Plant Physiol. (1999) 119: 859-872
A Single Limit Dextrinase Gene Is Expressed Both in the
Developing Endosperm and in Germinated Grains of Barley1
Rachel A. Burton2,
Xiao-Qi Zhang2,
Maria Hrmova, and
Geoffrey B. Fincher*
Department of Plant Science, University of Adelaide, Waite Campus,
Glen Osmond, South Australia 5064, Australia
 |
ABSTRACT |
The
single gene encoding limit dextrinase (pullulan 6-glucanohydrolase; EC
3.2.1.41) in barley (Hordeum vulgare) has 26 introns that range in size from 93 to 822 base pairs. The mature polypeptide encoded by the gene has 884 amino acid residues and a calculated molecular mass of 97,417 D. Limit dextrinase mRNA is abundant in
gibberellic acid-treated aleurone layers and in germinated grain.
Gibberellic acid response elements were found in the promoter region of
the gene. These observations suggest that the enzyme participates in
starch hydrolysis during endosperm mobilization in germinated grain.
The mRNA encoding the enzyme is present at lower levels in the
developing endosperm of immature grain, a location consistent with a
role for limit dextrinase in starch synthesis. Enzyme activity was also
detected in developing grain. The limit dextrinase has a presequence
typical of transit peptides that target nascent polypeptides to
amyloplasts, but this would not be expected to direct secretion of the
mature enzyme from aleurone cells in germinated grain. It remains to be
discovered how the enzyme is released from the aleurone and whether
another enzyme, possibly of the isoamylase group, might be equally
important for starch hydrolysis in germinated grain.
| |
INTRODUCTION |
Starch is the major carbohydrate reserve in cereal grains, where
it is located in the nonliving cells of the starchy endosperm and
constitutes up to 60% of total grain dry weight (Aman et al., 1985
).
Starch consists of the essentially linear (1
4)-
-glucan amylose,
together with the branched (14,16)--glucan amylopectin. The
two polysaccharides are organized in semicrystalline starch granules,
which in barley (Hordeum vulgare) grain contain 70% to 75%
amylopectin and 25% to 30% amylose (for review, see MacGregor and
Fincher, 1993).
Following germination, the glucosyl residues of amylose and amylopectin
are released to support seedling growth by the concerted action of
-amylases, 
-amylases, debranching enzymes, and -glucosidases. The debranching enzymes catalyze the hydrolysis of
(16)--glucosidic linkages in amylopectin or in
(14,16)--oligoglucosides released by -amylases. Because the
(16)--glucosyl linkages in these oligoglucosides, which are also
referred to as limit dextrins, are not hydrolyzed by - or
-amylases, and because the action of -glucosidase on branched
oligoglucosides is relatively slow, debranching enzymes are considered
to play a central role in the complete depolymerization of starch to
Glc. The debranched oligosaccharides are susceptible to further
hydrolysis by amylases and -glucosidases (Lee et al., 1971).
Starch-debranching enzymes have been divided into two groups based on
differences in their substate specificities and action patterns (Lee et
al., 1971). The first group includes the pullulanases (pullulan
6-glucanohydrolase; EC 3.2.1.41), endohydrolases capable of hydrolyzing
(16)--linkages in pullulan, a polysaccharide consisting of
maltotriosyl residues linked by (16)--linkages. The second group
includes isoamylases (glycogen 6-glucanohydrolase; EC 3.2.1.68),
endohydrolases that hydrolyze (16)--glucosyl linkages in glycogen
and amylopectin but not in pullulan.
The debranching enzyme described here from germinated barley grain
falls into the pullulanase group; however, the natural substrates of
the enzyme are amylopectin or oligomeric limit dextrins rather than
pullulan. For this reason, and because there is a marked preference for
the hydrolysis of (16)--glucosyl linkages in oligosaccharides
compared with polysaccharides, the enzyme is usually referred to as
limit dextrinase (Lee et al., 1971). This nomenclature will be used
here, although the enzyme is also referred to as the R-enzyme (Nakamura
et al., 1996).
As a result of the proposed role for limit dextrinases in starch
hydrolysis in germinated barley grains, and because of the associated
importance of this process in the malting and brewing industries,
considerable effort has been directed toward the purification and
characterization of the barley enzyme (Sissons et al., 1992; Longstaff
and Bryce, 1993; MacGregor et al., 1994a; Kristensen et al., 1998). The
enzyme has been reported in both developing (Sissons et al., 1993) and
germinated grain (Kristensen et al., 1998). Inactive "bound" forms
and active "soluble" forms have been described (Longstaff and
Bryce, 1993), and specific limit dextrinase inhibitors have also been
detected in grain extracts (MacGregor et al., 1994b). These latter
factors complicate the interpretation of limit dextrinase activity
measurements and therefore limited our ability to fully describe the
regulation of limit dextrinase gene expression in barley grain. The
availability of cDNA probes for northern- blot analyses would clearly
be advantageous for the provision of more direct information on the
transcriptional activity of the limit dextrinase gene.
In addition to their role in starch hydrolysis in germinated grain,
debranching enzymes have also been implicated in starch synthesis,
where it is proposed that an appropriate balance between branching and
debranching enzymes is required to achieve the final degree of
branching in amylopectin (Pan and Nelson, 1984; James et al.,
1995; Martin and Smith, 1995; Ball et al., 1996; Nakamura et al., 1997;
Rahman et al., 1998). A debranching enzyme would also be capable of
producing primers for the starch synthase reaction if required (Duffus
and Cochrane, 1993), a function that would explain the presence of
limit dextrinase in developing rice grain (Nakamura et al., 1996).
We have isolated cDNAs encoding a barley limit dextrinase and deduced
the complete primary structure of the enzyme, together with a number of
its physical and chemical properties. cDNA and PCR techniques have been
used to monitor levels of limit dextrinase mRNA transcripts in various
tissues, including developing and germinated grain, and to examine the
influence of the phytohormone GA3 on
transcriptional activity of the gene in isolated aleurone layers.
Enzyme activity was detected in both developing and germinated grain. A
limit dextrinase gene containing 26 introns ranging in size from 93 to
822 bp has been sequenced, and cis-acting elements that may
mediate the GA3 response have been identified in
the promoter region of the gene. Finally, a presequence with structural features similar to those of transit peptides has been detected. Although this can be reconciled with a function for the enzyme in
starch synthesis in amyloplasts in the developing grain, it is more
difficult to envisage how a transit peptide would direct secretion of
the enzyme from the aleurone layer in germinated grain and, therefore,
how limit dextrinase could participate in starch hydrolysis.
| |
MATERIALS AND METHODS |
Plant Material
Embryoless half-grains of barley (Hordeum vulgare L. cv
Haruna Nijo) were surface-sterilized for 30 min with 1% (v/v) NaOCl, washed thoroughly with 10 mM HCl and water, and
soaked in the dark for 3 d in sterile Petri dishes at 25°C. The
starchy endosperm was scraped away and the isolated aleurone layers
were incubated in sterile 10 mM
CaCl2, 2 µM
GA3, 10 µg/mL chloramphenicol, 100 µg/mL
neomycin sulfate, and 100 units/mL nystatin at 25°C in the dark with
continuous shaking for 48 h (Chrispeels and Varner, 1967).
Intact grains were also surface-sterilized and were then germinated in
the dark at 20°C for up to 7 d. Moisture content was maintained
at about 40% (v/v) using the antibiotic mixture described above.
Samples of developing grains at various stages of endosperm development
were harvested from 6 to 36 DPA, when grain fresh weights ranged from
approximately 1 to 72 mg. Because of differences in flowering times
within and between individual heads, developing grains were pooled for
RNA isolation according to their weight. Harvested grains were
immediately frozen in liquid N2 and stored at
80°C until RNA extraction.
Amino Acid Sequence Analysis
Barley limit dextrinase preparations were generously provided by
Dr. Michael Symons (Department of Primary Industries, Toowoomba, Queensland, Australia) and Dr. A.W. MacGregor (Canadian Grain Commission, Grain Research Laboratory, Winnipeg, Manitoba, Canada). Endoproteinase Asp-N fragments were generated essentially as described by Hagmann et al. (1995) and purified by microbore reversed-phase HPLC
on a C18 column (Chen et al., 1993). Selected
peptides and enzyme fractions were sequenced in a protein sequencer
(model G1005A, Hewlett-Packard) using the version 3.0 sequencing
routine, which is based on Edman degradation chemistry.
Detection of Limit Dextrinase Activity
Mature and developing grains and isolated aleurone layers were
homogenized with a Teflon pestle in Eppendorf tubes in 100 mM sodium acetate buffer, pH 5.5, containing 25 mM DTT, 50 mM CaCl2, and
0.05% (w/v) BSA. To ensure that limit dextrinase inhibitors were
released from the enzyme so that total activity could be more
accurately measured, homogenates were held for 5 h at 40°C (Schroeder and MacGregor, 1998). Homogenates were subsequently centrifuged for 30 min at 11,000g and supernatants were
precipitated with 85% saturated
(NH4)2SO4
at 4°C for 2 h. Precipitates were redissolved in extraction
buffer and desalted by dialysis before activity assays.
Limit dextrinase activity was measured with Red Pullulan (Megazyme
International, Bray, County Wicklow, Ireland) or DextriZyme tablets
(Deltagen, Boronia, Victoria, Australia) according to the method
of McCleary (1992) as modified by MacGregor et al. (1994a). Activity
remained linear with time for at least 2 h and is expressed as
milliunits per milligram of protein per hour. Protein was measured with
Coomassie brilliant blue (Bradford, 1976) using BSA as a standard.
Activity was also observed on IEF gels.
(NH4)2SO4-precipitated
proteins were separated on a flatbed IEF apparatus (Pharmacia Biotech)
in 1-mm polyacrylamide gels using a pH gradient of 3.5 to 9.5 (Hrmova
and Fincher, 1993). Prefocused gels were run at 600 V for 40 min,
followed by 800 V for another 10 min. Proteins were detected with
Coomassie brilliant blue after gels were fixed in 20% (w/v) TCA.
Apparent pI values were estimated by reference to marker proteins with
pI values in the range of 4.45 to 9.6 (Bio-Rad). Activity was detected
by immediately overlaying the IEF gels with a 1-mm, 2% (w/v) agar gel
containing 0.5% (w/v) Red Pullulan, and heated to 37°C. After the
active enzyme caused gel clearing, the agar overlay was fixed in 85%
(v/v) ethanol for 16 h.
RNA Extraction and Northern-Blot Analysis
Total RNA was extracted from tissue ground to a fine powder under
liquid N2 using sodium glycinate buffer (Chandler
and Jacobsen, 1991) for isolated aleurone layers or a commercially
available phenol/guanidine isothiocyanate procedure (Trizol, GIBCO-BRL) for developing grain and other tissues, including leaves, roots, and
coleoptiles from 5- and 10-d-old seedlings and scutella from 1-d-germinated grain.
For northern-blot analyses RNA samples (15 µg) were separated in 1%
(w/v) agarose gels containing 2.2 M formaldehyde and
transferred to nylon membranes (Hybond N+,
Amersham). Membranes were baked for 1 h at 80°C, RNA was fixed under short-wavelength UV light for 7 min, and the membranes were probed with 32P-labeled limit dextrinase cDNA
under conditions described previously (Banik et al., 1997). To ensure
approximately equal loading of RNA in individual lanes, gels were
stained with ethidium bromide for comparison of the intensities of rRNA
bands.
RT-PCR Amplification of mRNAs
To detect low levels of mRNA encoding limit dextrinase in the
starchy endosperm of developing grain and in young vegetative tissues,
cDNA was prepared from 3 µg of total RNA following the procedure of
Frohman et al. (1988) using RT (Superscript II, GIBCO-BRL) and a RACE
primer (TRACE, 5
-GACTCGAGTCGACATCGAT17-3; Frohman et al.,
1988) at 50°C. The final reaction volume was 50 µL. Aliquots of 2 µL were subsequently used in PCR reactions with the 5 and 3
limit dextrinase primers (5-GTGCATTTGCATATCAGG-3 and
5-TAAGGCTTTGAAGAGCAGA-3, respectively). The PCR program was 35 cycles
of 94°C for 40 s, 50°C for 40 s, and 72°C for 60 s. The single, 641-bp product was isolated from gels using Geneclean
(Bresatec, Adelaide, Australia), digested with diagnostic restriction
enzymes to check fragmentation patterns, and, following transfer to
Hybond N+ membranes, probed with a fragment of
the limit dextrinase cDNA at high-stringency conditions (0.1× SSC and
0.1% [w/v] SDS at 65°C). Selected PCR products were also sequenced
to confirm that they encoded the barley limit dextrinase.
To ensure that approximately equal amounts of RNA were used for
amplification, control PCR reactions in which mRNA encoding the
constitutive glycolytic pathway enzyme GAPDH from barley (accession no.
U36650) were also performed. The 5 and 3 GAPDH primers were
5-CCACCGGTGTCTTCACTGACAAGG-3 and 5-GCCTTAGCATCAAAGATGCTGG-3, respectively, and the 550-bp product was observed after either 28 or 35 cycles (94°C for 40 s, 65°C for 40 s, and 72°C for
60 s).
Isolation of cDNAs
A cDNA library prepared from poly(A+) RNA
from 12-d-old barley seedlings was purchased from Clontech Laboratories
(Palo Alto, CA) and screened with a cDNA encoding the rice endosperm
R-enzyme (clone D50602; Nakamura et al., 1996). A single positive clone containing a cDNA insert of 654 bp was isolated from 9 × 105 plaque-forming units using procedures
described by Banik et al. (1996). The cDNA was subsequently used to
screen a cDNA library from GA3-treated barley
aleurone layers, which was generously provided by Dr. Mitali Banik. The
aleurone library was prepared in 
ZAP (Uni-ZAP XR, Stratagene). In
both cases, libraries were plated out at high density on
Escherichia coli lawns. Duplicate plaque lifts on Hybond-C
nitrocellulose membranes (Amersham) were prehybridized, hybridized, and
washed under previously described conditions (Banik et al., 1996).
Following plaque purification of positive clones, cDNA inserts were
rescued into the phagemid pBluescript SK(+), and both strands were
sequenced using the dideoxynucleotide chain-termination procedure
(Sanger et al., 1977). Computer analyses of nucleotide sequences were
performed with Genetics Computer Group (Madison, WI) software (Devereux
et al., 1984) in the ANGIS suite of programs at the University of
Sydney.
Although the screening procedure yielded a cDNA of 2.6 kb from the
barley aleurone cDNA library, this was not a full-length cDNA. To
obtain longer cDNAs, 1 µL of the aleurone cDNA library consisting of
approximately 3 × 106 plaque-forming units
was used for a mass excision of cDNA inserts, and the resulting
phagemids were subjected to PCR amplification using the T3 primer as
the 5 PCR primer and an oligonucleotide corresponding to a 5
sequence of the 2.6-kb cDNA (5-TGGATGATACACGTCGAC-3) as the 3
PCR primer. This yielded a 316-bp PCR product, but its sequence still
did not extend to the 5 end of the mRNA, and it was concluded that the
aleurone cDNA library contained no full-length clones. Finally, 5-RACE
PCR (Frohman et al., 1988) from a single-stranded cDNA population
generated from total RNA extracted from
GA3-treated aleurone layers using a 3
oligonucleotide primer (5-CAAGGCTGGCGACGTCGACAG-3) based on the
sequence of the 316-bp PCR product resulted in the isolation of a
507-bp fragment that enabled the sequence of the 5 end of the mRNA to
be determined. The PCR products were purified from agarose gels using
Bresaclean (Bresatec), and ligated into pBluescript SK(+) for
nucleotide sequence analysis.
Isolation of Genomic Clones
The barley genomic library was prepared by Mr. Ron Osmond
(Department of Plant Science, University of Adelaide) from partially digested genomic DNA extracted from 7-d-old barley (cv Galleon) seedlings. The DNA fragments were ligated into the EcoRI
site of the bacteriophage DASH II vector (Stratagene). The library was plated out on lawns of E. coli XL1-Blue (P2) cells and
screened by hybridization of membrane filter plaque replicas using the 654-bp cDNA fragment as a probe (Sambrook et al., 1989). Phage DNA was
isolated from positive clones, digested with restriction endonucleases,
and subjected to Southern-blot analysis to identify limit dextrinase
gene fragments. The fragments were subcloned into the plasmid
pBluescript SK(+) for sequencing. Deletion series were generated with
the Erase-a-Base system (Promega).
Primer Extension
To determine the transcription start point of the gene, total RNA
preparations (12-24 µg) from GA3-treated
aleurone layers were annealed with 25 ng of end-labeled
oligonucleotides (complementary to sequences in the 5 region of the
cDNA) in deionized 50% (v/v) formamide at 34°C for 16 h after
heating the incubation mixture for 10 min at 85°C (Sambrook et al.,
1989; Slakeski et al., 1990). The RNA-oligonucleotide complex was
recovered by ethanol precipitation, and reverse transcription was
performed as described by Sambrook et al. (1989). The primer-extension
product was examined on a DNA-sequencing gel on which sequence
reactions of the gene itself, primed with the same 5 oligonucleotides,
were also subjected to electrophoresis.
| |
RESULTS |
Isolation and Characterization of cDNAs
Exhaustive screening of the libraries yielded a cDNA of
approximately 2.6 kb, but this was not the full-length cDNA. Subsequent 5 primer extension from cDNA preparations and 5 anchored PCR from
aleurone layer RNA finally produced DNA fragments containing the 5
region of the cDNA. The nucleotide sequences of all overlapping fragments were identical. A partial restriction map of the cDNA sequence and the relationships of the cDNA clones that were used to
determine the full nucleotide sequence are shown in Figure 1.
View larger version (18K):
[in this window]
[in a new window]
| Figure 1.
a, Partial restriction map of the 3429-bp cDNA
encoding barley limit dextrinase. b, Overlapping cDNAs and PCR products
used to obtain the complete nucleotide sequence of the cDNA for barley
limit dextrinase. The 2.6-kb 3 cDNA was screened from a library of
GA3-treated aleurone layers. The 316-bp "central"
sequence was obtained as a PCR product amplified from cDNA inserts
excised from the same aleurone cDNA library. The 507-bp 5 fragment was
prepared by 5 RACE PCR from aleurone total RNA. The arrows at the 3
end indicate multiple polyadenylation sites.
|
|
The complete nucleotide sequence of the nearly full-length, 3429-bp
cDNA and the deduced amino acid sequence are presented in Figure
2. Confirmation that the cDNA encoded a
barley limit dextrinase was provided by the exact match of the sequence
of the 20 NH2-terminal amino acids determined
directly from purified enzyme (Fig. 2). In addition, deduced amino acid
sequences corresponding to four endoproteinase Asp-N fragments
generated from the purified enzyme were found (data not shown).
View larger version (56K):
[in this window]
[in a new window]
View larger version (62K):
[in this window]
[in a new window]
| Figure 2.
Nucleotide sequence of the overlapping cDNAs for
barley limit dextrinase and the deduced amino acid sequence for the
enzyme. The NH2-terminal Ala of the mature polypeptide is
indicated by the vertical arrow, and the amino acid sequence obtained
directly from the purified enzyme is underlined and in bold.
Potential
N-glycosylation sites are underlined and amino
acid residues that are likely to be involved in catalysis are in bold.
The TGA stop codon is also shown in bold and the four large arrows near
the 3 end show the various positions of polyadenylic acid
tails in different cDNAs isolated. The putative polyadenylation signal
AATAAA is underlined. The cDNA accession number is AF122049.
|
|
The cDNA has an open reading frame extending from nucleotides 8 to
2893, and the sequence corresponding to the
NH2-terminal amino acid sequence of the enzyme
itself begins at nucleotide 242 (Fig. 2). The nucleotide sequence that
encodes a putative transit peptide has five in-frame ATG-Met codons
beginning at nucleotides 8, 14, 20, 29, and 179. The consensus sequence
for translation start codons in plant genes is AACAATGGC,
in which the most crucial requirements are believed to be a purine at
position 3 and a G at position 4; the A of the ATG codon is
designated as position +1 (Joshi, 1987; Lutcke et al., 1987; Cavener
and Ray, 1991). None of the ATG codons in the limit dextrinase cDNA has
both a purine at 3 and a G at position +4, but sequence context around the first ATG codon from the transcription start point beginning
at nucleotide 8 reasonably matches the consensus sequence and is
therefore considered to be the most likely translation start point
(Fig. 2).
The coding region for the mature enzyme has a relatively balanced
pattern of codon usage. Of the 884 codons in this region, 48% have C
or G in the wobble-base position, and all codons are used. The coding
region for the mature enzyme is followed by a TGA stop codon, which
forms part of a 3 untranslated region that ranges from approximately
260 to 530 bp in length. Polyadenylic acid tails were detected at four
different sites in the cDNA clones studied (Fig. 2). A potential
polyadenylation signal, AATAAA, begins 88 nucleotides downstream from
the stop codon.
Isolation and Analysis of the Limit Dextrinase Gene
From approximately 3 × 106 plaques
screened, nine positive genomic DNA clones were isolated when a 654-bp
limit dextrinase cDNA was used as a probe. Restriction digestion and
Southern-blot analyses showed that the gene was very large and was
dispersed over two cloned genomic DNA fragments. Overlap of the clones
themselves was performed by PCR from genomic DNA preparations, using
primers corresponding to a sequence in the exons immediately 5 and 3 to the ends of the original genomic clones (nucleotides 1912-1936 and
1996-1971 in Fig. 2). A fragment of approximately 900 bp was amplified
from DNA from four individual varieties and sequenced (data not shown).
Six fragments from the two genomic clones were subcloned into
pBluescript SK(+) for sequencing. Subsequent nucleotide sequence
analyses confirmed that overlapping fragments were identical and that
exon sequences were always identical to those determined from the cDNA.
It was concluded therefore that the fragments represent part of a
single limit dextrinase gene. The sequence of an 11.5-kb DNA fragment
carrying the gene isolated in the present study can be found in the
database under accession no. AF122050, and exhibits 98%
positional identity with the nucleotide sequence of a barley limit
dextrinase gene recently added to the database (accession no.
AF022725).
The barley limit dextrinase gene consists of 27 exons separated by 26 introns. Exon/intron boundaries were identified by reference to the
cDNA sequence shown in Figure 2. In all cases the 5 and 3 boundary
sequences of the introns were closely related to the consensus
sequences for monocotyledonous plants of AG
GTAAGT for the 5
boundary and TGCAGGT for the 3 boundary (Simpson and Filipowicz,
1996).
The structure of the barley limit dextrinase gene is represented
diagrammatically in Figure 3. The exact
positions of intron insertions, together with their lengths, are shown
in Table I. The introns range in length
from 93 to 822 bp: 3 are less than 100 bp, 11 are between 100 and 200 bp, 6 are between 200 to 299 bp, 4 are between 300 and 500 bp, and 2 are more than 500 bp (Table I).
View larger version (6K):
[in this window]
[in a new window]
| Figure 3.
Structure of an 11.5-kb genomic DNA fragment
carrying the barley limit dextrinase gene, showing the 27 exons
(black), the 26 introns (white), and the promoter region (shaded).
Positions of the GA3 response elements (GARE), the putative
TATA box of the promoter, the NH2 terminus of the mature
protein, and the stop codon are indicated.
|
|
Promoter Region of the Gene
The nucleotide sequence of the promoter region of the barley limit
dextrinase gene is shown in Figure 4a. To
define the transcription start point of the gene,
oligonucleotide-primed synthesis of cDNA from aleurone RNA preparations
was performed. Termination products were detected in three main
positions (Fig. 4b). The position that most closely matched the
transcription start point consensus sequence CTCATCA for
plant genes (Fig. 4a; compare Joshi, 1987) was the AGAATCT
sequence farthest from the translation start codon (Fig. 2). We have
therefore numbered the gene sequence from that point.
View larger version (55K):
[in this window]
[in a new window]
| Figure 4.
Promoter region of the barley limit dextrinase
gene. a, Nucleotide sequence of the gene upstream from the translation
start ATG codon. The putative transcription start point is indicated by
the arrowhead and is designated nucleotide +1. Underlined sequences
starting from the 5 end of the sequence include the CCTTT pyrimidine
box, the TAAAACAAA box and the TATCCAA box of the putative
GA3 response element, and the CATT box and the proposed
TATA box of the gene. b, Primer extension from total RNA of
GA3-treated aleurone layers using a primer specific for the
5 region of the cDNA. The top two lanes show the major primer
extension (PE) termination products (arrows), whereas the lower four
lanes represent the nucleotide sequence of the gene, primed with the
same oligonucleotide used for the primer extension.
|
|
A putative TATA box begins 142 bp upstream from the transcription start
point, but it only approximately matches the consensus sequence
ACTATATATAG for plant genes and is located farther upstream than most
TATA boxes (Fig. 4a; compare Joshi, 1987). A CATT sequence begins
approximately 20 bp 5 to the possible TATA box. In view of the
relatively poor correspondence of these promoter sequences with plant
consensus sequences, and because of the possibility that the 5 region
of the gene might be interrupted by introns, approximately 4 kb of
sequence upstream from that shown in Figure 4a was determined. However,
no additional transcription start point or potential TATA box sequences
were detected (data not shown). Furthermore, the promoter of the single
limit dextrinase gene from rice has similarly ill-defined promoter
sequences (Y. Nakamura, personal communication).
Northern-Blot Analysis
When RNA preparations from developing barley grain and
from GA3-treated aleurone layers were subjected
to northern-blot analysis, a 2-kb cDNA probe hybridized strongly with a
single RNA species of approximately 3.4 kb in
GA3-treated aleurone layers (Fig.
5a). No limit dextrinase gene transcripts
were detected in the developing grain (Fig. 5a), leaves, roots,
coleoptiles, or scutella (data not shown), and transcript levels in
untreated aleurone layers were low (Fig. 5a).
View larger version (48K):
[in this window]
[in a new window]
| Figure 5.
Northern blots of RNA preparations from developing
barley grain and GA3-treated aleurone layers using limit
dextrinase cDNA as a probe. a, Lanes 1 to 4, RNA from grain during the
period 19 to 40 DPA (grain weight 27-70 mg); lanes 5 and 6, RNA from
aleurone layers treated for 48 h with (+GA) or without (GA) 2 µM GA3. b, Time course of GA3
induction of limit dextrinase mRNA transcripts in isolated barley
aleurone layers.
|
|
In GA3-treated aleurone layers, limit dextrinase
gene transcripts could be detected after 12 h, were most abundant
at 24 h, and decreased thereafter (Fig. 5b). In view of the
apparent induction of limit dextrinase gene transcripts by
GA3 in the isolated barley aleurone layers, the
promoter sequence of the gene (Fig. 4a) was scanned for the presence of
cis-acting elements of the GA-response complex that are
commonly detected in gene promoters induced by the hormone in cereal
aleurone layers (Gubler et al., 1995). These elements include a
pyrimidine box, a TAACAAA box, and a TATCCAC box (Gubler and Jacobsen,
1992). The barley limit dextrinase gene promoter has similar sequences,
which include a CCCTTTCTCCC pyrimidine box beginning at nucleotide
594, a TAAAACAAA sequence at nucleotide 523, and a TATCCAA sequence
at nucleotide 379 (Fig. 4a).
RT-PCR Analysis of mRNAs in Developing Grain
Although no limit dextrinase mRNA transcripts were detected by
northern-blot analysis at a stage of grain development when starch
synthesis was likely to be in progress (Fig. 5a; compare MacLeod and
Duffus, 1988), Nakamura et al. (1996) isolated a rice limit dextrinase
cDNA from a library generated from RNA of developing grain, and it has
been widely suggested that the enzyme could be involved in starch
synthesis during grain development (Nakamura et al., 1997). RT-PCR was
therefore used in further attempts to detect limit dextrinase mRNA in
developing grain, and, to ensure that transcripts were not overlooked,
RNA preparations isolated over a broader time scale in starchy
endosperm development (6-36 DPA) were also examined. The resultant
amplification products are shown in Figure
6a, together with a Southern blot of the
PCR products (Fig. 6b). Nucleotide sequence analysis of the PCR
products confirmed that they corresponded to the limit dextrinase
sequence (data not shown). It is therefore clear that mRNA encoding
limit dextrinase is present at relatively low levels in developing
endosperm until the grain achieves a mass of 25 to 27 mg (Fig. 6a; this corresponds to approximately 19 DPA). Thereafter, the levels of limit
dextrinase mRNA transcripts decrease. The abundance of the control mRNA
for GAPDH also decreases after this time, and this was especially
apparent when PCR amplification of the GAPDH cDNA was reduced from 35 to 28 cycles (Fig. 6, c and d).
View larger version (31K):
[in this window]
[in a new window]
| Figure 6.
RT-PCR analysis of limit dextrinase mRNA
transcripts in the developing barley grain 6 to 36 DPA. Lane M, DNA
molecular mass markers; lane C, control PCR reactions; lanes 1 to 65, weight of the grain (in milligrams) at the time of RNA extraction.
Below are shown the corresponding times of RNA extraction (6-36 DPA).
a, RT-PCR products amplified with limit dextrinase (LD) primers; b,
Southern blot of the RT-PCR products shown after probing with a limit
dextrinase cDNA; c, RT-PCR products amplified with primers for the
constitutively expressed GAPDH after 35 cycles; d, RT-PCR products of
GAPDH primers after 28 cycles; e, limit dextrinase (LD) primers and 35 PCR cycles of RNA preparations extracted from the pericarp (P) and
endosperm (E) of developing grain, together with products amplified in
28 cycles using the GAPDH primers.
|
|
To ensure that the limit dextrinase mRNA did not originate from the
pericarp or other maternal tissues of the grain, the endosperm was
squeezed out from the developing grain before to RNA isolation, and
RT-PCR analysis was repeated on RNA extracted from both the endosperm
and the remaining pericarp and other tissues. The majority of mRNA
encoding limit dextrinase was shown to be in the endosperm, as noted in
previous immunological studies (Sissons et al., 1993), although some
could be detected in the pericarp (Fig. 6e). It should be noted that a
thin layer of aleurone tissue could be detected surrounding the starchy
endosperm that was squeezed out from the grain. However, the aleurone
layer could not be easily dissected away from the starchy endosperm,
and the results reported here therefore relate to the endosperm as a
whole. The RT-PCR procedure also allowed the detection of limit
dextrinase mRNA in young leaves, young roots, and the scutellum of
1-d-germinated grain; apparent levels in these tissues were relatively
low (data not shown).
Enzyme Activity
In situ detection of activity in IEF gels confirmed that active
enzyme was present during grain development, albeit at low levels (Fig.
7A). Similarly, active enzyme could be
detected in extracts of germinated grain (Fig. 7B). The faint bands
that are seen in the pI-6.5 to -7.0 region of the Red Pullulan gel in
Figure 7A and more faintly in Figure 7B are believed to be artifacts of
the gel system rather than another limit dextrinase isoform, because
the band was also present in blank lanes where no extract was loaded
(e.g. in the far left lane of the Red Pullulan panel in Fig. 7B).
Quantitative measurements of limit dextrinase activity in developing
endosperm and germinated grain are compared in Table II.
View larger version (72K):
[in this window]
[in a new window]
| Figure 7.
IEF of extracts of a barley developing grain (A)
and a germinated grain (B). The left panels show proteins stained with
Coomassie brilliant blue and the right panels show limit dextrinase
activity revealed by clearing of a Red Pullulan overlay gel. Lanes S,
pI protein standards; lanes M, mature grain and other extracts from
developing endosperm 8, 10, and 16 DPA. The number of days after the
initiation of germination are indicated in B.
|
|
View this table:
[in this window]
[in a new window]
|
Table II.
Limit dextrinase activity in tissue extracts from
developing endosperm and germinated grain
Assays were performed according to the method of McCleary (1992) and
MacGregor et al. (1994a).
|
|
Properties of the Encoded Enzyme
The barley limit dextrinase appears to have a relatively long
presequence before the NH2-terminal Ala residue
of the mature enzyme (Fig. 2). This sequence is 78 amino acid residues
in length and contains 13 Ala residues (17% of the presequence on a
molar basis), 12 Pro residues (15%), 12 Arg residues (15%), and 9 Ser residues (12%). Thus, these four residues account for nearly 60% of
the amino acids in the putative transit peptide (Fig. 2), and the Ala,
Pro, and Arg residues are two to three times more abundant in the
transit peptide than in the mature polypeptide (data not shown).
Furthermore, the major amino acids are asymmetrically distributed along
the transit peptide. The Arg and Pro residues are concentrated toward
the NH2-terminal end, and clusters of two or
three contiguous residues are apparent. Positively charged amino acid
residues, sometimes in clusters, are found at relatively regular
intervals along the NH2-terminal part of the
transit peptide (Fig. 2). Similarly, clusters of Ser residues are
observed in the NH2-terminal region. An internal
Met residue is present and, toward the COOH-terminus of the peptide,
negatively charged Glu residues are also present. Although the net
charge of the transit peptide at neutral pH is +10, there is a clear
delineation between positive charges at the NH2
terminus and negative charges at the COOH terminus. No extended
hydrophobic regions are seen.
The region of the cDNA from the codon specifying the
NH2-terminal Ala residue at nucleotide 242 to the
TGA stop codon at nucleotide 2894 encodes a polypeptide of 884 amino
acid residues. Seven potential N-glycosylation sites begin
at amino acid residues 229, 440, 524, 529, 642, 816, and 851 in the
deduced sequence (Fig. 2).
When the deduced amino acid sequence for the barley limit dextrinase
was compared with sequences in the DNA and protein databases, the best
matches were with the rice R-enzyme (81% positional identity; Nakamura
et al., 1996), a pullulanase from spinach (63% positional identity;
accession no. X83969; A. Renz, R. Schmid, J. Kossmann, and E. Beck,
unpublished data), a pullulanase from the bacterium Klebsiella
aerogenes (39% identity; Katsuragi et al., 1987), the sugary1 gene of maize (24% identity; James et al., 1995),
and an isoamylase from Pseudomonas amyloderamosa (25%
identity; Amemura et al., 1988). The alignment of these sequences,
shown in Figure 8, is restricted to the
catalytic site region, but blocks of approximately 2 to 10 identical
amino acids are also distributed along the rest of the polypeptide
chain. The blocks of highly conserved amino acid sequences noted in
starch-debranching enzymes of diverse origin by Nakamura et al. (1997)
are also conserved in the barley enzyme (data not shown).
View larger version (78K):
[in this window]
[in a new window]
| Figure 8.
Amino acid sequence alignments of the catalytic
domains of debranching enzymes from barley (Fig. 2), rice (Nakamura et
al., 1996), spinach (A. Renz, R. Schmid, J. Kossmann, and E. Beck,
unpublished data), and K. aerogenes (Katsuragi et al.,
1987), using the PileUp and PrettyBox programs (Devereux et al., 1984).
Shaded and hatched boxes indicate identical and homologous residues,
respectively.
|
|
| |
DISCUSSION |
The nucleotide sequence of the mRNA encoding barley limit
dextrinase has been obtained from overlapping cDNA clones and from PCR
amplification products. The clones were screened from a barley aleurone
library, which was generated from poly(A+) RNA
isolated from aleurone layers treated for 48 h with 2 µM GA3 (Banik et al., 1996) and
from a commercially available cDNA library from barley seedlings. The
largest cDNA isolated was 2.6 kb in length, and, despite extensive
screening of available cDNA libraries, full-length cDNAs could not be
found. This is probably attributable to the large size of the mRNA and
to the presence of secondary structures caused by GC-rich sequences in
the 5 region of the mRNA (Fig. 2). Complementary DNAs encoding the 5 region were eventually obtained from RNA preparations by PCR
procedures. Four independent polyadenylation sites were detected at the
3 end of the cDNAs (Fig. 2).
The cDNA sequence has an open reading frame that encodes a putative
transit peptide of 78 amino acid residues and a mature polypeptide of
884 amino acid residues. The calculated molcular mass of the mature
polypeptide is 97,417 D and the calculated pI is 5.0. These values are
comparable to a molecular mass of about 105 kD and a pI of 4.2 to 4.6 reported for purified barley limit dextrinase enzyme preparations
(Sissons et al., 1992; Kristensen et al., 1998; MacGregor et
al., 1994a). The apparent discrepancy between the deduced
Mr value and the value obtained for the
purified enzyme might result from glycosylation of the native enzyme;
seven potential N-glycosylation sites are found in the
deduced amino acid sequence (Fig. 2).
Alignment of the barley limit dextrinase amino acid sequence with other
sequences in the protein and DNA databases reveals significant
similarities with enzymes in the pullulanase and isoamylase groups of
debranching enzymes from higher plants and from microbial sources (Fig.
8). When the amino acid sequences of similar debranching enzymes were
examined for phylogenetic relatedness using the PileUp program
(Devereux et al., 1984), two distinct groups could be distinguished
(Fig. 9). Debranching enzymes of the
pullulanase type, including representatives from higher plants and
bacteria, were grouped together, whereas the isoamylase type fell into
a second group (Fig. 9).
View larger version (13K):
[in this window]
[in a new window]
| Figure 9.
Unrooted, radial phylogenetic tree of selected
starch-debranching enzymes as determined using the PileUp, EprotDist,
and EFitch programs of the University of Wisconsin package (version 8;
Devereux et al., 1984). The two major branches distinguish the
pullulanase type (upper branch) from the isoamylase type (lower
branch). The accession numbers of the sequences are shown.
|
|
Glycosyl hydrolases have been classified into distinct families based
on amino acid sequence similarities and hydrophobic cluster analyses
(Henrissat and Bairoch, 1993), which suggest that the barley limit
dextrinase is a member of the family 13 group of glycosyl hydrolases.
On this basis, the catalytic nucleophile is likely to be Asp-472 and
the catalytic acid Glu-509. In common with other family 13 enzymes, the
barley limit dextrinase would be expected to retain anomeric
configuration during hydrolysis of substrates (Jesperson et al., 1991).
A gene encoding the barley limit dextrinase was also isolated,
and the sequence of an 11.5-kb genomic DNA fragment has been determined
(Fig. 3). In an independent study, the cDNAs isolated here have been
used to show that there is only one limit dextrinase gene in barley and
that the gene is located on the long arm of chromosome 4H, where it has
now been mapped (C.-D. Li, X.-Q. Zhang, R.C.M. Lance, L.C. MacLeod,
G.B. Fincher, and P. Langridge, unpublished data). The most
striking structural feature of the barley limit dextrinase gene is its
highly fragmented nature: 26 introns are found in the coding region of
the nascent polypeptide (Fig. 3), ranging in size from fewer than 100 nucleotides to more than 800 bp (Table I). Whether the highly
fragmented nature of the barley limit dextrinase gene has evolutionary
or functional significance is not known, but the transcriptional unit
is certainly more complex than others that are expressed in aleurone
layers of germinated barley (Fincher, 1989).
A cDNA encoding the barley limit dextrinase was used in northern-blot
analyses to examine transcriptional patterns of the gene in various
tissues. Limit dextrinase mRNA transcripts could not be detected by
northern-blot analysis in developing grain (Fig. 5a), nor could they be
detected in coleoptiles, young leaves, young roots, or scutellum of
1-d-germinated grain (data not shown). Limit dextrinase mRNA was
observed only in northern blots of RNA preparations from isolated
aleurone layers (Fig. 5a). The abundance of limit dextrinase mRNA was
greatly enhanced by treatment of the aleurone layers with 2 µM GA3. The mRNA levels increased
in the first 24 h after treatment and thereafter decreased (Fig. 5b). The addition of ABA abolished the GA3
induction of limit dextrinase mRNA (data not shown). The induction
patterns of limit dextrinase gene transcription by
GA3 in isolated barley aleurone layers are
similar to those observed for (13,14)--glucanases (Mundy and
Fincher, 1986), (14)--xylan endohydrolases (Banik et al., 1996),
and -amylases (Chandler et al., 1984). The GA3 induction of the limit dextrinase gene is consistent with the presence
of nucleotide sequence elements in the promoter region of the gene that
have previously been implicated in the GA3
induction of gene expression in barley aleurone layers (Fig. 5; Gubler
and Jacobsen, 1992; Banik et al., 1997).
The expression patterns described above must be reconciled with the
likely functions of limit dextrinase in barley, each of which involve
the hydrolysis of (16)--glucosyl linkages in amylopectin or
derived oligosaccharides. It is widely assumed that in germinated grain
the enzyme is required for the complete depolymerization of stored
starch to Glc, and the induction of limit dextrinase mRNA transcripts
by GA3 in isolated aleurone layers (Fig. 5a) is
consistent with this function. A second role that has been suggested
for limit dextrinase is in starch synthesis in the developing grain.
This could involve the processing of preamylopectin (Martin and Smith,
1995; Mouille et al., 1996; Nakamura et al., 1996; Rahman et
al., 1998) or the provision of oligomeric primers for starch synthases
(Duffus and Cochrane, 1993). Any role in starch synthesis would
presumably be performed in amyloplasts of the developing grain, but
could also occur in chloroplasts of leaves and other photosynthetic
tissues. A third possible function of limit dextrinase is in the
turnover of starch that accumulates transiently in chloroplasts and in
amyloplasts in the parenchyma cells of the scutellum in germinated
grain (Smart and O'Brien, 1973).
The apparent absence of significant levels of limit dextrinase mRNA
in developing grains in the scutellum of germinated grain and in young
vegetative tissues, as shown by northern-blot analyses, was therefore
unexpected, particularly because the homologous limit dextrinase
(R-enzyme) from rice was purified from the starchy endosperm of
developing grain and the cDNA encoding the rice enzyme was generated
from RNA prepared from the same tissue (Nakamura et al., 1996). When
more sensitive RT-PCR methods were used, low levels of limit dextrinase
mRNA were detected in the developing endosperm until about 20 DPA, when
levels rapidly declined (Fig. 6a). These expression patterns correspond
closely to the timing of starch synthesis in developing barley grains,
although the latter is dependent to some extent on growth temperatures
(MacLeod and Duffus, 1988). Given the relatively low levels of limit
dextrinase mRNA in developing endosperm (Fig. 5), grain extracts were
subsequently assayed for the enzyme itself. Activity in developing
grain was detected at levels similar to those found initially in the
germinated grain, and the results confirm earlier immunological studies
suggesting that limit dextrinase protein is present in developing
barley kernels (Sissons et al., 1993). The presence of low levels of limit dextrinase mRNA and enzyme activity in developing endosperm at
times that coincide with the deposition of starch is consistent with
the proposed role for the enzyme in amylopectin synthesis in cereals
(Rahman et al., 1998), although it must be emphasized that the
endosperm tissue used here did include some associated aleurone cells.
The RT-PCR method also confirmed that limit dextrinase mRNA was present
in young leaves, young roots, and the scutellum of germinated grain
(data not shown), but levels were low and we have not yet begun
developmental studies of transcription in these tissues.
Clearly, the various functions that have been ascribed to limit
dextrinases in starch synthesis and hydrolysis would require targeting
of the enzyme to different subcellular compartments. Enzyme synthesized
in the aleurone layer would be expected to be secreted into the starchy
endosperm and to carry a signal peptide of the type that would target
the nascent polypeptide to the ER for eventual secretion from the cell.
On the other hand, targeting of the enzyme to amyloplasts in developing
endosperm or to chloroplasts in leaves would require a
transit-peptide-type targeting signal. Indeed, the 78-amino acid leader
sequence on the limit dextrinase polypeptide (Fig. 2) is more typical
of transit peptides that target polypeptides to plastids, and bears
little resemblance to normal ER-type signal peptides (Fig. 2).
Generally, transit peptides are longer than signal peptides, do not
have extended hydrophobic regions, are rich in Ser, Thr, and Pro
residues, and are highly charged (von Heijne et al., 1989; Baba et al.,
1993; Edwards et al., 1995). These features are found in the barley limit dextrinase presequence (Fig. 2). The limit dextrinase transit peptide is also notable for the clear separation of positive charges and Pro residues toward the NH2 terminus, and
negatively charged Glu residues toward the COOH terminus of the peptide
(Fig. 2). These characteristics are also found in the rice R-enzyme
presequence (Nakamura et al., 1996).
If a single limit dextrinase enzyme derived from a single gene were to
participate in starch synthesis or turnover in chloroplasts or
amyloplasts and were also secreted into the extracellular space from
aleurone layer cells in germinated grain, one might anticipate that
alternative splicing of pre-mRNA could be used to attach the
appropriate targeting signal to the NH2 terminus
of the nascent polypeptide. The highly fragmented nature of the barley
limit dextrinase gene described here would allow such a mechanism. It is also noteworthy that a potential intron 3 splicing site is located
precisely adjacent to the NH2-terminal Ala
residue of the mature enzyme (Fig. 2), where alternate splicing of
introns encoding different targeting signals could occur. With this
possibility in mind, we sequenced more than 4 kb upstream from the
promoter sequence of the barley limit dextrinase gene shown in Figure 4 in an attempt to find signal-peptide sequences that would indicate targeting to the ER, but found no such sequences. Furthermore, when PCR
and primer-extension techniques were applied to RNA from the
GA3-treated aleurone cells, the presence of the
transit-peptide sequence was always confirmed (data not shown).
Given that aleurone cells contain no starch (MacGregor and Fincher,
1993), are nonphotosynthetic, and that limit dextrinase is thought to
be secreted into the starchy endosperm to catalyze debranching of limit
dextrins, the presence of the transit peptide remains difficult to
explain. Plastids have been reported in isolated barley aleurone layers
after prolonged treatment with GA3 (Jones, 1969),
and these could be the sites for limit dextrinase targeting. But how
the enzyme would be released from the aleurone cells to be secreted
into the starchy endosperm is not known. In a recent and comprehensive
study of the development of limit dextrinase enzyme activity in
de-embryonated barley half-grains and whole, germinated grains,
Schroeder and MacGregor (1998) showed that GA3
caused a significant increase in total enzyme activity over the first
3 d of treatment, but little if any enzyme was secreted from the
aleurone layer. Enzyme activity in the starchy endosperm remained low
throughout and, because their extraction procedures were designed to
release any limit dextrinase-inhibitor complexes, it seemed unlikely
that much activity originated from the developing endosperm. The
results of Schroeder and MacGregor (1998) are entirely consistent with
those reported here.
It might be argued that limit dextrinase is not required in the starchy
endosperm until the final stages of reserve mobilization, when the
enzyme could mediate in the recovery of the last few glucosyl residues
that are bound in the limit dextrins. Perhaps the enzyme might be
released if, after their reserves were depleted, aleurone layer cells
eventually burst. Morphological examinations of senescing aleurone
layers from both wheat and barley show that the cells do indeed die
after prolonged treatment with GA3 (Kuo et al.,
1996; Wang et al., 1996), but this process is characteristic of
programmed cell death rather than necrosis. In programmed cell death,
DNA fragmentation occurs to produce oligonucleosome-sized fragments in
a fashion similar to apoptosis, the cytoplasm and nuclei of dead and
dying cells collapse and shrink, but membrane integrity appears to be
maintained and there is little leakage of cell contents (Pennell and
Lamb, 1997).
It has been proposed that the retention of membrane structure would
offer some protection against pathogen attack, in contrast to necrotic
cell death, in which cell leakage usually occurs and releases nutrients
that could support microbial growth (Pennell and Lamb, 1997). The
programmed cell death that is observed in cereal aleurone (Kuo et al.,
1996; Wang et al., 1996) is therefore unlikely to result in the
wholesale release of accumulated limit dextrinase. In support of
this, Schroeder and MacGregor (1998) showed that only very small
amounts of limit dextrinase were secreted from aleurone cells after
5 d of treatment with GA3, and at this stage
programmed cell death would have been well advanced (Wang et al.,
1996).
In conclusion, the results presented here are consistent with a role
for limit dextrinase in starch synthesis in developing barley grain,
but raise some questions as to the involvement of the enzyme in starch
degradation during the mobilization of the starchy endosperm in
germinated grain. Perhaps a previously undetected, isoamylase-like
enzyme is more important for amylopectin degradation in germinated
barley grain. Isoamylase has been described in maize (James et al.,
1995; Rahman et al., 1998) and, although it could debranch amylopectin
during starch degradation, it would go undetected in the pullulanase
assays that are commonly used to quantitate limit dextrinase
activity.
| |
FOOTNOTES |
1
This work was supported by a grant from the
South East Australian Malting Barley Quality Improvement Program, which
includes contributions from Joe White Maltings, the Grains Research and Development Corporation, the Strategic Research Foundation,
Barrett-Burston International, the Australian Associated Brewers, and
the Adelaide Malting Company.
2
These two authors contributed equally to this
work.
*
Corresponding author; e-mail gfincher{at}waite.adelaide.edu.au;
fax 61-8-8303-7109.
Received September 2, 1998;
accepted December 2, 1998.
| |
ABBREVIATIONS |
Abbreviations:
DPA, days postanthesis.
GAPDH, glyceraldehyde
3-phosphate dehydrogenase.
RACE, rapid amplification of cDNA ends.
RT, reverse transcriptase.
| |
ACKNOWLEDGMENTS |
We gratefully acknowledge the assistance and skill of Drs. Lin
Chen, Ann MacGregor, and Peilin Xu with various aspects of the
work.
| |
LITERATURE CITED |
Aman P,
Hesselman K,
Tilly A-C
(1985)
The variation in chemical composition of Swedish barleys.
J Cereal Sci
3:
73-77
Amemura A,
Chakraborty R,
Fujita M,
Noumi T,
Futai M
(1988)
Cloning and nucleotide sequence of the isoamylase gene from Pseudomonas amyloderamosa SB-15.
J Biol Chem
263:
9271-9275
[Abstract/Free Full Text]
Baba T,
Nishihara M,
Mizuno K,
Kawasaki T,
Shimada H,
Kobayashi E,
Ohnishi S,
Tanaka K-I,
Arai Y
(1993)
Identification, cDNA cloning, and gene expression of soluble starch synthase in rice (Oryza sativa L.) immature seeds.
Plant Physiol
103:
565-573
[Abstract]
Ball S,
Guan H-P,
James M,
Myers A,
Keeling P,
Mouille G,
Buleon A,
Colonna P,
Preiss J
(1996)
From glycogen to amylopectin: a model for the biogenesis of the plant starch granule.
Cell
86:
349-352
[CrossRef][ISI][Medline]
Banik M,
Garrett TPJ,
Fincher GB
(1996)
Molecular cloning of cDNAs encoding (14)--xylan endohydrolases from the aleurone layer of germinated barley (Hordeum vulgare).
Plant Mol Biol
31:
1163-1172
[CrossRef][ISI][Medline]
Banik M,
Li C-D,
Langridge P,
Fincher GB
(1997)
Structure, hormonal regulation, and chromosomal location of genes encoding barley (14)--xylan endohydrolases.
Mol Gen Genet
253:
599-608
[CrossRef][ISI][Medline]
Bradford MM
(1976)
A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding.
Anal Biochem
72:
248-254
[CrossRef][ISI][Medline]
Cavener DR,
Ray SC
(1991)
Eukaryotic start and stop translation sites.
Nucleic Acids Res
19:
3185-3192
[Abstract/Free Full Text]
Chandler PM,
Jacobsen JV
(1991)
Primer extension studies on -amylase mRNAs in barley aleurone. II. Hormonal regulation of expression.
Plant Mol Biol
16:
637-645
[Medline]
Chandler PM,
Zwar JA,
Jacobsen JV,
Higgins TJV,
Inglis AS
(1984)
The effects of gibberellic acid and abscisic acid on -amylase mRNA levels in barley aleurone layers studied using an -amylase cDNA clone.
Plant Mol Biol
3:
407-418
[CrossRef][ISI]
Chen L,
Fincher GB,
Hoj PB
(1993)
Evolution of polysaccharide hydrolase substrate specificity: catalytic amino acids are conserved in barley (13,14)-glucanase and (13)--glucanase.
J Biol Chem
268:
13318-13326
[Abstract/Free Full Text]
Chrispeels MJ,
Varner JE
(1967)
Gibberellic acid enhanced synthesis and release of -amylase and ribonuclease by isolated barley aleurone layers.
Plant Physiol
42:
398-406
[Abstract/Free Full Text]
Devereux J,
Haeberli P,
Smithies O
(1984)
A comprehensive set of sequence analysis programs for the VAX.
Nucleic Acids Res
12:
387-395
Duffus CM, Cochrane MP (1993) Development, structure and
composition of the barley kernel. In AW MacGregor, RS
Bhatty, eds, Barley: Chemistry and Technology. American Association of
Cereal Chemists, St. Paul, MN, pp 31-72
Edwards A,
Marshall J,
Sidebottom C,
Visser RGF,
Smith AM,
Martin C
(1995)
Biochemical and molecular characterisation of a novel starch synthase from potato tubers.
Plant J
8:
283-294
[CrossRef][ISI][Medline]
Fincher GB
(1989)
Molecular and cellular biology associated with endosperm mobilization in germinating cereal grains.
Annu Rev Plant Physiol Plant Mol Biol
40:
305-346
[CrossRef][ISI]
Frohman M,
Dush M,
Martin G
(1988)
Rapid amplification of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer.
Proc Natl Acad Sci USA
85:
8998-9002
[Abstract/Free Full Text]
Gubler F,
Jacobsen JV
(1992)
Gibberellin-re
Top
Abstract
Introduction