Download Molecular and biochemical characterization of cytosolic

Journal of Experimental Botany, Vol. 54, No. 386, pp. 1351±1360, May 2003
DOI: 10.1093/jxb/erg151
RESEARCH PAPER
Molecular and biochemical characterization of cytosolic
phosphoglucomutase in wheat endosperm
(Triticum aestivum L. cv. Axona)
Emma J. Davies1, Ian J. Tetlow2, Caroline G. Bowsher1 and Michael J. Emes2,3
1
School of Biological Sciences, 3.614 Stopford Building, Oxford Road, University of Manchester, Manchester
M13 9PT, UK
2
Department of Botany, College of Biological Sciences, University of Guelph, Guelph, Ontario, N1G 4C4,
Canada
Received 26 August 2002; Accepted 12 February 2003
Abstract
Introduction
Evidence from a number of plant tissues suggests
that phosphoglucomutase (PGM) is present in both
the cytosol and the plastid. The cytosolic and plastidic isoforms of PGM have been partially puri®ed from
wheat endosperm (Triticum aestivum L. cv. Axona).
Both isoforms required glucose 1,6-bisphosphate for
their activity with Ka values of 4.5 mM and 3.8 mM for
cytosolic and plastidic isoforms, respectively, and
followed normal Michaelis±Menten kinetics with glucose 1-phosphate as the substrate with Km values of
0.1 mM and 0.12 mM for the cytosolic and plastidic
isoforms, respectively. A cDNA clone was isolated
from wheat endosperm that encodes the cytosolic
isoform of PGM. The deduced amino acid sequence
shows signi®cant homology to PGMs from eukaryotic
and prokaryotic sources. PGM activity was measured
in whole cell extracts and in amyloplasts isolated
during the development of wheat endosperm. Results
indicate an approximate 80% reduction in measurable
activity of plastidial and cytosolic PGM between 8 d
and 30 d post-anthesis. Northern analysis showed a
reduction in cytosolic PGM mRNA accumulation during the same period of development. The implications
of the changes in PGM activity during the synthesis
of starch in developing endosperm are discussed.
Phosphoglucomutase (PGM, EC 2.7.5.1) catalyses the
interconversion of glucose 1-phosphate (G1P) and glucose
6-phosphate (G6P), with glucose 1,6-bisphosphate
(G16BP) being a cofactor in this reaction (Ray et al.,
1983). In plant tissues, PGM is present in the cytosol and
the plastid (MuÈhlbach and Schnarrenberger, 1978;
Sangwan and Singh, 1987; Popova et al., 1998). The
PGM reaction is thought to be important in the partitioning
of carbon between the pathways of starch synthesis and
carbohydrate oxidation (Tetlow et al., 1998; Esposito et al.,
1999); alterations in the levels of plastidic and cytosolic
PGM in dicotyledenous species and mutations in plastidic
PGM have been shown to have a deleterious effect on
starch synthesis (Harrison et al., 2000; Tauberger et al.,
2000; Fernie et al., 2001, 2002).
The plastidic PGM isoform provides the substrate (G1P)
for the ADP-glucose pyrophosphorylase (AGPase) reaction, which is the ®rst committed step of the starch
synthetic pathway. In plants that rely upon the import of
G6P into amyloplasts, there is an absolute requirement for
the plastidial isoform of PGM. The absence or reduction of
plastidial PGM in mutants of Arabidopsis thaliana (Caspar
et al., 1985), Nicotiana sylvestris (Hanson and McHale,
1988), pea (Harrison et al., 1998, 2000), and in transgenic
potato plants (Tauberger et al., 2000; Fernie et al., 2001)
have resulted in starchless or near-starchless phenotypes,
indicating that in these dicotyledonous tissues the plastidic
isoform plays an essential role in starch synthesis.
Key words: Phosphoglucomutase, starch synthesis, Triticum
aestivum.
3
To whom correspondence should be addressed. Fax: +1 519 767 1991. E-mail: [email protected]
Abbreviations: ADH, alcohol dehydrogenase; ADPG, ADPglucose; AGPase, ADPglucose pyrophosphorylase; APPase, alkaline inorganic pyrophosphatase;
cyt c oxidase, cytochrome c oxidase; dpa, days post-anthesis; Na2-EDTA, disodium ethylenediaminetetra-acetic acid; G1P, glucose 1-phosphate; G6P,
glucose 6-phosphate; G16BP, glucose 1,6-bisphosphate; PGM, phosphoglucomutase; PMSF, phenyl methyl sulphonyl ¯uoride; UGPase, uridine 5¢
diphosphate glucose pyrophosphorylase.
Journal of Experimental Botany, Vol. 54, No. 386, ã Society for Experimental Biology 2003; all rights reserved
1352 Davies et al.
The antisense repression of the cytosolic PGM isoform
in potato tubers results in dramatic alterations in both plant
morphology and in carbon partitioning between sink and
source organs. As G6P is viewed as the ®rst glycolytic
intermediate it is suggested that the cytosolic PGM isoform
may play an important role in the partitioning of carbon
between the starch synthetic (G1P-utilizing) and glycolytic
pathways (G6P-utilizing) in these tissues (Fernie et al.,
2002).
There is now evidence that in the developing endosperm
of maize (Denyer et al., 1996), barley (Thorbjùrnsen et al.,
1996) and wheat (Beckles et al., 2001; Tetlow et al., 2003)
ADP-glucose (ADPG) can be synthesized in the cytosol
and taken up directly by the amyloplast as well as being
synthesized within the organelle. The possession of a
cytosolic AGPase could bypass the need for the G1P
produced by the plastidial PGM reaction, since ADPG
synthesis could be directly linked to the uridine 5¢
diphosphate glucose pyrophosphorylase (UGPase) reaction. Wheat endosperm can import G1P, G6P and ADPG
into amyloplasts, but whilst G6P can be transported by
these organelles it is not able to support starch synthesis
in vitro (Tetlow et al., 1994, 1998). PGM catalyses a
reaction that at equilibrium results in a 20-fold excess of
G6P over G1P (King, 1970) and, therefore, in this tissue,
PGM may be a potential drain on the G1P pool either in the
cytosol or the amyloplast with respect to starch synthesis.
Consequently, the relationship between PGM and starch
synthesis in developing endosperm is likely to be different
from that observed in other species and organs such as the
potato tuber.
This study aims to determine how the plastidic and
cytosolic isoforms are regulated in relation to starch
synthesis in the developing wheat grain. The puri®cation
and biochemical characterization of plastidic and cytosolic
PGM isoforms from wheat endosperm amyloplasts, are
described here. The isolation of a full-length cDNA
encoding cytosolic PGM of wheat endosperm is also
described. RNA analysis was performed to examine the
developmental regulation of cytosolic PGM mRNA accumulation. The biochemical activity of PGM was measured
in isolated amyloplasts and whole cell extracts to
examine the developmental changes occurring during
grain development.
Materials and methods
Materials
All reagents, substrates and enzymes were purchased from Sigma±
Aldrich Co. Ltd. (Poole, Dorest, UK) and BDH±Merck Ltd.
(Lutterworth, Leicestershire, UK), unless otherwise stated. All
coupling enzymes used in enzymes assays had (NH4)2SO4 removed
by centrifugation at 13 000 g for 5 min; the supernatant was
discarded and the pellet resuspended in the appropriate assay buffer.
All other reagents were of analytical grade and aqueous solutions
were made up with distilled water.
Protease inhibitors (bestatin, E63, pepstatin, and leupeptin) were
purchased from Roche Diagnostics Ltd. (Lewes, East Sussex, UK).
Agarose, RNA ladder (high range) and restriction enzymes (EcoRI
and XhoI) were purchased from Helena Biosciences (Sunderland,
Tyne and Wear, UK). Phenol was purchased from BioGene Ltd.
(Kimbolton, UK).
Plant material, amyloplast isolation, and whole cell extract
preparation
Spring wheat (Triticum aestivum L. cv. Axona) was grown under
conditions previously described (Tetlow et al., 1993) and the
developing ears tagged at the onset of anthesis (the ®rst appearance
of anthers). Endosperm tissue for amyloplast preparations was
obtained from developing grains taken from the mid-ear region of
the head between 5 d and 30 d post-anthesis (dpa). Amyloplasts were
puri®ed by centrifuging twice through a layer of 3% (w/v) 5-(N-2,3dihydroxypropylacetamido)-2, 4, 6-triiodo-N, N¢-bis-(2,3-dihydroxypropyl) isophthalamide (Nycodenz, Robbins Scienti®c, Knowle,
Sohihul, UK) as previously described except that the 0.1% (w/v)
bovine serum albumin was omitted (Tetlow et al., 1993).
Amyloplast stroma was prepared from lysates using the methods
described by Tetlow et al. (2003). To determine total enzyme
activity from wheat endosperm whole cell extracts, 10±20 grains
were isolated, the endosperm dissected out and homogenized in 1
cm3 of ice-cold rupturing buffer (100 mM N-[2-hydroxy-1,1-bis
(hydroxymethyl) ethyl] glycine (Tricine)±NaOH (pH 7.8), 1 mM
disodium-ethylenediaminetetra-acetic acid (Na2-EDTA), 1 mM
dithiothreitol, 1 mM phenyl methyl sulphonyl ¯uoride (PMSF), 5
mM MgCl2, 1 mM bestatin, 1 mM E64, 1 mM pepstatin, and 1 mM
leupeptin). The extract was clari®ed by centrifugation at 13 000 g for
30 min at 4 °C, and the supernatant removed and assayed
immediately for PGM and subcellular marker enzymes (see below).
Enzyme assays
PGM (EC. 2.7.5.1) was assayed at 25 °C in a 1 cm3 reaction mixture
containing 30 mM 1,3-diaza 2,4-cyclopentadiene (imidazole)±HCl
(pH 7.4), 3.3 mM magnesium chloride (MgCl2), 0.85 mM
nicotinamide adenine dinucleotide, oxidized (NAD), 0.9 mM
Na2-EDTA, 15 mM G16BP, and 0.7 U glucose 6-phosphate
dehydrogenase (from Leuconostoc mesenteroides). Assays were
initiated with 1.5 mM G1P and increased absorbance was measured
at 340 nm.
The substrate af®nities for each partially puri®ed isoform were
measured by varying G1P and G16BP independently. In all assays
the G1P used was a preparation essentially free from G16BP (Sigma,
grade IV).
For each preparation of isolated amyloplasts the following
enzyme assays were carried out at 25 °C according to methods
described previously: alkaline inorganic pyrophosphatase (APPase,
EC 3.6.1.1) (Gross and ap Rees, 1986), UGPase (EC 2.7.7.9)
(MuÈller-RoÈber et al., 1992), AGPase (EC 2.7.7.27) (Journet and
Douce, 1985, except that G16BP was omitted), alcohol dehydrogenase (ADH, EC 1.1.1.1) (MacDonald and ap Rees, 1983) and
cytochrome c oxidase (cyt c oxidase, EC 1.9.3.1) (Darley-Usmar
et al., 1987).
Partial puri®cation of PGM isoforms
Unless otherwise stated, all enzyme puri®cation steps were carried
out at 4 °C.
Solid ammonium sulphate was added to wheat endosperm whole
cell extracts and isolated amyloplast stroma to 40% saturation and
allowed to precipitate for 30 min. The precipitate was removed by
centrifugation at 10 000 g for 20 min and the supernatant brought to
60% saturation by adding further amounts of ammonium sulphate.
The precipitate, collected by centrifugation at 10 000 g for 20 min,
Characterization of cytosolic phosphoglucomutase in wheat endosperm 1353
3
was dissolved in 1 cm buffer containing 50 mM 2-amino-2hydroxymethylpropane-1,3-diol (Tris)±HCl (pH 8.0), 1 mM
Na2-EDTA and 1 mM PMSF (buffer A).
The above enzyme preparations (2 cm3) from both whole cell
extracts and amyloplasts were applied to a 1 cm3 phenyl sepharose
column (Amersham Pharmacia, Little Chalfont, Buckinghamshire,
UK) previously equilibrated with buffer A containing 1 M
ammonium sulphate at a ¯ow rate of 1 cm3 min±1. The column
was washed with 2 vols of buffer A with 1 M ammonium sulphate
and then 1 cm3 fractions collected at a ¯ow rate of 1 cm3 min±1 using
a stepwise gradient of ammonium sulphate from 1.0 to 0 M.
Fractions containing PGM activity were pooled and dialysed for 2 h
in 2.0 l of buffer A.
The pooled fractions (2 cm3) were applied to a 1 cm3 HiTrap Q
column (Amersham Pharmacia, Little Chalfont, Buckinghamshire,
UK) pre-equilibrated with buffer A at a ¯ow rate of 1 cm3 min±1. The
column was washed with 2 vols of buffer A and then 1 cm3 fractions
collected at a ¯ow rate of 1 cm3 min±1 using a stepwise gradient of
KCl from 0 to 0.2 M.
Starch and protein determination
Starch was extracted and measured according to the methods
described in Tetlow et al. (1994). Protein was measured using a
Bio-Rad protein assay dye reagent (Bio-Rad, Hemel Hempstead,
Hertfordshire, UK) based on the method of Bradford (1976) using
thyroglobulin as the standard.
Library screening
A cDNA library was initially prepared with mRNA isolated from
wheat (Triticum aestivum cv. Axona) endosperm 10±14 dpa using a
Stratagene lZap library kit and according to the manufacturer's
instructions. The library was hybridized at 50 °C overnight with a
full-length 2.1 kbp oilseed rape plastidic PGM cDNA clone
(Harrison et al., 2000) radioactively labelled with [a-32P]-dCTP
using a random primer (Stratagene). Filters were washed under low
stringency conditions for three 30 min washes using 23 SSC and
0.1% SDS at 50 °C.
RNA isolation and hybridization
Total RNA was isolated from different developmental stages (5±30
dpa) of wheat endosperm according to the method described by
Lahners et al. (1988). For RNA blot hybridization 10 mg of the
isolated total RNA from each developmental stage was denatured
with formaldehyde and subjected to electrophoresis and blotted onto
a Hybond N+ membrane as described previously (Fonseca et al.,
1997). The probe used for the hybridization was either the isolated
wheat cPGM or a wheat rRNA cDNA clone. In all cases, DNA
probes were radiolabelled with [a-32P]-dCTP using a random
priming kit (Stratagene), prehybridized, hybridized and exposed to
®lm as described previously (Fonseca et al., 1997). Autoradiograms
were scanned and relative amounts of mRNAs determined as
described previously (Fonseca et al., 1997).
Results
Enzyme puri®cation
Prior to puri®cation of the plastidial PGM isoform from
isolated amyloplasts, marker enzyme assays were carried
out to check that the amyloplasts were essentially free from
contamination by other cell compartments (Table 1). Less
than 1% of the total activities of the cytosolic marker
enzymes, ADH and UGPase, and the mitochondrial marker
enzyme, cyt c oxidase, were recovered in the pellet,
Table 1. Distribution of enzymes between cytosol and
amyloplasts in wheat endosperm
Activities of marker enzymes in preparations of amyloplasts puri®ed
from cell-free extracts of developing wheat endosperm at
approximately 14 dpa. Values in bold parentheses are estimates of the
percentage of the total AGPase and PGM activity that is plastidial,
and are calculated from the following equation of Beckles et al.
(2001) which is used to estimate the percentage of extra-plastidial
AGPase activity (using APPase as the plastidial marker enzyme and
UGPase as the cytosolic marker enzyme): % Extra-plastidial
activity=(% Plastid marker activity in pellet±% Total AGPase activity
in pellet)/(% Plastid marker activity in pellet±% Cytosol marker
activity in pellet) All values are means 6SEM of three amyloplast
preparations with each sample assayed in triplicate.
Enzyme
Activity recovered
in supernatant and
pellet as % of
whole cell extracts
% Recovery in
amyloplasts
from whole
cell extracts
APPase
UGPase
ADH
Cyt c oxidase
AGPase
PGM
98.963.7
111.566.2
102.364.1
89.361.3
97.863.2
105.267.1
25.3964.61
0.2460.07
0.1560.06
0.2060.08
8.2062.12 [32%]
6.7360.99 [26%]
indicating that the level of contamination of plastids by
cytosol and mitochondria was low. The sum of the enzyme
activities in all the supernatant and pellet fractions
obtained during an amyloplast preparation was approximately 100% of those in the original homogenate,
indicating that there was no serious loss of enzyme activity
during the preparation of amyloplasts. Allowing for
organelle breakage and cytosolic contamination, approximately 26% of total cellular PGM is calculated as being
present in amyloplasts (calculated from Table 1).
PGM activity was precipitated between 40% and 60%
ammonium sulphate saturation using either wheat endosperm tissue extracts or isolated amyloplasts. The total
PGM activity of the amyloplast preparations more than
doubled following ammonium sulphate fractionation, but
this was not the case in the whole cell extracts (Table 2).
PGM activity eluted in a single peak from the phenyl
sepharose column between 1.0 M and 0.8 M ammonium
sulphate (Fig. 1A). The enzyme preparation from whole
cell extracts was resolved into two peaks of PGM activity
following anion exchange chromatography (Fig. 1B).
PGM peak 1 was eluted at approximately 0.1 M KCl and
PGM peak 2 at approximately 0.15 M KCl. The enzyme
preparation from isolated amyloplasts when loaded onto a
Hi Trap Q column was resolved in a single peak of PGM
activity eluting at 0.1 M KCl (Fig. 1C), coincident with the
®rst peak obtained from whole cell extracts. This suggests
that PGM peak 1 is the plastidic isoform of PGM and PGM
peak 2 is the cytosolic isoform of PGM. The two isoforms,
cytosolic and plastidic were puri®ed 120-fold and 50-fold
respectively (Table 2).
1354 Davies et al.
Table 2. Puri®cation of PGM isoforms from wheat endosperm whole cell extracts and amyloplast stroma
Wheat endosperm whole cell extracts (WSE) were prepared from 200 ears of developing wheat taken at approximately 12±20 dpa. Puri®cation
of the two PGM isoforms was carried out as described in the Materials and methods section. Amyloplasts (AP) from wheat endosperm (isolated
at approximately 12±20 dpa) were isolated and puri®cation of the plastidic PGM isoform was carried out as described in the Materials and
methods section. Data shown is from a single representative puri®cation experiment.
Puri®cation stage
Homogenate
Ammonium sulphate
precipitation (40±
60% saturation)
Phenyl sepharose
Hi Trap Q
PGM peak 1
PGM peak 2
Volume
(ml)
Total
protein
(mg)
Speci®c
activity
(mmol
min±1 mg±1
protein)
WSE
AP
WSE
AP
430
14
10
4
229.6
31.0
22.64
3.39
0.24
0.78
0.13
1.85
1
3.22
1
14.57
2
2
3.5
0.78
2.93
6.84
12.12
53.86
2
0.23
0.36
0.33
±
26.92
12.04
13.59
±
111.20
49.96
106.61
±
2
2
±
Biochemical characterization of the partially puri®ed
PGM isoforms
The two partially puri®ed isoforms of PGM were
characterized with respect to pH, substrate af®nities and
molecular weight. The two isoforms of PGM had the same
pH optimum, but in different buffers. The optimum assay
conditions for the plastidic isoform was found to be pH 8.0
in glycylglycine buffer whereas the cytosolic isoform had
optimal activity at pH 8.0 in Tris buffer. Both isoforms of
PGM, at their respective pH optima, followed normal
Michaelis±Menten kinetics with respect to G1P as the
substrate. Km values for the cytosolic and plastidic
isoforms as determined by Lineweaver±Burk plots were
99.761 mM (n=3) and 12269 mM (n=3), respectively. The
two isoforms had an absolute requirement for G16BP for
their activity and the Ka values were 4.4760.31 mM (n=3)
and 3.7860.36 mM (n=3), respectively, for the cytosolic
and plastidic isoforms. The two isoforms had identical
molecular weights of 61 kDa as determined by gel
permeation chromatography and SDS electrophoresis
(data not shown). A number of metabolites (ATP, AMP,
ADP, 3-phosphoglyceric acid, inorganic pyrophosphate,
ribulose 1,5-bisphosphate, and inorganic orthophosphate)
were added to the PGM assay to determine whether they
caused an activation or inhibition of the PGM isoforms.
However, it was found that none of these metabolites had
any signi®cant effect on PGM activity (data not shown).
Sequence analysis of wheat endosperm cytosolic
PGM
Screening of a wheat endosperm cDNA library with a
plastidic oil seed rape cDNA clone that had sequence
similarity to known PGM sequences resulted in the
isolation of a cDNA clone that encoded a cytosolic
isoform of PGM (EMBL accession number AJ313311).
The full-length cDNA clone of cytosolic PGM from the
WSE
Puri®cation
-fold
AP
WSE
AP
wheat endosperm library was 2358 bp in length. This
comprised a 339 bp 5¢ untranslated region, a 1743 bp open
reading frame and a 276 bp 3¢ untranslated region. The
open reading frame encoded a predicted protein of 581
amino acids (Fig. 2), with a predicted molecular mass of 63
kDa. Further con®rmation that the PGM cDNA clone
isolated was cytosolic came from analysis using PSORT
and Chloro V1.1 computer programs (Emanuelsson et al.,
1999). Neither of these transit peptide site prediction
programs identi®ed any sequence area that would suggest
the isolated clone could be targeted to the plastid.
Comparison of the deduced amino acid sequences of the
cytosolic PGM from wheat endosperm to that of PGM
sequences of cytosolic Bromus inermis (AF197925),
plastidic Solanum tuberosum (AJ240053), Homo sapiens
(M83088), and Saccharomyces cerevisiae (P33401),
revealed long stretches of identical residues as shown in
Fig. 2. The wheat endosperm cytosolic PGM had the
highest amino acid sequence identity (80±90%) with the
deduced cytosolic sequences of PGMs of Solanum
tuberosum (AJ240054), Pisum sativum (AJ250769),
Arabidopsis thaliana (AJ242650), Populus tremula
(AF097938), Mesembryanthemum crystallinum (U84888),
Zea mays 1 (U89341), Zea mays 2 (U89342), and Bromus
inermis (AF197925). The cytosolic cDNA clone showed
approximately 60% homology to plastidic sequences and
approximately 50% homology to sequences from Homo
sapiens, Saccharomyces cerevisiae and Agrobacterium
tumefaciens (P39671). However, it had considerably lower
homology (approximately 25%) to PGM sequences of
E. coli (U08369) and spinach chloroplasts (X75898).
Phylogenetic analysis of PGM proteins
An alignment of protein sequences with homology to
wheat endosperm cytosolic PGM was analysed to
determine the probable phylogeny of the proteins by
Characterization of cytosolic phosphoglucomutase in wheat endosperm 1355
plastidic isoforms of the enzymes, indicating how the
different PGMs have diverged during evolution.
Expression of cytosolic PGM mRNA transcripts during
wheat endosperm development
A PGM transcript of approximately 2 kb was identi®ed in
the wheat endosperm and was found to be developmentally
regulated. Figure 4 shows the normalized densitometry
readings and indicates that the transcript levels peak at 8
dpa and decline thereafter throughout the period of
endosperm development studied.
Fig. 1. Elution pro®les of PGM activity. (A) Elution pro®le of PGM
activity from wheat endosperm whole cell extract following
chromatography on a 1 cm3 phenyl sepharose column. Approximately
2 mg of protein was loaded onto the column. Data are means of three
replicates 6SEM and 1 unit is equivalent to 1 mmol min±1. In (A) to
(C), the ®lled squares show PGM activity and the open squares show
protein content. (B) Elution of PGM activity from wheat endosperm
whole cell extracts following anion exchange chromatography.
Approximately 1 mg of protein containing PGM activity eluted from
the phenyl sepharose column was loaded onto a 1 cm3 HiTrap Q
column. (C) Elution of PGM activity from wheat endosperm
amyloplast stroma following anion exchange chromatography.
Approximately 1 mg of amyloplast stromal protein containing PGM
activity was loaded onto a 1 cm3 HiTrap Q column following elution
from the phenyl sepharose column.
the parsimony method (Clustal W). The relationship
between PGM sequences from different species is
shown in Fig. 3.
Analysis of an unrooted phylogenetic tree generated
from an alignment of PGM protein sequences revealed
three distinct classes corresponding to plant, bacterial and
mammalian origin. The plant class of PGM is further
divided into two sections corresponding to cytosolic and
Activity of PGM isoforms during wheat endosperm
development
Whole cell extracts and amyloplasts were prepared from
wheat endosperm at different stages of development.
Marker enzyme assays were performed on isolated
amyloplasts and whole cell extracts from each developmental stage, which showed the same level of contamination from cytosol and mitochondria in the plastids and the
same enzyme recovery in the supernatant and pellet to
those ®gures shown in Table 1.
The total activity of PGM, in whole cell extracts, peaks
at 11 dpa and then declines by approximately 80% over the
period of development studied (Fig. 5A). Plastidial PGM
activity declines between 8 and 30 dpa by approximately
90% (Fig. 5B). The activity pro®le observed for PGM
activity in whole cell extracts follows the same pro®le
obtained for RNA expression of the cytosolic isoform of
PGM (Fig. 4), and is consistent with the observation that
the cytosolic enzyme constitutes the greater part of the
total activity (Table 1).
The amount of PGM activity in amyloplasts varied
from 13% to 32% during development based on marker
enzyme recoveries (data not shown). At 14 dpa approximately 26% of the total PGM activity is plastidial, which is
consistent with previous determinations made in developing wheat endosperm (Esposito et al., 1999). Taking
account of this and the measured total PGM activity per
endosperm the calculated changes in the amount of
cytosolic and plastidic PGM isoforms were determined
(Fig. 5C). Between 5 and 30 dpa the starch content of the
developing endosperm increased steadily from 3.9660.90
(n=4) mg. per endosperm (5 dpa) to 16.7260.62 (n=4) mg
per endosperm at 30 dpa (Fig. 5C).
The relative changes of both isoforms of PGM, when
activity is expressed per endosperm, peak at approximately
11 dpa and then steadily decline throughout the period of
development studied (Fig. 5C). The relative changes of
plastidial PGM are slightly different to those obtained
when expressed per mg protein (Fig. 5B). This may be due
to the fact that the protein content of the tissue also changes
during development.
1356 Davies et al.
Discussion
PGM activity in wheat endosperm was found to be present
in two cell compartments (cytosol and plastid) and the
properties of the two isoforms are very similar. Many of
these properties are shared with PGMs from other plant
species, mammalians and micro-organisms. The two
partially puri®ed PGM isoforms were characterized
biochemically and it was shown that they have many
basic properties in common with each other and with
PGMs from other species such as potato tubers (Pressey,
1967) and developing seeds of Cassia corymbosa (Small
and Matheson, 1979): with respect to pH optima, molecular size and substrate and activation af®nities.
The Km values for G1P for the partially puri®ed
isoforms from wheat endosperm are approximately
2-fold higher than those reported for the PGM isoforms
of spinach leaves (MuÈhlbach and Schnarrenberger, 1978),
which have Km values of 40 mM and 60 mM for plastidic
and cytosolic isoforms respectively, and a PGM isoform
from isolated amyloplasts of wheat endosperm (40 mM,
Esposito et al., 1999). Two isoforms of PGM have been
identi®ed in wheat endosperm previously (Sangwan and
Singh, 1987). In that report, the peaks of PGM activity
were resolved after anion exchange chromatography of
tissue extracts, but the authors assigned the ®rst peak
(corresponding to peak 1 of Fig. 1B) as being cytosolic in
origin and the second peak as plastidic. No sequence
analysis of the isoforms, or separation of organelles was
performed to con®rm this and the conclusions drawn were
based on small differences in kinetic properties of the two
isozymes. It has been clearly shown in this report that this
annotation was incorrect and that it is the PGM isozyme
which elutes, from anion exchange chromatography, at the
lower salt concentration which is plastidic in origin. Ka
values for G16BP reported here (4.5 mM and 3.8 mM for
cytosolic and plastidic isoforms, respectively) are lower
than those reported for spinach leaf PGM (12.5 mM for the
cytosolic isoform and 10 mM for the plastidic isoform)
(MuÈhlbach and Schnarrenberger, 1978), but substantially
higher than that previously reported for the plastid PGM in
wheat endosperm (0.6 mM, Esposito et al., 1999).
Although these Ka values for G16BP in wheat amyloplasts
are in the same order of magnitude, they may be a result of
Fig. 2. Amino acid sequence alignment of PGMs from different
species. Key to ®gure: 1, wheat endosperm cytosolic PGM
(AJ313311); 2, Bromus inermis cytosolic PGM (AF197925); 3,
Solanum tuberosum plastidic PGM (AJ240053); 4, Homo sapiens
PGM (M83088), and 5, Saccharomyces cerevisiae PGM (P33401).
The star (*) represents amino acids that are identical and the dot (.)
indicates where the amino acids are similar or conserved. The
catalytic centre and the metal ion-binding site, the two conserved
regions of PGM, are indicated by the grey boxes. The sequence
alignment was performed using the program Clustal W (Thompson
et al., 1994).
Characterization of cytosolic phosphoglucomutase in wheat endosperm 1357
Fig. 3. Unrooted phylogenetic tree of PGMs from different species.
An optimal amino acid alignment was created with the program
Clustal W and a phylogenetic tree was constructed. The tree was
displayed with the Treeview program. The plant cytosolic PGM
sequences used in the alignment were wheat (AJ313311), M.
crystallinum (U84888), Bromus grass (AF197925), maize 1 (U89341),
maize 2 (U89342), potato (AJ240054), Populus (AF097938), pea
(AJ250769), and Arabidopsis (AJ242650); and the plastidic sequences
used were pea (AJ250770), potato (AJ240053), oilseed rape
(AJ250771), Arabidopsis (AJ242601), and spinach (X75898). The
other PGM sequences used in the alignment were human (M83088),
yeast (P33401), E. coli (U08369), and A. tumefaciens (P39671).
differences in enzyme preparation; the PGM isoforms in
the present study had each been puri®ed several-fold from
whole cell extracts, whereas the plastid PGM analysed by
Esposito et al. (1999) was prepared by lysing puri®ed
amyloplasts. The intracellular concentrations of the
activator are not known and it cannot be concluded that
these differences are signi®cant. However, estimates of the
cytosolic G16BP concentration in developing wheat
endosperm cells (approximately 8 mM) would suggest
that both wheat PGM isoforms will be fully activated
in vivo (Esposito et al., 1999). Although it has previously
been observed that the mass action ratio for PGM in wheat
endosperm amyloplasts can be displaced signi®cantly from
the equilibrium constant (Tetlow et al., 1998), no simple
explanation for this phenomenon was found based on
kinetic properties and response to other metabolites and it
is concluded that other mechanisms, possibly including
post-translational modi®cation, must be involved in the
regulation of this enzyme.
The sequence of the cytosolic wheat endosperm cDNA
cloned here contains many regions that are conserved in
human, yeast, cytosolic Bromus inermis and plastidic
potato PGM sequences (Fig. 2). Regions of high homology
Fig. 4. RNA expression of the cytosolic isoform of PGM during
wheat endosperm development between 5 and 30 dpa. (A) Northern
analysis of PGM expression, (B) hybridization of rRNA speci®c
cDNA probe to the same blot in (A), and (C) normalized densitometry
readings of probe hybridization shown in (A) and (B). Approximately
10 mg of RNA was loaded in each lane.
included the catalytic centre (T/S-A-S-H-N motif), as
identi®ed in rat PGM by Milstein and Sanger (1961), and
the metal ion-binding loop (D-G-D-G/A-D motif), as
described in the crystal structure of muscle PGM by Dai
et al. (1992). The serine (Ser) located at residue 116 in
human PGM serves as the phosphate acceptor/donor in the
catalytic process (Ray and Peck, 1972). The catalytic site
was found at positions 121 to 125 in cytosolic wheat
endosperm PGM, with Ser-123 being the acceptor/donor
and the metal ion-binding loop was located at residues 298
to 302, similar to the enzyme from maize (Manjunath et al.,
1998).
Phylogenetic analysis (Fig. 3) indicates the separation of
the PGMs into classes corresponding to plant, bacterial and
mammalian origin. The plant enzyme is further divided
into two classes corresponding to plastidic and cytosolic,
as there is a distinct separation between the two isoforms.
The plastidic sequences have approximately equal distances between plant and prokaryote branches and is
consistent with the hypothesis that plastids are derived
from ancient prokaryote symbionts. Branches from
cytosolic and plastidic plant sequences, yeast, bacteria,
and mammalian have approximate equal distances indicating the general homology between all PGM sequences,
and suggesting that the PGM sequences from the different
1358 Davies et al.
Fig. 5. Changes in PGM activity during the development of wheat
endosperm. (A) PGM activity measured in whole cell extracts of
wheat endosperm between 8 and 30 dpa; (B) plastidial PGM activity
measured in amyloplast stroma between 8 and 30 dpa and (C) changes
in the relative levels of plastidic (®lled squares) and cytosolic (®lled
diamonds) PGM compared with the starch content (®lled circles)
measured in the endosperm between 5 and 30 dpa. PGM assays were
performed as described in the Materials and methods section under
optimized conditions with saturating concentrations of G1P and
G16BP. Each value is the mean 6 SEM of four independent
experiments.
kingdoms diverged from each other early in evolution and
then evolved independently.
A PGM transcript of approximately 2 kb was identi®ed
(Fig. 4) in the wheat endosperm and was found to be
developmentally regulated con®rming that regulation is
transcriptional. Interestingly, the major period of starch
synthesis in wheat endosperm begins at approximately
10 dpa (Briarty et al., 1979) when the PGM transcript
levels start to decline. The reason for the failure to isolate a
wheat plastidic cDNA clone using the 2.1 kbp oilseed rape
plastidic cDNA clone is unclear. It may be that, at the
developmental stage the endosperm cDNA library was
prepared (10±14 dpa), mRNA transcript levels for this
protein are too low for its isolation.
Recent work carried out by Tetlow et al. (2003)
measured the activity of AGPase throughout wheat
endosperm development. This work together with the
data presented here suggests that there is an inverse
relationship between the two enzymes. When PGM
activity is high, the AGPase activity is low giving rise to
a high ratio between PGM and AGPase. The ratio between
PGM and AGPase changes as AGPase activity increases
and PGM activity decreases (as the starch content of the
endosperm is increasing). The high activity of the PGM
isoforms early in development may be due to the fact that,
at this stage of development, there is a high demand for
substrates for energy, cell division, respiration, and
glycolysis.
In tissues containing both a cytosolic and a plastidic
AGPase, for example, maize (Denyer et al., 1996), barley
(Thorbjùrnsen et al., 1996), rice (Sikka et al., 2001), and
wheat (Tetlow et al., 2003), it may be that neither the
plastidic nor cytosolic PGM isoforms contribute greatly to
starch synthesis to the same extent as in other tissues, for
example, potato, where AGPase is exclusively plastidial
(Fernie et al., 2001, 2002). The decline observed in PGM
expression and activity in the cytosol and amyloplasts of
developing wheat endosperm during the major period of
grain ®lling, is consistent with the notion that its role is not
as important as in tissues containing only plastidic
AGPase. Interestingly, Pan et al. (1990) tested a number
of maize inbred lines and found that the majority had only
one form of PGM activity and that this activity was due to
the plastidic isozyme. The lack of the cytosolic isozyme in
these maize inbred lines had no discernible phenotypic
effects and caused no reduction in starch content (Pan et al.,
1990) reinforcing the view that its role in starch synthesis
is not critical in cereal endosperms. The import of ADPG
from the cytosol in such tissues would bypass the need for
the import of hexose phosphates, therefore decreasing the
need for the interconversion of hexose phosphates by the
PGM isoforms in the cytosol and the amyloplast.
Acknowledgements
This work was ®nancially supported by the Biotechnology and
Biological Sciences Research Council (BBSRC) through a RASP
studentship. We thank Dr T Wang (John Innes Centre, Norwich) for
the gift of the plastidic oil seed rape clone and Mr TW Heaton for
Characterization of cytosolic phosphoglucomutase in wheat endosperm 1359
growing the wheat at the Botany Experimental Grounds
(Manchester).
References
Bradford MM. 1976. A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein±dye-binding. Analytical Biochemistry 72,
248±254.
Briarty LG, Hughes CE, Evers AD. 1979. The developing
endosperm of wheatÐa stereological analysis. Annals of Botany
44, 641±658.
Beckles DM, Smith AM, ap Rees T. 2001. A cytosolic ADPglucose pyrophosphorylase is a feature of graminaceous
endosperms, but not of other starch storing organs. Plant
Physiology 125, 818±827.
Caspar T, Huber SC, Somerville C. 1985. Alterations in growth,
photosynthesis and respiration in a starchless mutant of
Arabidopsis
thaliana
(L.)
de®cient
in
chloroplast
phosphoglucomutase activity. Plant Physiology 79, 11±17.
Dai J, Liu Y, Ray WJ, Konno M. 1992. The crystal structure of
muscle phosphoglucomutase re®ned at 2.7 angstrom resolution.
Journal of Biological Chemistry 267, 6322±6337.
Darley-Usmar VM, Capaldi RA, Takamiya S, Millet F, Wilson
MT, Malatesta F, Sart P. 1987. Reconstitution and molecular
anlaysis of the respiratory chain. In: Darley-Usmar VM,
Rickwood D, Wilson MT, eds. MitochondriaÐa practical
approach. Oxford, UK: IRL Press Ltd., 137.
Denyer K, Dunlap F, Thorbjùrnsen T, Keeling P, Smith AM.
1996. The major form of ADP-glucose pyrophosphorylase in
maize endosperm is extra plastidial. Plant Physiology 112, 779±
785.
Emanuelsson O, Nielsen H, von Heijne G. 1999. ChloroP, a
neural network-based method for predicting chloroplast transit
peptides and their cleavage sites. Protein Science 8, 978±984.
Esposito S, Bowsher CG, Emes MJ, Tetlow IJ. 1999.
Phosphoglucomutase activity during development of wheat
grains. Journal of Plant Physiology 154, 24±29.
Fernie AR, Roessner U, Trethewey RN, Willmitzer L. 2001. The
contribution of plastidial phosphoglucomutase to the control of
starch synthesis within the potato tuber. Planta 213, 418±426.
Fernie AR, Tauberger E, Lytovchenko A, Roessner U,
Willmitzer L, Trethewey RN. 2002. Antisense repression of
cytosolic phosphoglucomutase in potato (Solanum tuberosum)
results in severe growth retaration, reduction in tuber number and
altered carbon metabolism. Planta 214, 510±520.
Fonseca F, Bowsher CG, Stulen I. 1997. Impact of elevated CO2
on nitrate reductase transcription and activity in leaves and roots
of Plantago major. Physiologia Plantarum 100, 940±948.
Gross P, ap Rees T. 1986. Alkaline inorganic pyrophosphatase and
starch synthesis in amyloplasts. Planta 167, 140±145.
Hanson KR, McHale NA. 1988. A starchless mutant of Nicotiana
sylvestris containing a modi®ed plastid phosphoglucomutase.
Plant Physiology 88, 838±844.
Harrison CJ, Hedley CL, Wang TL. 1998. Evidence that the rug 3
locus of pea (Pisum sativum L.) encodes plastidial
phosphoglucomutase con®rms that the imported substrate for
starch synthesis in pea amyloplasts is glucose 6 phosphate. The
Plant Journal 13, 753±762.
Harrison CJ, Mould RM, Leech MJ, et al. 2000. The rug 3 locus
of pea encodes plastidial phosphoglucomutase. Plant Physiology
122, 1187±1192.
Journet E, Douce R. 1985. Enzymic capacities of puri®ed
cauli¯ower bud plastids for lipid synthesis and carbohydrate
metabolism. Plant Physiology 79, 458±467.
King J. 1970. Phosphoglucomutase. In: Bergmeyer HU, ed.
Methoden der Enzymatischen Katalyse. Weinheim, FRG:
Verlag Chemie, 764±785.
Lahners K, Kramer V, Back E, Privalle L, Rothstein S. 1988.
Molecular cloning of complementary DNA encoding maize
nitrite reductase. Molecular analysis and nitrate induction. Plant
Physiology 88, 741±746.
MacDonald FD, ap Rees T. 1983. Enzymic properties of
amyloplasts from suspension cultures of soybean. Biochimica et
Biophysica Acta 755, 81±89.
Manjunath S, Lee CK, VanWinkle P, Bailey-Serres J. 1998.
Molecular and biochemical characterization of cytosolic
phosphoglucomutase in maize. Plant Physiology 117, 997±1006.
Milstein C, Sanger F. 1961. An amino acid sequence in the active
centre of phosphoglucomutase. Biochemistry Journal 79, 456±
469.
MuÈhlbach H, Schnarrenberger C. 1978. Properties and
intracellular distribution of two phosphoglucomutases from
spinach leaves. Planta 141, 65±70.
MuÈller-RoÈber B, Sonnewald U, Willmitzer L. 1992. Inhibition of
the ADP-glucose pyrophosphorylase in transgenic potatoes leads
to sugar storing tubers and in¯uences tuber formation and
expression of tuber storage protein genes. EMBO Journal 11,
1229±1238.
Pan D, Strelow LI, Nelson OE. 1990. Many maize inbreds lack an
endosperm cytosolic phosphoglucomutase. Plant Physiology 93,
1650±1653.
Popova TN, Matasova LV, Lapot'ko AA. 1998. Puri®cation,
separation and characterization of phosphoglucomutase and
phosphomannosemutase from maize leaves. Biochemistry and
Molecular Biology International 46, 461±470.
Pressey
R.
1967.
Puri®cation
and
properties
of
phosphoglucomutase from potato tubers. Journal of Food
Science 32, 381±385.
Ray WJ, Peck EJ. 1972. Phosphomutases. In: Boyer PD, ed. The
enzymes, 3rd edn. Vol. 6. New York, USA: Academic Press, 407±
477.
Ray WJ, Hermodson MA, Puvathingal JM, Mahoney WC. 1983.
The complete amino acid sequence of rabbit muscle
phosphoglucomutase. Journal of Biological Chemistry 258,
9166±9174.
Sangwan RS, Singh R. 1987. Multiple forms of
phosphoglucomutase from immature wheat endosperms. Plant
Physiology and Biochemistry 25, 745±751.
Sikka VK, Choi S, Kavakli IH, Sakulsingharoj C, Gupta S, Ito
H, Okita TW. 2001. Subcellular compartmentation and allosteric
regulation of the rice endosperm ADPglucose pyrophosphorylase.
Plant Science 161, 461±468.
Small DM, Matheson NK. 1979. Phosphomannosemutase and
phosphoglucomutase in developing Cassia corymbosa seeds.
Phytochemistry 18, 1147±1150.
Tauberger E, Fernie AR, Emmermann M, Renz A,
Kossmann J, Willmitzer L, Trethewey RN. 2000.
Antisense inhibition of plastidial phosphoglucomutase
provides compelling evidence that potato tuber amyloplasts
import carbon from the cytosol in the form of glucose 6
phosphate. The Plant Journal 23, 43±53.
Tetlow IJ, Blissett KJ, Emes MJ. 1993. A rapid method for the
isolation of puri®ed amyloplasts from wheat endosperm. Planta
189, 597±600.
Tetlow IJ, Blissett KJ, Emes MJ. 1994. Starch synthesis and
carbohydrate oxidation in amyloplasts from developing wheat
endosperm. Planta 194, 454±460.
Tetlow IJ, Blissett KJ, Emes MJ. 1998. Metabolite pools during
starch synthesis and carbohydrate oxidation in amyloplasts
isolated from wheat endosperm. Planta 204, 100±108.
1360 Davies et al.
Tetlow IJ, Davies EJ, Vardy KA, Bowsher CG, Burrell MM,
Emes MJ. 2003. Subcellular localization of ADPglucose
pyrophosphorylase in developing wheat endosperm and analysis
of the properties of a plastidial isoform. Journal of Experimental
Botany 54, 715±725.
Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W:
improving the sensitivity of progressive multiple sequence
alignment through sequence weighting, position speci®c gap
penalties and weight matrix choice. Nucleic Acids Research 22,
4473±4680.
Thorbjùrnsen T, Villand P, Denyer K, Olsen O, Smith AM.
1996. Distinct forms of ADPglucose pyrophosphorylase occur
inside and outside the amyloplasts in barley endosperm. The
Plant Journal 10, 243±250.