A novel embryo-specific barley cDNA clone encodes a protein with homologies to bacterial glucose and ribitol dehydrogenase

Planta (1994)192:519-525 P l ~ n m

�9 Springer-Verlag 1994

A novel embryo-specific barley eDNA clone encodes a homologies to bacterial glucose and ribitol dehydrogenase Roland Alexander, Josefa M. Alamillo, Francesco Salamini, Dorothea Bartels Max-Planck-Institut f/ir Ziichtungsforschung, Carl-von-Linn&Weg 10, D-50829 K61n, Germany

Received: 20 September 1993 / Accepted: 8 October 1993

protein with

Abstract. In order to analyze the genetic programme ex- pressed during the early stages of embryogenesis a eDNA clone bank was constructed from desiccation-tolerant ex- cised barley embryos 18d after pollination (D. Bartels et al., 1988, Planta 175, 485-492). One of the selected eDNA clones pG31 encodes a transcript of 1300 nude- otides and a protein of 31 kDa, both are specifically ex- pressed in developing embryos and are not detected in other tissues. The expression of the pG31 mRNA is not modulated by the plant hormone eis-abscisic acid but it ceases to be expressed in germinating embryos. The protein sequence deduced from the pG31 transcript shows substantial sequence homologies to bacterial glu- cose dehydrogenase and ribitol dehydrogenase. Bio- chemical analysis indicates that glucose dehydrogenase activity is present in protein extracts from embroys 18d after pollination. This glucose dehydrogenase activity is inhibited by antiserum raised against the recombinant pG31 protein. These findings provide evidence for the discovery of a novel pathway in carbohydrate metabolism acting specifically during embryogenesis.

Key words: Embryo specific transcript Glucose/ribitol dehydrogenase - Hordeum (embryo development)

Introduction

Plant embryogenesis involves the commitment of cells to particular developmental programmes. Different stages of embryogenesis are characterized by the expression of specific sets of genes (Goldberg et al. 1989). Among the regulatory factors coordinating gene expression during embryogenesis the plant hormone eis-abscisic acid

Abbreviations : ABA = cis-abscisic acid; DAP = days after pollina- tion; GlcDH=glucose dehydrogenase; LEA=late embryogenesis abundant Correspondence to: D. Bartels; FAX: 49(221)5062413

(ABA) has been identified as one of the potential signals (Quatrano 1986; Black 1991). Many of the isolated em- bryo-specific genes are inducible by exogenous ABA, and studies of their promoters led to the identification of conserved ABA-responsive elements (Marcotte et al. 1989; Mundy et al. 1990; Lam and Chua 1991). One of the best-studied genes in this group is the Em gene of wheat (Cuming 1984; Litts et al. 1987). A putative tran- scription factor for this gene represents a leucine zipper protein which binds to the ABA-responsive elements of the Em gene (Guiltinan et al. 1990). Other factors mediat- ing the activation of embryo maturation genes are re- presented by the maize Vpl gene product and the Ara- bidopsis homologue Abi3 which encode potential tran- scription activators (McCarty et al. 1991 ; Giraudat et al. 1992). Evidence that ABA levels are not the only factors regulating the embryogenesis programme comes from a careful examination of cotton embryogenesis which sug- gests that an unidentified maternal maturation factor and a post-abscission factor may be involved in the regulation of embryogenesis (Hughes and Galau 1991).

A specific feature of embryogenesis is the acquisition of desiccation tolerance. We are using immature develop- ing barley embryos as a model system to identify genes leading to desiccation tolerance. Embryos isolated from barley grains 12 d after pollination (DAP) do not ger- minate after a severe dehydration treatment (< 10% water content), but desiccated 18-DAP embryos ger- minate and are defined as desiccation-tolerant embryos (Barrels et al. 1988). This change in the sensitivity to water loss occurs before the embryo starts to loose water and before most of the late-embryogenesis-abundant (LEA) genes are expressed (Galau et al. 1986; Dure et al. 1989; Skriver and Mundy 1990). In barley a subset of new gene products was observed when the protein ex- pression of 18-DAP embryos was compared with that of 12-DAP embryos (Barrels et al. 1988). A cDNA bank was constructed from 18-DAP embryos and eDNA clones abundantly expressed in 18-DAP embryos were selected. One of these clones was modulated by ABA and displayed substantial structural homologies to a gene

520 R. Alexander et al.: A barley cDNA clone encodes a glucose dehydrogenase

family encoding N A D P H - d e p e n d e n t aldose reductase (Bartels et al. 1991). In this paper we repor t the charac- terization o f a second barley c D N A clone which is ex- pressed in 18-DAP embryos at similar levels as the aldose reductase m R N A ; in cont ras t to the latter transcript a high expression level was already found in 12-DAP em- bryos and exogenous A B A does not appreciably change its level o f t ranscript ion. The gene described here for the first time in plants is embryo specific and encodes a prote in with s tructural h o m o l o g y to bacterial glucose and ribitol dehydrogenase . Biochemical evidence is provided that in vivo the gene encodes a protein with glucose dehydrogenase (G l c DH) activity.

Materials and methods

Plant material. Culture of barley (Hordeum vulgare L. cv. Aura) plants and isolation of stage-specific barley embryos are described by Bartels et al. (1988). If required, 10 -4 M ABA was added to the embryo culture (germination) medium (GM).

Isolation of cDNA clones. A cDNA library was constructed using poly(A) § RNA from 18-DAP barley embryos. The library was dif- ferentially screened using probes from 12-DAP (not desiccation tolerant) and 18-DAP (desiccation tolerant) embryos (Bartels et al. 1991). For the isolation of full-length cDNA clones a second library was constructed in the lambda zap vector system (Stratagene, Heidelberg, Germany).

Recombinant-DNA techniques. Isolation of plasmid DNA, prepara- tion of DNA fragments, ligation and transformation of Eseherichia coli were carried out according to Maniatis et al. (1982). Barley RNA and DNA were isolated and used in Northern and Southern blot experiments as described by Bartels et al. (1991). The DNA probes were radioactively labelled according to Feinberg and Vogelstein (1983).

Sequencin 9 (?f DNA and computer analysis'. Nucleotide sequences were determined on both strands by subcloning of restriction frag- ments into pUC19 (Messing and Vieira 1982) followed by dide- oxynucleotide sequencing with the T7 polyrnerase kit (Pharmacia LKB, Freiburg, Germany). The program WISGEN of the Univer- sity of Wisconsin genetic's computer group (Madison, Wis., USA) was used for nucleic-acid and protein-sequence analysis (Devereux et al. 1984), and amino-acid comparisons were done with the TFAS- TA program (Pearson and Lipman 1988).

Construction of E. coli expression clones, purification of recombinant fusion proteins and raisin9 of polyclonal antibodies. The insert of pG31 was ligated into the Smal site of the E. coli plasmid expression vector pGEX-2 (Smith and Johnson 1988) to yield a translational fusion with the glutathione-S-transferase gene (GST). The ex- pression of the fusion protein was induced by the addition of isopropyl-~-D-thiogalactopyranoside (IPTG) to 0.4 mM. Purifica- tion of the fusion protein by preparative gel electrophoresis and electroelution, immunization and Western blot analysis were per- formed as described by Bartels et al. (1991).

In-situ analysis'. The insert of pG31 was subcloned into the pBlue- script plasmid vector. The recombinant cDNA clone was linearized and transcribed in sense and antisense directions in the presence of ct-[3sS]UTP (Transprobe SP Kit; Pharmacia LKB) using SP6- or T7-RNA-Polymerase. Further treatment of transcripts, preparation of sections, hybridization conditions and photomicroscopy were as described by Schneider et al. (1993).

Determination of enzymatic activity. For assay of GlcDH activity, barley embryos were ground to a fine powder under liquid nitrogen.

Proteins were extracted with adequate amounts of 150 mM Tris- HC1 buffer (pH 7.5). The suspension was centrifuged (I 1 500 - 9 for 15 rain at 4 ~ C). The supernatent was desalted on a Sephadex G-50 column (medium, Pharmacia) and subsequently the proteins were concentrated using a Centricon microconcentrator. The protein concentration was determined according to Bradford (1976) using the BioRad assay kit. Glucose dehydrogenase (GlcDH) activity was assayed spectrophotometrically in the protein extract following the formation of NAD(P)H at 340 or 366 nm at 30 ~ C. The standard assay mixture contained 150 !umol Tris, 50 lamol D-glucose, 5 p.mol NAD and the appropriate amounts of protein extracts in 1 ml. One unit of GIcDH activity was defined as the amount of enzyme producing 1 lamol NAD(P)H per minute.

Immunoprecipation. Aliquots of protein extracts from 18-DAP em- bryos were incubated overnight at 4~ with antiserum raised against the pG31 recombinant protein. When indicated the pre- cipitated proteins were isolated with protein A-Sepharose.

Results

Isolation o f a cDNA clone with homologies to GlcDH and ribitol dehydrogenase. A c D N A library was constructed f rom p o l y ( A ) + R N A of 18-DAP desiccation-tolerant barley embryos (Bartels et al. 1991); a selected c D N A clone (pG31) encodes a m R N A abundant ly expressed at 1 8 D A P . The nucleotide and deduced amino-acid sequence o f this clone are presented in Fig. 1. One possi- ble open reading frame predicts a protein o f 31 kDa. D a t a b a n k searches revealed regions o f homologies spread th roughou t the coding region between pG31 and a representative bacterial G l c D H gene f rom Bacillus megaterium and ribitol dehydrogenase f rom Klebsiella aerogenes (Fig. 1). Out o f 256 compared amino acids, 32 % and 25 % amino acids, respectively, were identical to the barley pG31.

Genomic Southern analysis and expression pattern o f the pG31 transcript and protein. Barley genomic D N A was digested with three different restriction enzymes and probed with the insert o f pG31. The hybr idizat ion pat- tern suggests that pG31 is represented in one to three copies in the barley genome (Fig. 2).

Nor the rn blot experiments were performed to deter- mine the tissue-specific and developmental expression o f the pG31-encoded transcript. The R N A was abundan t ly expressed in embryos but no t in leaves, roots or endo- sperm. In the embryo the pG31 transcript started to accumulate early during embryogenesis and increased during the matura t ion phase (Fig. 3); at all stages a single transcript o f 1300 nucleotides was detected. Incuba t ion o f 12-DAP barley embryos with A B A did not affect the level o f pG31 transcripts, while precocious germinat ion o f 12- and 18-DAP embryos led to a drastic decrease o f the pG31 m R N A level (Fig. 3). Expression in leaves could not be triggered by osmotic-stress treatments.

Polyclonal antibodies raised against the pG31 recom- binant protein were used to moni to r the expression o f the pG31 protein during embryogenesis. Protein accumula- t ion followed the R N A pat tern and th roughou t em- bryogenesis a protein o f 31 k D a was detected (Fig. 4). The detected molecular mass corresponds to that de-

R. Alexander et al. : A barley cDNA clone encodes a glucose dehydrogenase 521

1 GTCGTCGCCAAGAGCACCGCCCGCT CGCCGGGGGACCAGAGCAATGGCGT CGCAGAAGTTCCCGCCGCAGCAG a M A S Q K F P P Q Q 75 CAGGACTGCCAGCCCGGCAAGGAGCACGCCATGGACCCCCGCCCCGAGGCCATCATCAAGAACTACAAGTCGG

a Q D C Q P G K E H A M D P R P E A I I K N Y K S

c M K H S V S S 150 GCCAACAAGCTCCAGGGCAAGGTGGCGCTGGTGACCGGCGGCGACTCGGGCATCGGGCGCGCGGTGTGCCTG

b c

22 T!CCiCOiGCiOO CTiCAIGT!CO!O OO!GClCOIGOICAI iCG G c b a

375 GGGTACGAGGAGAACTGCCGCAGGGTGGTGGAGGAGGTGGCCAACGCGCACGGCGGCCGCGTGGACATCCTC

b �9 ~ ~ ~ ~ v ~ �9 ~ ~ ~ ~ ~ ~ �9 .___=__~__/ ~ C M Q A D Q ~ D ~ L ~ G I L Q . L T ~ F

450 GTG~CAACGCGGCCGAGCAGTACGTCCGCCCCTGCATCACCGAGATCACCGAGCAGGACCTGGAGCGCGTG a

b Z v Pv s H ~ S ~ ~ c H : ~ G ~ v ~ G O ~ v o

525 TTCCGCACCAACATCTTCTCCTACTTCCTCATGACCAAGTTCGCCGTGAAGCACATGGGGCCCGGG: �9 : �9 : :

b X D T G G S E I Y F V E N D c L H N A C V S L H L I A Q K

600 : : :TCCAGCATCATCAACACCACCTCCGTGAACGCGTACAAGGGCAACGCGACGCTGCTGGACTACACGGCC

b K N H E M I W P L F V H c S D A G V V V I W E P V

675 ACCAAGGGCGCCAT CGTGGCCTTCACCCGCGCGCTGTCGATGCAGCTGGCGGAGAAGGGGATCCGCGTCAAC

a ~ ~ F ~ I V m ~ M V R ~ T S M E~ L ~ E ~ b M K E A L Y P C V Q H R R ~ V Q Y G V ~ G

7 5 0 GGCGTGGCGCCGGGGCCCATCTGGACGCCCCTCATCCCGGCCTCCTTCCCGGAG...GAGAAGGTGAAGCAG a G ~ A ~ I W ~ ' ~ I P A S ~ P E . ~ K V K Q b ~ ~ G ~---X ~ ~ ~ ~ ~ ~ ~ ~ ~ o ~ ~ Q ~ ~ o

A ~ ~ ~ V V ~m ~ ~ ~ . W . . . . ~ a ~ ~ 825 TTCGGGTCCGAGGTGCCCATGAAGCGCGCCATGCAGCCCAGCGAGGTCGCGCCCAGCTTCGTCTTCCTTGCC

b V E M I Y I E A A A c M D A L S L I E ~ V L

900 AGCGAGCAGGACTCCTCCTACAT CTCCGGCCAGATCCT CCACCCCAACGGTGGTACCATCGTCAATAGCTAG a!E!OSV S I!H L/vTIVNS* b S �9 A T T F M TKYPSFQAGRG* o R ~ N R v L ~ o L * 975 ATCGAGGTTGGAGAAGCTCGGG~GAACAGGGTGAAGTGTGCCCGTGGTGTGCGAGTCTGTAGTACGAGCAA

1050 GTGT~GCGTGT~AGTCTCGGAGTC~GCT~GTAGTGTTTTGTGTaGCAGTAGTACCGTTGTGTGZ~CGG~ 1 1 2 5 GACTTTTGGCGCTGGAGCCACGAGGGCAAGCGTCAAACGTTAAATA~AT~TGTAACGTGCATGCATTGCA

AGTTG~CTATG GTCATACTCCCTCTAT

Fig. 1. Nucleotide sequence (mRNA strand) and predicted amino-acid sequences of pG31. The nucleotide sequence is shown in the upper line and numbered, the corre- sponding amino-acid sequence (a) is given below and is aligned with amino- acid sequences from GlcDH of Bacillus mega- terium (b) (Heilmann et al. 1988) and ribitol dehydro- genase from Klebsiella aerogenes (c) (Loviny et al. 1985). Identical amino ac- ids are indicated in black. Gaps were introduced to optimize matches

duced from the cDNA clone. Like the m R N A levels the protein disappeared during germination, although the turnover of the protein appears to be slower in older embryos (Fig. 4, compare lanes 2 and 8). When embryo proteins were separated by a two-dimensional electro- phoresis the single band of the pG31 protein could be resolved into two spots with isoelectric points between 7.0 and 6.8 (Fig. 5).

The spatial distribution of the pG31 transcript in developing embryos was investigated by RNA in-situ hybridizations (Fig. 6). The transcript was detected in all tissues of the embryo with a particular high accumula- tion in the scutellum and a low expression in the coleo- rhiza. This distribution did not change in 12-DAP and 18-DAP embryos.

Glucose dehydrogenase activity. The homology of the protein deduced from the barley cD N A clone pG31 to bacterial GlcDH and ribitol dehydrogenases raised the question whether the presence of the barley protein can be linked with related enzymatic activity. The activity of GlcDH was measured in crude protein extracts of 12- DAP and 18-DAP embryo and leaf samples. The results from at least three independent experiments are shown in Fig. 7. The highest level of activity was found in 18- DAP embryos, while 12-DAP embryos gave a lower activity. No activity was found in germinated 12- or 18-DAP embryos and in leaves. The enzymatic activity in the 12-DAP and 18-DAP embryos was dependent on glucose as a substrate. Addition of antiserum against the pG31 recombinant fusion protein to the enzyme assays

522 R. Alexander et al. : A barley cDNA clone encodes a glucose dehydrogenase

Fig. 4. Western blot analysis of the pG31 protein during barley embryo development. Protein extracts were subjected to PAGE and analysed via immunoblotting in the presence of the pG31 protein for the following samples : lane 1, 12-DAP embryos; lane 2, 12-DAP embryos on germination medium for 3 d; lane 3, 14-DAP embryos; lane 4, 16-DAP embryos; lane 5, 18-DAP embryos; lane 6, 20-DAP embryos; lane 7, 30-DAP embryos; lane 8, 30-DAP embryos on germination medium for 3 d; lane 9, mature embryos. Reference molecular-size markers are given in kDa

Fig. 2. Southern analysis of genomic DNA from barley. The DNA was cut with Bam HI (1), Eco RI (2) and Hind III (3) and probed with the 32p-labelled insert of pG31; the fragment sizes are given in kilo base-pairs

Fig. 5. Western blot analysis of barley embryo proteins extracted from 18-DAP embryos and separated by two-dimensional electro- phoresis. The embryo proteins were first separated by isoelectric focusing (IEF) and then in a 12% SDS-polyacrylamide gel (SDS- PAGE) according to Bartels et al. (1988); the proteins were trans- ferred to a nitrocellulose membrane, which was incubated with antibodies raised against the pG31-encoded protein

Fig. 3. A Expression of the pG31 transcript in the barley embryo. RNAs (1 lag poly(A) +) from embryos of different developmental stages were separated on a denaturing agarose gel, blotted to a nylon membrane and hybridized with the 32P-labelled fragment of pG31. RNAs were extracted from the following tissues: lane 1, 12-DAP embryos; lane 2 12-DAP embryos incubated on germina- tion medium for 3 d; lane 3, 15-DAP embryos; lane 4, 18-DAP embryos; lane 5, 18-DAP embryos on germination medium; lane 6, 30 DAP-embryos. B The same blot as in A was hybridized with a ribosomal probe to assure equal loadings of RNA

reduced the G l c D H act iv i ty subs tan t i a l ly ; af ter r emov ing the p ro t e in - an t i gen complex wi th P ro te in A - S e p h a r o s e no ac t iv i ty was f o u n d in the s u p e r n a t a n t (Table 1). W h e n , in con t ro l exper iments , p r e i m m u n e serum was

used ins tead o f an t i se rum the supe rna t an t re ta ined the G l c D H activity.

The abi l i ty o f different sugars to reduce N A D ( P ) was tested in enzyme assays using bo th p ro te in ext rac ts f rom 18-DAP embryos and ge rmina ted embryos . N A D was reduced by D-glucose only in 18 -DAP e m b r y o extracts but no t in extracts f rom ge rmina t ed e m b r y o s ; this speci- ficity was no t found for D-sorbi tol , D-xylose, D-fructose and D-ribi tol for which N A D reduc t ion was observed bo th with extracts f rom 18-DAP embryos and with ger- m ina t ed embryos as enzyme sources (Fig. 8). W i t h re- spect to the specificity o f the co-subs t ra te in the presence o f D-glucose, bo th N A D and N A D P can be reduced, a l t hough N A D P at lower efficiency (40%). The a p p a r e n t Km values for G l c D H were de t e rmined to be 3 m M for glucose and 0.3 m M for N A D . The act ivi ty o f G l c D H was no t affected by an i ncuba t ion up to 56 ~ C for 10 min whereas the act ivi ty using D-ribi tol was abo l i shed at this t empera tu re .

R. Alexander et al.: A barley eDNA clone encodes a glucose dehydrogenase 523

Table 1, Glucose dehydrogenase (GlcDH) activity in protein ex- tracts of 18-DAP barley embryos after the addition of antiserum and protein A-Sepharose (Prot. A.S.)

Serum added Prot. A.S. GlcDH activity (mg) (%)

none none 100% 2.5 lal immune serum none 30% 5.0 lal immune serum none 12% 2.5 gl immune serum 2.5 mg 0% 5.0 gl immune serum 2.5 mg 0% 2.5 tal preimmune serum 2.5 mg 100% 5.0 lal preimmune serum 2.5 mg 100%

Fig. 6A, B. In-situ RNA hybridizations of the pG31 transcript in developing barley embryos. Cross-sections of 18-DAP barley em- bryos were hybridized with pG31 3sS-labelled sense (A) and anti- sense (B) transcripts. Indicated in A are different embryonic tissues : a, embryonic roots; b, shoot apex; c, scutellar tissue; d, coleorhiza; • bar=0.11 mm

12

9 7

E

6

s

3 < z

GIe Sor Xyl Fru Rib

Fig. 8. Sugar specificity in the enzymatic assay: the reduction of NAD was measured in protein extracts of 18-DAP embryos (shaded bars) and of 18-DAP embryos germinated (open bars) in the presence of D-glucose (Glc), D-sorbitol (Sor), D-xylose (Xyl), D-fruc- tose (Fru) and D-ribitol (Rib). The results represent the average values of several independent experiments

4.5

r

' 3 .0

o 31.5

0.0 18DAP lSGM 12DAP 12GM L e a v e s

Fig. 7. Glucose dehydrogenase activities in protein extracts prepared from different barley tissues: 18-DAP embryos (18-DAP), 18-DAP embryos germinated for 3 d (18GM), 12-DAP embryos (12DAP), 12-DAP embryos germinated for 3 d (12GM) and leaves. The enzyme assays were carried out with glucose (shaded bars) and without glucose (open bars)

Discussion

In the search for sequences which are expressed early dur ing barley embryo development and possibly asso- ciated with the acquisi t ion o f desiccation tolerance, we have identified a novel barley e D N A clone, pG31, with h o m o l o g y to bacterial G l c D H and ribitol dehydroge- nase. This pG31 transcript is highly specific for em- bryogenesis ; it ceases to be expressed when immature or mature embryos are germinated. Moreover , in embryos the transcript could no t be detected in vegetative tissues or in endosperm, nor could its expression be triggered by osmot ic stress. In cont ras t to several o ther genes ex- pressed in the embryo the plant h o r m o n e A B A did not appreciably modula te the expression level o f pG31 tran- scripts and therefore its expression mus t be regulated independent ly o f ABA. While the ABA-induc ib le L E A - type transcripts accumula te dur ing late stages o f em- bryogenesis (Dure et al. 1989; Skriver and M u n d y 1990) the levels o f pG31 m R N A and protein did no t change appreciably between 12 D A P and matur i ty .

524 R. Alexander et al.: A barley cDNA clone encodes a glucose dehydrogenase

An interesting feature and a clue to the function of the pG31 gene is the similarity of the encoded protein with GlcDH, as well as ribitol dehydrogenase, from bacteria. Besides the 82 and 65 amino acids identical between the pG31 protein, GlcDH and ribitol dehydrogenases, re- spectively, all three sequences share a number of similar amino acids. Glucose dehydrogenase catalyses the oxida- tion of D-glucose using NAD or N A D P as co-substrate and is considered a key enzyme during the early stage of sporogenesis of Bacillus megaterium (Jany et al. 1984). Meanwhile four isoforms of GlcDH with possibly dif- ferent physiological functions have been identified from B. megaterium (Nagao et al. 1992). Ribitol dehydroge- nase catalyses the oxidation of ribitol and the structural gene was isolated from the bacterium Klebsiella aer- ogenes (Loviny et al. 1985). Both, GlcDH and ribitol dehydrogenase are structurally related and can be grouped into the superfamily of short alcohol-polyol- sugar dehydrogenases (J6rnvall et al. 1984). Our finding is the first report that this family of genes extends to plants. The secondary structure of GlcDH has been analysed in detail (J6rnvall et al. 1984; H6nes et al. 1987): the NAD(P)+-binding domain, as well as motifs involved in the dimerisation of subunits, seem conserved in the predicted barley protein. These observations sup- port the conclusion that functionally important D N A sequences have been conserved among plants and bac- teria, and similar metabolic pathways are also expected to be active in both types of organism. Besides the sequences discussed above, GlcDH genes have been characterized from other sources such as Acinetobacter calcoaceticus (Cleton-Jansen et al. 1989), Archebacteria (Bright et al. 1993) and Drosophila (Krasney et al, 1990), but they do not display significant structural homology to the barley pG31 protein or the glucose dehydrogenase from Bacillus; possibly these proteins are involved in different metabolic pathways.

It is widely accepted that mainly sugar-phosphates are metabolized during plant carbohydrate metabolism (e.g. Turner and Turner 1980). To obtain an indication that the GlcDH-related transcript and protein described here are involved in a novel pathway, not involving phos- phorylated intermediates, protein fractions were analysed for their abilities to oxydize D-glucose in the presence of NAD(P) +. Specific enzymatic activity was found in extracts from embryos but not from leaves. These biochemical data are in agreement with the ex- pression pattern of the pG31 transcript and proteins (Figs. 3, 4), Although the enzymatic measurements were done with crude protein fractions, the specificity of the assay is largely substantiated by the fact that the GIcDH activity was inhibited by the addition of pG31-specific antiserum. When different sugars were tested as sub- strates, only D-glucose was oxidized specifically in em- bryo protein extracts whereas, in the presence of other substrates, N A D was reduced by extracts from embryos and germinated embryos. This observation suggests that the dehydrogenase activity in embryos related to the gene pG31 is restricted to D-glucose as substrate. This is in agreement with the degree of sequence homology found: pG31 seems more homologous to GlcDH than to ribitol

dehydrogenase. Further biochemical characterization (apparent Km values and temperature stability) supports the relatedness between the barley protein and the characterized bacterial GlcDH enzymes (Smith et al. 1989; Nagao et al. 1992).

The GlcDH-related cDNA clone is a novel sequence specifically related to embryogenesis and pointing to a metabolic pathway not yet described for higher plants. Previously, an aldose-reductase-related barley embryo transcript was isolated and characterized by our group (Bartels et al. 1991). Although the function of this gene is still not clear, specific enzymatic activity can be attrib- uted to the aldose-reductase-encoded protein (data not shown). Both the GIcDH and the aldose-reductase- related transcripts are abundantly expressed in develop- ing barley embryos, reaching their maximum steady-state levels before the well-characterized LEA transcripts (Hughes and Galau 1989; Skriver and Mundy 1990). This indicates that the pG31- and aldose-reductase- related genes contribute to specific functional require- ments of early embryo development and suggests the existence in the carbohydrate metabolism specific to em- bryogenesis of pathways that have not yet been charac- terized.

We thank B. Eilts and M. Feck for excellent technical assistance and M. Pasemann for patiently typing the manuscript. J. Alamillo acknowledges the receipt of a grant from the European Economic Community in the Human Capital and Mobility Program and a Formacion de Personal Investigador-grant from Ministerio de Edu- cacion y Ciencia (Spain).

References

Bartels, D., Singh M., Salamini, F. (1988) Onset of desiccation tolerance during development of the barley embryo. Planta 175, 485-492

Bartels, D., Engelhardt, K., Roncarati, R., Schneider, K., Rotter, M., Salamini, F. (1991) An ABA and GA modulated gene expressed in the barley embryo encodes an aldose reductase. EMBO J. 1037, 1043

Black, M. ( 1991) Involvement of A BA in the physiology of develop- ing and mature seeds. In : Abscisic acid physiology and biochem- istry, pp. 99-124, Davies, W.J., Jones, H.G., eds. Bios Scientific Publishers, Oxford, UK

Bradford, M.M. (1976) A rapid and sensitive method for the quan- tification of microgram quantities of protein utilizing the princi- ple of protein-dye binding. Anal. Biochem. 72, 248-254

Bright, J.R., Byrom, D., Danson, M.J., Hough, D.W., Towner, P. (1993) Cloning, sequencing and expression of the gene encod- ing glucose dehydrogenase from the thermophilic archaeon Thermoplasma acidophilum. Eur. J. Biochem. 211, 549-554

Cleton-Jansen, A.-M., Goosen, N., Vink, K., van de Putte, P. (1989) Cloning, characterization and DNA sequencing of the gene encoding the Mr 50000 quinoprotein glucose dehydrogenase from Aeinetobacter calcoaceticus. Mol. Gen. Genet. 217, 430-436

Cuming, A.C. (1984) Developmental regulation of gene expression in wheat embryos. Eur. J. Biochem. 145, 351-357

Devereux, J., Haeberli, P., Smithies, O. (1984) A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12, 387-395

Dure III, L., Crouch, M., Harada, J., Ho, T.H., Mundy, J., Qua- trano, R.S., Thomas, T., Sing, Z.R. (1989) Common amino acid sequence domains among the lea proteins of higher plants. Plant Mol. Biol. 12, 475-486

R. Alexander et al. : A barley cDNA clone encodes a glucose dehydrogenase 525

Feinberg, A.P., Vogelstein, B. (1983) A technique for radiolabelling DNA restriction endonuclease fragments to high specific activ- ity. Anal. Biochem. 132, 6-13

Galau, G.A., Hughes, D.W., Dure III, L. (1986) Abscisic acid induction of cloned cotton late embryogenesis abundant (Lea) mRNAs. Plant Mol. Biol. 7, 155-170

Giraudat, J., Hauge, B.M., Valon, C., Smalle, J., Parcy, F., Good- man, H.M. (1992) Isolation of the Arabidopsis ABI3 gene by positional cloning. Plant Cell 4, 1251-1261

Goldberg, R.B., Barker, S.J., Perez-Grau, H. (1989) Regulation of gene expression during plant embryogenesis. Cell 56, 149-160

Guiltinan, M.J., Marcotte, W.R., Quatrano, R.S. (1990) A plant leucine zipper protein that recognizes an abscisic acid response element. Science 250, 267-271

Heilmann, H.J., Mggert, H.J., Gassen, H.G. (1988) Identification and isolation of glucose dehydrogenase genes of Bacillus mega- terium M1286 and their expression in Escherichia coil Eur. J. Biochem. 174, 485-490

H6nes, J., Jany, K.D., Pfleiderer, G., Wagner, A.F.V. (1987) An integrated prediction of secondary, tertiary and quaternary structure of glucose dehydrogenase. FEBS Lett. 212, 193-198

Hughes, D.W., Galau, G.A. (1989) Temporally modular gene ex- pression during cotyledon development. Genes Dev. 3, 358-369

Hughes, D.W., Galau, G.A. (1991) Developmental and environ- mental induction of Lea and LeaA mRNAs and the post ab- scission program during embryo culture. Plant Cell 3, 603-618

Jany, K,D., Ulmer, W., Fr6schle, M., Pfleiderer, G. (1984) Com- plete amino acid sequence of glucose dehydrogenase from Bacillus rnegaterium. FEBS Lett. 165, 6-10

Jtrnvall, H., yon Bahr-Lindstrtm, H., Jany, K.D., Ulmer, W., Frtschle, M. (1984) Extended superfamily of short alcohol- polyol-sugar dehydrogenases: structural similarities between glucose. FEBS Lett. 165, 190-196

Krasney, P.A., Carr, C., Cavener, D.R. (1990) Evolution of the glucose dehydrogenase gene in Drosophila. Mol. Biol. Evol. 7, 155-177

Lam, E., Chua, N.H. (1991) Tetramer of a 21-base pair synthetic element confers seed expression and transcriptional enhance- ment in response to water stress and abscisic acid. J. Biol. Chem. 266, 17131-17135

Litts, J.C., Colwell, G.C., Chakerian, R., Quatrano, R.S. (1987) The nucleotide sequence of a cDNA clone encoding the wheat Em protein. Nucl. Acids Res. 15, 3607 3618

Loviny, R., Norton, P.M., Hartley, B.S. (1985) Ribitol dehydroge- nase of Klebsiella aerogenes. Biochem. J. 230, 579-585

Maniatis, T., Fritsch, E.F., Sambrock, J. (1982) Molecular cloning: a laboratory manual. Cold spring Harbour laboratory Press, N.Y., USA

Marcotte, W.R., Russel, S.M., Quatrano, R.S. (1989) Abscisic acid-responsive sequences from the Em gene of wheat. Plant Cell 1, 969-979

McCarty, D.R., Hattory, T., Carson, C.B., Vasil, V., Lazar, M., Vasil, I.K. (1991 ) The Viviparous I developmental gene of maize encodes a novel transcriptional activator. Cell 66, 895-905

Messing, J., Vieira, J. (1982) A new pair of U 13 vectors for selecting either DNA strand or double-digest restriction fragments. Gene 19, 269-278

Mundy, J., Yamaguchi-Shinozaki, K., Chua, N.H. (1990) Nuclear proteins bind conserved elements in the abscisic acid-responsive promoter of a rice rab gene. Proc. Natl. Acad. Sci. USA 87, 406-410

Nagao, T., Mitamura, T., Wang, X.H., Negoro, S., Yomo, T., Urabe, I., Okada, H. (1992) Cloning, nucleotide sequences, and enzymatic proterties of glucose dehydrogenase isozymes from Bacillus megateriurn IAM1030. J. Bacteriol. 174, 5013-5020

Pearson, W.R., Lipman, D.J. (1988) Improved tools for biologi- cal sequence comparison. Proc. Natl. Acad. Sci. USA 85, 2444-2448

Quatrano, R.S. (1986) Regulation of gene expression by abscisic acid during angiosperm embryo development. Oxford Surv. Plant Cell Mol. Biol. 3, 457-477

Schneider, K., Wells, B., Schmelzer, E., Salarnini, F., Bartels, D. (1993) Desiccation leads to the rapid accumulation of both cytosolic and chloroplastic proteins in the resurrection plant Craterostigrna plantagineum Hochst. Plant 189, 120-131

Skriver, K., Mundy J. (1990) Gene expression in response to abscis- ic acid and osmotic stress. Plant Cell 2, 503-512

Smith, D.B., Johnson, K.S. (1988) Single step purification ofpoly- peptides expressed in Escherichia coli as fusions with glutatione- S-transferase. Gene 67, 31-40

Smith, L.D., Budgen, N., Bungard, S.J., Danson, M.J., Hough, D.W. (1989) Purification and characterization of glucose dehy- drogenase from the thermoacidophilic Archaebacterium Thermoplasrna acidophilurn. Biochern. J. 261, 973-977

Turner, J.F., Turner, D.H. (1980) The regulation of glycolysis and the pentose phosphate pathway. In: The biochemistry of plants, vol. 2, pp. 279-316, Davis, D.D., ed. Academic Press, New York