Maniatis T, Fritsch EF, Sambrook J: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1982). 19. Miller JS, Paterson BM, Ricciardi RP, Cohen L, Roberts BE: Methods utilizing cell-free protein-synthesizing systems for the identification of recombinant DNA molecules. In: Methods in Enzymology Vol 101:650-674 (1983). 20. Mozer T J: Control of protein synthesis in barley aleurone layers by the plant hormones gibberellic acid and abscisic acid. Cell 20:479-485 (1980). 21. Mundy J, Rogers JC: Selective expression of a probable amylase/protease inhibitor in barley aleurone cells: comparison to the barley amylase/subtilisin inhibitor. Planta 169:51-63 (1985). 22. O'Farrell PH: High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007-4012 (1975). 23. Partridge J, Shannon L, Grumpf D: A barley lectin that binds free amino sugars. 1. Purification and characterization. Biochim Biophys Acta 451:470-483 (1976). 24. Raikhel NV, Quatrano RS: Localization of wheat-germ agglutinin in developing wheat embryos and those cultured in abscisic acid. Planta 168:433-440 (1986). 25. Rogers JC: Two barley c~-amylasegene families are regulated differently in aleurone cells. J Biol Chem 260:3731-3738 (1985). 26. Rogers JC, Milliman C: Coordinate increase in major transcripts from the high pI c~-amylasemultigene family in barley aleurone cells stimulated with gibberellic acid. J Biol Chem 259:12234-12240 (1984). 27. Sanger FS, Nicklen S, Coulson AR: DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74: 5463-5467 (1977). 28. Singh NK, LaRosa PC, Handa AK, Hasegawa PM, Bressan RA: Hormonal regulation of protein synthesis associated

506

29.

30.

31.

32. 33.

34.

with salt tolerance in plant cells. Proc Natl Acad Sci USA 84: 739-743 (1987). Southern EM: Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98: 503-517 (1975). Southgate R, Ayme A and Voellmy R." Nucleotide sequence analysis of the Drosophila small heat shock gene cluster at locus 67B. J Mol Biol 165:35-37 (1983). Sussex IM, Dale RMK, Crouch ML: Developmental regulation of seed storage protein synthesis in seeds. In: Leaver CJ (ed) Genome Organization and Expression in Plant, pp. 283-289. Plenum Press, New York (1980). Theologis A: Rapid gene regulation by auxin. Ann Rev Plant Physiol 37:407-438 (1986). Uknes SJ, Ho THD: Mode of action of abscisic acid in barley aleurone layers: abscisic acid induces its own conversion to phaseic acid. Plant Physiol 75:1126-1132 (1984). Varner JE, Ho THD: The role of hormones in the integration

35. 36.

37.

38.

of seedling growth. In: Papaconstantinou J (ed) The Molecular Biology of Hormone Action, pp. 173-174. Academic Press, New York (1976). Walton DC: Biochemistry and physiology of abscisic acid. Ann Rev Plant Physiol 31:453-489 (1980). Watson C J, Jackson JF: An alternative procedure for the synthesis of double-stranded cDNA for cloning in phage and plasmid vectors. In: Glover DM (ed) DNA Cloning Techniques: A Practical Approach, pp. 79-88. IRL Press, Washington DC (1985). Weselake R J, MacGregor AW, Hill RD, Duckwirth HW: Purification and characteristic of endogenous c~-amylaseinhibitor from barley kernels. Plant Physiol 73:1008-1012 (1983). Williamson JD, Quantrano RS, Cuming AC: Em polypeptide and its messenger RNA levels are modulated by abscisic acid during embryogenesis in wheat. Eur J Biochem 152:501-507 (1985).