Báo cáo khoa học: Regulation of maize lysine metabolism and endosperm protein synthesis by opaque and floury mutations potx

Eur. J. Biochem. 270, 4898–4908 (2003) Ó FEBS 2003

doi:10.1046/j.1432-1033.2003.03890.x

Regulation of maize lysine metabolism and endosperm protein
synthesis by opaque and floury mutations
´
Ricardo A. Azevedo1, Catherine Damerval2, Jacques Landry3, Peter J. Lea4, Claudia M. Bellato5,
Lyndel W. Meinhardt6, Martine Le Guilloux2, Sonia Delhaye3, Alejandro A. Toro1, Salete A. Gaziola1
and Bertha D. A. Berdejo1
1

Departamento de Gene´tica, Escola Superior de Agricultura Luiz de Queiroz, Universidade de Sa˜o Paulo, Sa˜o Paulo, Brazil;
Station de Ge´ne´tique Ve´ge´tale INRA/UPS/INA-PG/CNRS UMR 8120, La Ferme du Moulon, Gif-sur-Yvette; 3INRA, Laboratoire
de Chimie Biologique, INA-PG, F78850 Thiverval-Grignon, France; 4Department of Biological Sciences, University of Lancaster,
Lancaster, UK; 5Centro de Energia Nuclear na Agricultura, Universidade de Sa˜o Paulo, Sa Paulo, Brazil; and 6Departamento de
˜o
Gene´tica e Evoluca˜o, Universidade Estadual de Campinas, Campinas, Brazil
¸

2

The capacity of two maize opaque endosperm mutants
(o1 and o2) and two floury (fl1 and fl2) to accumulate lysine
in the seed in relation to their wild type counterparts
Oh43+ was examined. The highest total lysine content was
3.78% in the o2 mutant and the lowest 1.87% in fl1, as
compared with the wild type (1.49%). For soluble lysine, o2
exhibited over a 700% increase, whilst for fl3 a 28% decrease
was encountered, as compared with the wild type. In order to
understand the mechanisms causing these large variations in

both total and soluble lysine content, a quantitative and
qualitative study of the N constituents of the endosperm has
been carried out and data obtained for the total protein,
nonprotein N, soluble amino acids, albumins/globulins,
zeins and glutelins present in the seed of the mutants. Following two-dimensional PAGE separation, a total of 35

different forms of zein polypeptides were detected and considerable differences were noted between the five different
lines. In addition, two enzymes of the aspartate biosynthetic
pathway, aspartate kinase and homoserine dehydrogenase
were analyzed with respect to feedback inhibition by lysine
and threonine. The activities of the enzymes lysine 2-oxoglutate reductase and saccharopine dehydrogenase, both
involved in lysine degradation in the maize endosperm were
also determined and shown to be reduced several fold with
the introduction of the o2, fl1 and fl2 mutations in the
Oh43+ inbred line, whereas wild-type activity levels were
verified in the Oh43o1 mutant.

Maize production is the highest of all crop plants and serves
as an important source of dietary protein for human and
livestock consumption. However, the nutritional quality is
not adequate, due to the lack of the essential amino acids
lysine and tryptophan in the seed proteins [1].
Zeins, which account for 50–70% of the endosperm
proteins in maize seeds, have a characteristic amino acid
composition, being rich in glutamine and hydrophobic
amino acids, whilst being very poor in lysine and tryptophan
[2]. Based on their solubility, genetic properties, and the

apparent molecular masses, zeins have been classified into
a- (22 and 19 kDa), the most abundant, b- (14 kDa), c- (27

and 16 kDa) and d-zein (10 kDa) [3].
Four main strategies have been attempted in order to
obtain plants with a high lysine seed content: plant breeding,
characterization of naturally occurring mutants, induction
of biochemical mutants and the production of transgenic
plants [4,5]. Perhaps the most exciting result obtained
during this research was the identification of the high-lysine
opaque 2 (o2) maize mutant [6]. Unfortunately, the highlysine trait was negatively correlated with other agronomic
characteristics, such as resistance to plant pathogens and
yield [1]. More recently, quality protein maize (QPM)
varieties have been produced which maintain the high-lysine
and high-tryptophan characteristics conditioned by the o2
mutation in a modified-vitreous endosperm, with favorable
agronomic characteristics [7–9].
The amino acid lysine is derived from aspartate and the
biosynthetic pathway involves the action of several strongly
regulated enzymes [10]. The enzyme aspartate kinase (AK;
EC 2.7.2.4), which converts aspartic acid into b-aspartyl
phosphate, can exist in at least two distinct isoforms, one (or
two) sensitive to lysine feedback inhibition and the other
sensitive to threonine feedback inhibition, the latter being a
bifunctional polypeptide with the threonine-sensitive homoserine dehydrogenase isoenzyme (HSDH; EC 1.1.1.3) [11].

´
Correspondence to R. A. Azevedo, Departamento de Genetica, Escola
Superior de Agricultura Luiz de Queiroz, Universidade de Sao Paulo,
˜
Piracicaba CEP 13418–900, SP, Brazil.
Fax: + 55 19 3433 6706, Tel.: + 55 19 3429 4475,
E-mail:

Abbreviations: AK, aspartate kinase; DHDPS, dihydrodipicolinate
synthase; HSDH, homoserine dehydrogenase; LOR, lysine 2-oxoglutarate reductase; N, nitrogen; NPN, nonprotein nitrogen; PVPP,
insoluble polyvynylpyrrolidone; SDH, saccharopine dehydrogenase;
SAA, soluble amino acids.
Enzymes: AK (EC 2.7.2.4); HSDH (EC 1.1.1.3); LOR (EC 1.5.1.8);
SDH (EC 1.5.1.9); DHDPS (EC 4.2.1.52).
(Received 2 September 2003, accepted 22 October 2003)

Keywords: lysine metabolism; maize; storage proteins.


Ó FEBS 2003

Lysine metabolism in maize mutants (Eur. J. Biochem. 270) 4899

The AK isoenzymes have been characterized at both the
biochemical and molecular level in several plant species
[4,5,10,12], and shown to be a major factor in the regulation
of the carbon flux through the aspartate pathway [4,10].
HSDH catalyses the conversion of aspartate semialdehyde
to homoserine in the presence of the coenzymes NADH or
NADPH and is present in plant species in two isoforms,
resistant and sensitive to threonine inhibition [10]. The
first enzyme unique to lysine synthesis, dihydrodipicolinate synthase (DHDPS; EC 4.2.1.52) has also been extensively studied and characterized in plants catalyzing the
condensation of pyruvate and aspartate semialdehyde
into dihydrodipicolinic acid [4]. DHDPS is also subject
to feedback inhibition by micromolar concentrations of
lysine [4].
Several mutants that overproduce and accumulate threonine have been obtained by selection on media containing
amino acids or their analogues and this phenomenon has

been shown to be due to alteration in the feedback pattern of
the lysine-sensitive AK isoenzyme [4]. However, in the case of
cereal seeds, the mutants failed to accumulate lysine in higher
concentration [10,13,14]. The development of plant transformation techniques has allowed the production of transgenic plants expressing the enzymes of lysine biosynthesis
that are insensitive to feedback regulation analogous to the
biochemical mutants. Again, most of the plants did not
exhibit significant accumulation of lysine in the seed [4,12].
Positive results were however, obtained with barley, canola
and soybean transgenic seeds in which dramatic increases in
the lysine content were observed [5,15].
Very little was known about lysine catabolism in plant
until recently [5,12,16]. The first two enzymatic steps are
catalyzed by the bifunctional protein lysine 2-oxoglutarate
reductase–saccharopine dehydrogenase (LOR–SDH; EC
1.5.1.8 and EC 1.5.1.9, respectively). LOR–SDH protein
has been studied in some plant species [17–21] where the
activity was particularly high in the endosperm tissue in
cereal crops [17,18]. The regulation of the LOR activity has
been shown to be complex, involving several distinct
mechanisms [5,12,16].
Recent studies have confirmed that in order to obtain
lysine overproduction in cereal seeds, manipulation of lysine
degradation is needed [5,12,16]. This suggestion is supported
by five main points [5]: (a) The cereal mutants or transgenic
plants do not exhibit significant accumulation of lysine in
the seeds; (b) LOR–SDH activities are endosperm specific in
cereal crops only; (c) LOR–SDH activities are drastically
reduced in the high-lysine o2 maize mutant as compared
with the wild-type; (d) lysine catabolism intermediates
accumulate in the seeds of lysine overproducing plants of

soybean and canola, indicating reduced LOR–SDH activities; and (e) LOR–SDH activities are lower in legume
plants and rice, which is the cereal crop with the highest
concentration of lysine in the seed.
The product of the o2 gene is specifically expressed in the
endosperm and the protein was shown to activate the
transcription of the 22 kDa a-zein [22] and 14 kDa b-zein
genes [23], together with the b-32 [24] and cyPPDK1 (one of
two cytosolic isoforms of pyruvate orthophosphate dikinase) genes [25]. Other possible direct or indirect target genes
of the o2 factor have been shown to belong to various
metabolic pathways [26–28]. In the o2 mutant, LOR–SDH

mRNA and protein quantities were severely reduced (about
90%), and the expression pattern during grain development
was markedly modified [29]. The genomic sequence of the
gene and its 5¢ regulatory regions revealed the presence of o2
boxes in the upstream promoter, confirming the hypothesis
of a transcriptional control of the Lor/Sdh gene by the o2
protein [16]. These large effects suggest that o2 protein may
play an important role in the developing grain, as a
coordinator of the expression of storage protein, and
nitrogen and carbon metabolism genes [30].
Although there is now plenty of information available
about o2, information related to lysine metabolism for
several other similar mutants that have been classified as
high-lysine and exhibit the opaque phenotype are very
scarce. A comprehensive investigation into these mutants
was initiated, with the aim of obtaining new insights into the
regulation of lysine metabolism in maize. During the course
of this work Hunter et al. [31] published an analysis of some
of these mutants. Our work now extends the studies of

Hunter et al. [31] and provides further insights into the
complex but critical regulation of lysine accumulation
within the maize seed, reporting for the first time the
biochemical characterization of these mutants using proteomic and enzymological approaches.

Experimental procedures
Maize mutants
Seeds of the mutant genotypes opaque (o1 and o2) and
floury (fl1 and fl2) and the respective wild type, Oh43 +,
were kindly provided by the Maize Genetics Cooperation
Seed Stock Center (Urbana, IL, USA). Plants of all
genotypes were grown in the glasshouse at ESALQ-USP,
Brazil and self-pollinated. Maize ears were harvested
20 days after pollination (DAP) directly into liquid nitrogen
and stored at ) 80 °C until used for enzyme extraction. The
experiments were repeated over three summer seasons
(1999–2000, 2000–01 and 2001–02).
Preparation of endosperm samples
Endosperms were isolated from mature grains previously
soaked in water for 30 min by peeling off the outer
tegument and excising the germ. After freeze drying, the
endosperms were ground to a powder using a ball mill.
Fractionation of nitrogen (N) constituents
The isolation of endosperm N constituents was undertaken
in duplicate as previously described by Landry et al. [32].
Quantitation of N constituents
For an accurate quantitation, the nonprotein nitrogen
(NPN) and protein content were determined by the
ninhydrin assay of a-amino N released after alkaline
digestion (3 M NaOH, 130 °C, 45 min) for the TCA, E1,2,

E4 extracts [32] or acid digestion (6 M HCl, 110 °C, 18 h) for
the E3 extract and residues, according to Landry et al. [33].
Soluble amino acids (SAA) were quantitated by ninhydrin
without previous digestion of the TCA extracts [32].


4900 R. A. Azevedo et al. (Eur. J. Biochem. 270)

Amino acids analysis
Soluble amino acids from mature seeds were extracted and
analyzed exactly as described by Gaziola et al. [9]. As the
OPA-lysine derivative is rapidly degraded, a second analysis
was performed using a 15-min elution time. Four replicates
were analyzed.
Protein extraction and two-dimensional polyacrylamide
gel electrophoresis of zein polypeptides
The procedure for two-dimensional polyacrylamide gel
electrophoresis (2D-PAGE) analysis of zein isoforms
followed that published previously by Consoli and Damerval [34]. Briefly, three sets of three mature kernels were
combined for each genotype and individually analyzed,
generating three independent replicates. Embryos and
pericarp were manually excised, and the endosperms were
crushed in liquid nitrogen for each genotype. The proteins
were resuspended in a urea-Triton X-100–2 ME buffer.
Isoelectric focusing was performed in 10-cm long rod gels in
a pH gradient ranging from 5.5 to 8.5. Approximately
40 mg of total proteins were loaded on each gel. The SDS
dimension was separated using a 14% acrylamide slab gel,
and staining was adapted from the colloidal Coomassie blue
method of Neuhoff et al. [35]. Images of the 2D patterns

were recorded and image analysis and spot detection were
carried out as described by Consoli and Damerval [34].
Specific zein protein extraction was previously used to
confirm the zein identity of the polypeptide spots visualized
in 2D gels [34].
Statistical analyses of zein isoform amounts
A previous analysis of colloidal Coomassie blue staining
intensity as a function of protein loading was carried out for
zein spots and it was demonstrated that for 86% of the
isoforms, a linear relationship was obtained [34]. Differences
in total zein amounts loaded onto the gels were compensated by scaling the raw integrated optical density of every
spot i in each gel j according to Consoli and Damerval [34].
One-way analysis of variance with the genotype as the
factor, were then performed for each spot on their scaled
integrated optical density, and a significant effect was
retained at P < 0.05.
Enzymes partial purification and assays
All procedures were carried out at 4 °C unless stated
otherwise. Four replicates each composed of five selected
maize ears, which were harvested (20 DAP), combined, and
mixed, were used for enzyme analysis.
For the extraction of AK, frozen seeds were extracted in
five volumes of buffer A [50 mM Tris/HCl, 200 mM KCl,
0.1 mM phenylmethanesulfonyl fluoride, 0.1 mM EDTA,
1 mM dithiothreitol, 2 mM L-lysine, 2 mM L-threonine, 10%
(v/v) glycerol and 5% (w/v) insoluble polyvynylpyrrolidone
(PVPP), pH 7.4]. The extract was filtered through three
layers of miracloth, and centrifuged at 16 000 g for 30 min
to remove the cellular debris. Solid ammonium sulfate was
added slowly to 30% saturation with gently stirring for at

least 30 min. The suspension was centrifuged at 16 000 g for

Ó FEBS 2003

30 min and the supernatant subjected to a second ammonium sulfate precipitation at 60% saturation for 30 min
with continuous stirring. Precipitated protein was recovered
by centrifugation at 16 000 g for 30 min and the protein
pellets were dissolved in a small volume of buffer B [25 mM
Tris/HCl, 1 mM dithiothreitol, 0.1 mM L-lysine, 0.1 mM
L-threonine and 10% (v/v) glycerol, pH 7.4]. The sample
was loaded onto a Sephadex G25 column (2.5 · 20 cm)
equilibrated with five column volumes of buffer B and run
under gravity. The desalted sample was collected and
assayed for AK activity.
AK activity was assayed routinely in a final volume of
500 mL as described by Brennecke et al. [28]. Controls
containing lysine and threonine were included to identify the
isoenzymes sensitive to lysine and threonine. Activity was
expressed as nmolỈmin)1Ỉmg)1 protein. Four replications
were carried out for each assay.
For the extraction of HSDH, frozen seeds were extracted
in five volumes of buffer C [50 mM Tris/HCl, 200 mM KCl,
0.1 mM phenylmethanesulfonyl fluoride, 1 mM EDTA,
3 mM dithiothreitol, 5 mM L-threonine, 10% (v/v) glycerol
and 5% (w/v) PVPP, pH 7.5]. The extract was filtered
through three layers of miracloth, and centrifuged at
16 000 g for 30 min to remove completely cellular debris
from the extract. Solid ammonium sulfate was added slowly
to 30% saturation with gently stirring for at least 30 min.
The suspension was centrifuged at 16 000 g for 30 min and

the supernatant subjected to a second ammonium sulfate
precipitation at 60% saturation for 30 min with continuous
stirring. Precipitated protein was recovered by centrifugation at 16 000 g for 30 min and the protein pellets were
dissolved in a small volume of buffer D [25 mM Tris/HCl,
1 mM EDTA, 1 mM dithiothreitol, 0.1 mM L-threonine and
10% (v/v) glycerol, pH 7.5]. The sample was loaded onto a
Sephadex G25 column (2.5 · 20 cm) equilibrated with five
column volumes of buffer D and run under gravity. The
desalted sample was collected and assayed for HSDH
activity.
HSDH activity was assayed routinely spectrophotometrically at 340 nm in a final volume of 1.1 mL at 30 °C as
described by Azevedo et al. [11]. The effect of threonine on
the HSDH activity was determined by the addition (10 mL
of a 1 M solution) of the amino acid to the assay mixture.
Activity was expressed as nmolỈmin)1Ỉmg)1 protein. Four
replications were carried out for each assay.
For the extraction of LOR–SDH, frozen seeds were
extracted in five volumes of buffer E [100 mM potassium
phosphate, 50 mM KCl, 1 mM EDTA, 1 mM dithiothreitol,
0.1 mM phenylmethanesulfonyl fluoride, 10% (w/v) glycerol
and 5% (w/v) PVPP, pH 7.0]. The homogenate was first
filtered through three layers of miracloth and then centrifuged at 15 000 g for 30 min to remove cellular debris. The
supernatant was adjusted to 30% ammonium sulfate
saturation by gently stirring for at least 30 min. The
suspension was centrifuged at 15 000 g for 30 min and
the supernatant subjected to a second ammonium sulfate
precipitation at 55% saturation for 30 min with continuous
stirring. After centrifugation at 15 000 g for 30 min, the
sedimented proteins were dissolved in 10 mL of buffer E
(minus phenylmethanesulfonyl fluoride and PVPP). The

sample was then loaded onto a Sephadex G50 column
(2.6 · 20 cm) previously equilibrated with buffer F


Ó FEBS 2003

Lysine metabolism in maize mutants (Eur. J. Biochem. 270) 4901

[100 mM Tris/HCl, 1 mM dithiothreitol, 1 mM EDTA and
10% (v/v) glycerol, pH 7.4] and run under gravity. The
desalted sample was collected and assayed for LOR and
SDH activities.
LOR activity was routinely assayed spectrophotometrically in a 1 mL cuvette at 30 °C by following the change in
absorbance at 340 nm over a 15-min period, with appropriate adjustments for a lysine-free blank as described by
Gaziola et al. [18]. Activity was expressed as nmol NADPH
oxidizedỈmin)1Ỉmg)1 protein. Four replications were carried
out for each assay.
SDH activity was measured spectrophometrically in a
1 mL cuvette by following the rate of substrate-dependent
reduction of NAD+ to NADH monitored at 340 nm at
30 °C over a 15-min period, with appropriate adjustments
for a saccharopine-free blank as described by Gaziola
et al. [18]. Activity was expressed as nmol NAD+ reducedỈ
min)1Ỉmg)1 protein. Four replications were carried out for
each assay.
Protein determination
Protein concentrations of the samples were determined as
described by Bradford [36] using bovine serum albumin as
a standard.

Results
Distribution patterns of N constituents
Table 1 provides data concerning the percentage contribution of the main N constituents present in opaque (o),
floury (fl) and wild-type (+) endosperms. The amounts of
SAA and NPN were of the same magnitude for the wildtype inbred line and all mutants. The albumins and
globulins of the mutants exhibited variable amounts,
ranging from a value similar to that of the wild-type
inbred line (Oh43fl1), to a value 3.5-fold higher (Oh43o2).
The same mutant genotypes also marked the boundaries
of variation of zein for the mutants, with the Oh43o2
endosperm being the poorest in zeins, whereas Oh43fl1
exhibited the highest amounts of zeins, but still lower than
that of the wild-type Oh43+. In general, the mutants had
protein distribution patterns varying between that of
Oh43fl1, similar to that of the wild type, and that of

Oh43o2. From these results it was possible to assess the
importance of lysine-rich nonzeins with accuracy, because
of the quantitation of nonprotein N and the exhaustive
extraction of zeins. Thus, the ratio of the nonzein content
of the mutants compared with that of Oh43+ varied from
1.2 to 1.5 for most mutants, whereas for Oh43o2, a ratio
of 2.6 was calculated.
Soluble lysine concentration
The Oh43o2 mutant exhibited the highest relative concentration of soluble lysine followed by the Oh43fl2 mutant,
whereas the Oh43o1 mutant exhibited the lowest relative
concentration of soluble lysine, but still higher than that of
the wild-type counterpart (Table 1). The Oh43o2 mutant
also exhibited the highest absolute concentration of soluble
lysine (7.35 nmolỈmg)1 dry weight) followed by the Oh43fl2

mutant (5.29 nmolỈmg)1 dry weight), whereas the Oh43o1
mutant exhibited the lowest absolute concentration of
soluble lysine (1.65 nmolỈmg)1 dry weight), but still higher
than that of the wild-type counterpart (0.82 nmolỈmg)1 dry
weight). However, the total SAA pool also varied among
the genotypes, indicating clear differences between the
mutations. The total SAA pool was increased slightly
following the introduction of the mutations o2, fl1 and fl2,
but was reduced by 20% by the o1 mutation (Table 1),
however, the relative soluble lysine concentrations were
increased considerably following the introduction of each
mutation (Table 1).
2D-page
Thirty-five zein polypeptides were detected in wild type
Oh43+ and Oh43o1, Oh43o2, Oh43fl1 and Oh43fl2
mutants. Four polypeptides were identified as c27 kDa
zeins, 10 as a22 kDa zeins, 15 as a19 kDa zeins, two as
b14 kDa zeins, two as c16 kDa zeins and two as d10 kDa
zeins, according to their apparent molecular masses in the
SDS dimension (Fig. 1).
Between 20 and 31 zein isoforms were detected according
to the genotype (Table 2). The mutations decreased the
number of zein isoforms detected on the 2D gels as
compared with the wild-type, indicating a decrease in zein
amount and diversity (Fig. 2). The effect of each mutation
on the amount of every isoform was tested using analyses of

Table 1. Quantitation of N constituents. Percentage contribution of N constituents present in opaque (o), floury (fl) and wild-type (Oh43+)
endosperms. Data expressed as percentage (± standard deviation) of endosperm total N. N constituents: SAA, soluble amino acids; NPN,
nonprotein N; A + G, albumins + globulins corresponding to E1,2 – NPN; non- zeins corresponding to glutelins (Glu) + albumins + globulins;

P, endosperm total proteins expressed as percentage of dry matter. Soluble lysine is expressed as percentage of total soluble amino acids pool
(± standard deviation) followed by the percentage increase in soluble lysine in relation to the wild type.
N constituents
Genotypes

SAA

Oh43+
Oh43o1
Oh43o2
Oh43fl1
Oh43fl2

0.75
0.60
0.95
0.82
0.94

NPN
(0.07)
(0.07)
(0.00)
(0.07)
(0.14)

A+G

Zeins

1.6
2.4
2.9
1.9
2.7

3.2
4.0
11.1
3.7
6.7

77.4
68.6
41.5
71.9
64.8

17.8
25.0
44.5
22.5
25.8

Non-zeins

Glu

(0.28)
(0.20)

(0.35)
(0.28)
(0.35)

(1.8)
(0.4)
(1.2)
(0.8)
(0.2)

(1.0)
(0.7)
(1.6)
(1.1)
(0.6)

P% DM

Lysine

% increase

21.0
29.0
55.6
26.2
32.5

10.8
11.6

8.7
12.8
11.2

0.33
0.53
2.70
0.83
1.46

–
61
718
151
342

(0.6)
(0.5)
(1.2)
(0.8)
(0.2)

(0.02)
(0.02)
(0.07)
(0.02)
(0.10)


4902 R. A. Azevedo et al. (Eur. J. Biochem. 270)

Ó FEBS 2003

Fig. 1. Two-dimensional separation of zein
isoforms isolated from the endosperms of maize
seeds in the Oh43 background, using isoelectric
focusing and SDS/PAGE. Wild-type Oh43+
key isoforms are indicated with black arrows.
The white arrow points to an isoform
appearing specifically in the fl2 mutant.
Molecular masses and pH range are indicated
along the gel.

variance on standardized spot volumes (see Methods). The
o2 and fl2 mutations had large effects, as about 60% of the
isoforms differed in amount as compared with the wild type.
Conversely, Oh43o1 and Oh43fl1 mutants exhibited zein
patterns and contents similar to those of their wild-type
counterpart, in agreement with the data of Table 1.
The pattern of c27 kDa isoforms was the most strongly
affected by the fl2 mutation, as all the polypeptides
disappeared, in contrast, fl1 had little effect. Among the
10 a22 kDa zein class isoforms, only one appeared in a
similar amount in all of the mutants and wild type (a22z2,
Fig. 1, 2). The mutations generally decreased the amount of
the isoforms as compared with the wild type. Among the 15
a19 kDa zein class isoforms, five were unaffected in the
mutants. The Oh43fl2 mutant exhibited the strongest effect
on this zein class, altering the amount of nine isoforms,
amongst which, two occurred specifically in this mutant (e.g.

a19z114, Fig. 1, 2). The mutations fl2 and o2 had a parallel
effect on b14 kDa and d10 kDa zein isoforms, but the effect
of fl2 was less pronounced than that of o2. In all, the o2
mutation markedly altered the pattern of low molecular
mass zeins, as compared with the wild type.
Lysine metabolism
In this study, the enzymes AK, HSDH, LOR and SDH
were extracted initially from the developing seeds (16, 20
and 24 DAP) of the wild type, which indicated that the
main peak of activity of AK (4.32, 7.91 and
3.10 nmolỈmin)1Ỉmg)1 protein at 16, 20 and 24 DAP, respectively), HSDH (5.24, 16.31 and 6.16 nmolỈmin)1Ỉmg)1
protein at 16, 20 and 24 DAP, respectively), LOR (2.05,

3.53 and 2.18 nmolỈmin)1Ỉmg)1 protein at 16, 20 and 24
DAP, respectively) and SDH (2.37, 3.51 and 2.08 nmolỈ
min)1Ỉmg)1 protein at 16, 20 and 24 DAP, respectively) was
at 20 DAP. The activities of the enzymes involved in lysine
metabolism have been studied in maize endosperm, exhibiting a peak of activity between 16 and 24 DAP depending
on the enzyme [9,28]. In this study, the activities of the
enzymes AK, HSDH, LOR and SDH were determined in
extracts isolated from the wild-type 16, 20 and 24 DAP and
the maximum activity for all enzymes was confirmed as 20
DAP. Based on this peak of enzyme activity, all genotypes
were subsequently analyzed at 20 DAP.
The activity of AK varied considerably among all
genotypes, ranging from 2.78 nmolỈmin)1Ỉmg)1 protein in
Oh43fl1)15.29 nmolỈmin)1Ỉmg)1 protein in the Oh43fl2
(Table 3). The Oh43o1 and Oh43fl2 mutants exhibited
higher activities (12 and 85%, respectively) when compared
with the wild type, whereas the mutants Oh43o2 and

Oh43fl1 exhibited lower activities (40 and 66%, respectively). The inhibition by lysine was shown to be reduced in
all four mutants when compared with the wild type (60.8%
inhibition), ranging from 28.9% in the Oh43fl1 mutant to
55.6% in the Oh43o1 mutant. The inhibitory effect of
threonine on AK activity was much lower when compared
with the effect of lysine, resulting in 11.1% inhibition of AK
activity in the wild type, 15.6% in the Oh43fl2 mutant and
3.34% in the Oh43fl1 mutant, whereas a slight stimulation
of AK activity was induced by threonine in the Oh43o2
mutant. When both amino acids were added together, a
more intense inhibitory effect was observed, with the wildtype Oh43+ exhibiting the highest inhibitory effect (92.5%)
and the Oh43fl1 the lowest (41.8%) (Table 3).


Ó FEBS 2003

Lysine metabolism in maize mutants (Eur. J. Biochem. 270) 4903

Table 2. Two dimensional separation of zeins isolated from maize endosperms. Thirty-five zein isoforms were revealed. Mean values of spot volumes
are indicated. Each isoform number is prefixed by the name of the zein class. Statistical analyses were performed to test for significant differences in
isoform amounts, and genotypes sharing a same letter did not differ significantly (a indicates an amount significantly greater than b).
Spot name

WT

o1

o2

c27z45

c27z38
c27z28
c27z21
a22z4
a22z37
a22z31
a22z3
a22z2
a22z18
a22z12
a22z115
a22z11
a22z1
a19z93
a19z9
a19z8
a19z7
a19z6
a19z5
a19z43
a19z30
a19z23
a19z20
a19z19
a19z17
a19z15
a19z114
a19z10
c16z44
c16z13

b14z33
b14z14
d10z16
d10z109

2262.89(a)
8423.21(a)
7009.92(a)
29520.46(a)
51353.85(a)
4126.22(a)
7140.23(a)
26824.27(a)
26413.60(a)
9363.84(a)
18884.57(a)
0
23365.92(a)
81282.15(a)
0
51483.41(a)
25934.51(c)
65109.57(a)
65917.49(a)
92886.09(a)
2339.87(a)
7646.58(b)
10831.41(b)
16394.55(a)
11402.26(b)

19406.44(a)
12065.24(a)
0
45064.00(a)
4072.74(a)
31426.21(b)
7480.22(a)
71467.31(ab)
43782.08(bc)
0

688.72(a)
0
8712.80(a)
10278.86(b)
50168.37(a)
0
4266.46(a)
24194.50(a)
32646.71(a)
6607.25(a)
19780.78(a)
0
18853.00(a)
73752.64(ab)
0
43735.68(a)
18027.17(c)
50561.34(a)
78276.12(a)

99287.19(a)
0
0
22279.91(a)
19023.76(a)
0
20261.95(a)
4832.44(a)
0
43462.53(a)
2180.72(a)
37841.49(b)
7535.85(a)
83184.64(a)
52390.91(bc)
3303.82(b)

0
0
8342.94(a)
0
41558.61(a)
0
0
4797.66(b)
26387.42(a)
0
13033.15(a)
0
0

44042.71(b)
0
50534.32(a)
72930.41(a)
70261.18(a)
79297.66(a)
79080.27(a)
0
17683.49(a)
8024.70(b)
2706.90(b)
0
10550.99(b)
0
0
51245.73(a)
0
118849.72(a)
0
39052.79(c)
109105.04(a)
20394.88(a)

The activity of HSDH varied considerably among all
genotypes, ranging from 17.4 nmolỈmin)1Ỉmg)1 protein
in Oh43+ to 38.4 nmolỈmin)1Ỉmg)1 protein in Oh43o2
(Table 3). All mutants exhibited higher activities when
compared with the wild-type, with the Oh43o2 mutant
exhibiting a 2.2-fold higher HSDH activity. The effect of
threonine was tested on HSDH activity, exhibiting an

inhibitory effect in the wild-type and Oh43o2, Oh43fl1 and
Oh43fl2 mutants, but stimulating HSDH activity in the
Oh43o1 mutant (Table 3).
Table 3 shows the activities of the enzymes LOR and
SDH, both involved in lysine degradation, which were also
measured for all genotypes. Large variations were observed
for LOR activity, varying from 0.49 nmol NADPH oxidized min)1Ỉmg)1 protein in Oh43fl2)3.83 nmol NADPH
oxidized min)1Ỉmg)1 protein in Oh43o1, which was even
higher than the activity in the wild type (3.50 nmol NADPH
oxidized min)1Ỉmg)1 protein) (Table 3). The Oh43o2
mutant exhibited a sixfold reduction of LOR activity, a
reduction that was even higher (7.1-fold) in the Oh43fl2

fl1
0
25013.17(a)
8406.33(a)
16517.54(ab)
60342.63(a)
0
5461.84(a)
24681.31(a)
32919.48(a)
7301.23(a)
25436.17(a)
0
23972.68(a)
101923.68(a)
0
42954.38(a)

33846.97(bc)
50230.13(a)
75690.34(a)
63968.14(a)
0
6122.26(b)
8822.52(b)
12926.97(a)
0
19055.38(ab)
0
0
51198.85(a)
0
44790.63(b)
6828.42(a)
73327.86(ab)
27284.15(c)
0

fl2
0
0
0
0
0
0
0
25758.31(a)
28709.67(a)

0
0
31431.73(a)
19571.06(a)
88989.89(a)
47289.65(a)
52618.48(a)
47679.73(b)
46370.74(a)
62619.74(a)
61067.52(a)
0
0
27136.28(a)
15690.45(a)
54852.13(a)
18178.69(ab)
0
22304.39(a)
46990.69(a)
0
23034.16(b)
0
51025.04(bc)
75793.39(ab)
0

mutant. Reduction of LOR activity was also observed in the
Oh43fl1 mutant (40% lower) when compared with the wild
type, whereas in the Oh43o1 mutant the activity was slightly

higher than the wild type. Similar reductions in SDH
activity along with LOR, were also induced by the o2, fl1
and fl2 mutations and thus the LOR/SDH ratio did not
exhibit major variations.

Discussion
The opaque and floury mutations and their respective wildtype (Oh43+) were obtained from the Maize Genetics
Cooperation Seed Stock Center (USA) and cultivated in
Brazil for three successive summer seasons. Very little
variation among the genotypes was observed for time of
flowering indicating a similar developmental behavior,
which would be expected as all mutants are in the same
genetic background. This also allowed the self-pollination
and production of seeds for all genotypes. The content of
the various N constituents in the endosperm is dependent on


Ó FEBS 2003

4904 R. A. Azevedo et al. (Eur. J. Biochem. 270)

Fig. 2. Mutants Oh43o1, Oh43o2, Oh43fl1
and Oh43fl2. The arrows point to the isoforms
indicated on the wild-type gel (Fig. 1).

Table 3. Determination of activity of enzymes involved in lysine metabolism. AK specific activity (nmolỈmin)1Ỉmg protein)1), HSDH specific activity
(nmolỈmin)1Ỉmg protein)1), LOR specific activity (nmol NADPH oxidizedỈmin)1Ỉmg protein)1) and SDH specific activity (nmol NAD +
reducedỈmin)1Ỉmg protein)1) were determined in extracts of 20 DAP maize endosperms and following the addition of lysine (L) and/or threonine
(T). Standard deviation (SD) values were all below 5% for the L, T and LT treatments.
Genotypes

Oh43 +

Enzyme
AK
Control (SD)
% inhibition by +
% inhibition by +
% inhibition by +
HSDH
Control (SD)
% inhibition by +
LOR (SD)
SDH (SD)
LOR/SDH ratio
a

5 mM L
5 mM T
5 mM LT

5 mM T

Oh43o1

Oh43o2

Oh43fl1

Oh43fl2

8.282 (0.314)
60.8
11.1
92.5

9.240 (0.371)
55.6
13.5
78.3

4.956 (0.121)
42.2
+ 4.73a
83.2

2.783 (0.120)
28.9
3.34
41.8

15.290 (0.414)
37.5
15.6
49.5

17.4 (0.71)
31.0
3.505 (0.121)
3.490 (0124)
1.00

27.6 (0.88)
+ 26.1
3.830 (0.133)
3.050 (0.090)
1.25

38.4 (1.43)
25.0
0.590 (0.016)
0.705 (0.071)
0.83

19.8 (0.57)
6.1
2.115 (0.030)
1.870 (0.097)
1.13

21.0 (0.57)
17.1
0.490 (0.013)
0.890 (0.033)
0.55

Indicates activation of enzyme activity.

genetic and environmental factors. With the view of
dissociating these two factors, the present results were
compared with data taken from the literature and concerning the same genotypes, but cultivated at diverse locations:

Bergamo, Italy [37]; Orsay, France [32]; LaFayette, USA
[38]; and Tucson, USA [31]. Furthermore, for a better
comparison, the genotypes were ranked according to an
increasing content of zeins (Table 4): (a) Zein percentages


Ó FEBS 2003

Lysine metabolism in maize mutants (Eur. J. Biochem. 270) 4905

Table 4. Zein and lysine determinations in distinct studies. Percentage
of zein and protein lysine in maize seeds. References: PS: present study;
Misra et al. [38]; Di Fonzo et al. [37]; Landry et al. [32]; Hunter et al.
[31]. Lysine percentage true proteins (estimated), values in parentheses
correspond to lysine percentage crude proteins (assayed).
Genotype

Zeins

Lysine percentage

[Reference]

W64Ao2
Oh43o2
Oh43o2
Oh43o2
W64Afl2
Oh43fl2

W64Ao1
W64+
W22+
Oh43fl2
Oh43o1
W22+
Oh43fl1
W22+
Oh43+
Oh43+

28.7
41.5
47.1
49.3
50.0
64.8
66.4
67.8
65.5
65.7
68.6
68.7
71.9
74.3
77.4
77.5

3.8
3.78

3.14
(3.5)
2.8
2.34
1.7
1.5
–
(2.3)
2.08
(2.3)
1.87
1.68
1.49
(1.6)

31
PS
32
38
31
PS
31
31
37
38
PS
38
PS
32
32

38

ranged from 28.7% (W64Ao2) to 77.5% (Oh43+). The
difference between the minimum and maximum percentages
was almost the same as that found by Balconi et al. [39]
between Illinois low protein (40%) and Illinois high protein
(74.5%) genotypes, taking into account that these values are
slightly (5%) underestimated as E4 proteins were excluded
from zeins by the authors. (b) The effect of environmental
conditions upon the content of zeins for a given genotype
can result in a discrepancy of 8–9% in the case of Oh43o2,
W22o2 and W22 + or be negligible in the case of Oh43fl2
or Oh43+. (c) For a given mutant gene the genetic
background can have a considerable impact upon the zein
content, however, this is not always the case as can be seen
with W64Ao1 and Oh43o1. (d) More generally, the gradual
increase in zein content would indicate a progressive change
in the relative proportions of soft and hard endosperms,
respectively, poor and rich in zeins. Therefore, the effect of
one gene upon the distribution pattern of protein fractions
cannot be generalized from that found for only one genetic
background.
The opaque and floury mutants used in this study have
been classified as high-lysine endosperm mutants, however,
such higher concentrations of lysine can be due to alterations in the storage protein fractions and/or in the concentration of soluble lysine in the endosperm. In previous
studies, the soluble lysine concentration has been shown to
be increased in the o2 maize mutant when compared with
the wild-type maize [9,30,31]. Estimating the percentage of
lysine in true proteins by assuming the lysine content of
nonzeins is independent of genotype and equal to 7 g per

100 g of proteins, and based on the distribution of the
endosperm proteins, the mutants exhibited higher
concentrations of total lysine when compared with their
wild-type counterpart. We have also observed a significant
variability in the absolute and relative soluble lysine
concentrations among the mutants analyzed. The o2
mutation led to an increase in the total SAA pool and in

the soluble lysine concentration in the endosperm, confirming the previous reports for this mutant [2,13,30,31],
although such increases may vary depending on the genetic
background to which the gene is introduced [30,31]. In the
other mutants, distinct responses were observed in relation
to lysine concentration, showing that the mutants Oh43fl1
and Oh43fl2, exhibited increases in total SAA and soluble
lysine concentration, in a similar way to the Oh43o2 mutant,
leading to higher lysine concentrations in the endosperm,
but not to the same extent. However the Oh43o1 mutant,
exhibited a lower concentration of total SAA, but an
increased concentration of relative and absolute contents of
soluble lysine, which on balance indicates that the Oh43o1
mutant has a small significant increase (101%) in soluble
lysine. The results observed for the o1 mutation are similar
to that reported by Hunter et al. [31], who observed an
amino acid composition similar to the wild-type counterpart. On the other hand, Balconi et al. [39] reported an
increased concentration of total lysine in the o1 mutant to
the same extent as that for the o2 mutant. In general, all
mutants can be classified as high-lysine mutants, but the
increases in lysine observed were not as great as that
observed for the o2 mutation.
Hunter et al. [31] used one dimensional SDS/PAGE to

compare qualitative and quantitative differences in zein
patterns among a range of opaque mutants. Except for o2,
little effect of the mutations was observed. The analysis was
refined by immunoblotting with specific antisera, which
demonstrated that in o2 there was a decreased amount
a22 kDa, b14 kDa and d10 kDa isoforms, whereas in fl2
the a22 kDa zeins were reduced. Using 2D electrophoresis,
we were able to observe complex patterns of alterations in
the mutants as compared with the wild type. The various
isoforms detected are not due to artifacts during protein
isolation and/or fractionation, but to genetic differences in
charge and amino acid content [40]. A given mutation can
increase or decrease the relative amount of different
isoforms belonging to the same class of zein, indicating
very specific effects. In the Oh43 background fl1 and o1
mutations had very little effect. A similar low effect was also
observed for o1 in the W64A background [31]. The
mutations o2 and fl2 had their largest effects on the
a22 kDa and c27 kDa zeins, mostly decreasing the amount
of the isoforms present. The effect of o2 on b14 kDa
isoforms was consistent with a regulatory role of this
transcriptional activator on these zein genes [23]. In contrast
to Hunter et al. [31], we found that o2 increased rather than
decreased the relative amount of the d10 kDa isoforms.
This may be due to a specific effect of the background, as we
used Oh43 while Hunter et al. [31] used W64A. A large
background effect on the range of o2 effects had already
been observed (e.g. [34]).
The enzymes of lysine metabolism have been studied and
characterized in several plant species [10]. As wild-type

maize and the o2 mutant were the only sources of
information in the literature as far as lysine metabolism is
concerned, we have used them as controls for our analysis of
the other mutants. The data in Table 3 provide evidence
that there is a wide variation in terms of AK activity, with
several-fold variation in AK activity among the genotypes
studied, which, in the case of the low rates in the Oh43o2
mutant, agreed with previous results published by other


4906 R. A. Azevedo et al. (Eur. J. Biochem. 270)

authors [9,28]. Two mutants, Oh43o1 and Oh43fl2 exhibited
increases in AK activity, whereas the mutants Oh43o2
and Oh43fl1 exhibited a reduction in AK activity when
compared with their wild type, Oh43+. AK activity has
been shown to be determined by the action of at least two
separate isoenzymes, one that is sensitive to lysine inhibition
and the other sensitive to threonine inhibition [10]. Furthermore, in higher plants, the lysine-sensitive isoenzyme
normally accounts for 50–80% of the total AK activity,
with the exception of AK activity in coix endosperm, in
which the threonine-sensitive isoenzyme predominates [41].
Independent of the mutation, in the Oh43 genetic background, lysine produced the stronger inhibition of AK
activity, suggesting that the lysine-sensitive isoenzyme is
predominant in this genetic background and that such
distribution of isoenzymes activities is not affected by any of
the introduced mutations.
Threonine inhibition of AK was low in all the lines, but
there was evidence of a further reduction, caused by the o2
and fl1 mutations. However, lysine inhibition was reduced

in all the mutants when compared with the wild-type,
particularly in fl1 and fl2. The presence of AK activity more
insensitive to lysine and threonine inhibition was confirmed
when both amino acids were tested together, resulting in less
than 50% inhibition of the total AK activity in fl1 and fl2,
with lesser reductions being detected in the opaque mutants.
Stimulation of HSDH activity by threonine was observed
for the Oh43o1 mutant, however, no major effects on AK
activity were observed in this mutant, which might indicate
a specific effect of the o1 gene on the HSDH domain of the
bifunctional polypeptide. Apart from these results, HSDH
activity does not appear to be under any particular influence
from the mutations analyzed. All the genotypes tested
exhibited variations for threonine inhibition, suggesting the
presence of both HSDH isoenzymes. It has been suggested
that HSDH does not have a regulatory role in the
biosynthesis of lysine, although this enzyme shares the
same substrate (aspartate semialdehyde) with DHDPS,
which could eventually be a key point in determining
the flux of carbon through the pathway, leading to
threonine or lysine biosynthesis [10]. Although a recent
study using transgenic Arabidopsis thaliana expressing
bacterial DHDPS and having knockout mutation for lysine
catabolism produced high increases in soluble lysine and
methionine [42], no evidence of an increase in soluble
methionine was detected in the opaque and floury mutants
analyzed in this work (data not shown).
Evidence has been obtained from biochemical and
molecular analyses that AK activity is possibly regulated
by the o2 gene [13], intensifying its effect on the total pool of

SAA and free threonine accumulation in maize endosperm
[13]. Moreover, one of the genes encoding a lysine-sensitive
AK isoenzyme was linked to the o2 gene in chromosome 7
[13]. Wang et al. [43] also observed that AK activity varied
in its sensitivity to lysine inhibition, even between distinct
lines in which the o2 was introduced. Furthermore, several
quantitative trait loci for SAA content have been identified,
one of them linked to another AK-HSDH encoding gene
[43]. The analysis of o2 mutants has indicated that the
lysine-sensitive AK isoenzyme, but not the bifunctional
threonine-sensitive AK-HSDH isoenzyme, is affected by the
mutation [43].

Ó FEBS 2003

The enzymes of lysine catabolism, LOR and SDH, were
also analyzed in all genotypes and exhibited significant
alteration in activity depending on the mutant. LOR and
SDH were initially identified as one bifunctional enzyme
containing both enzyme domains [12,17,18], whilst later
monofunctional LOR and SDH enzymes were identified
[12]. The results reported in the literature generally indicated
that SDH activity is more stable than LOR activity [5]. The
dramatic sixfold reduction of LOR activity in the mutant
Oh43o2 also confirmed the results observed for the effect of
the o2 gene on LOR activity [9], which is due to a decreased
mRNA and enzyme protein synthesis [29]. SDH activity
was also influenced by the o2 gene, exhibiting a 4.9-fold
decrease in enzyme activity, which is a greater reduction
when compared with previous work with this mutant [9].

The reduction in LOR and SDH activities observed for the
Oh43o2 mutant was also observed in Oh43fl1 and Oh43fl2
mutants, with the latter exhibiting a LOR activity even
lower that of the Oh43o2 mutant (7.1-fold).
Our results suggest that the catabolism of lysine catalyzed
by the enzyme LOR, may be under the regulation of the
opaque and floury mutations. This is in addition to the
biosynthetic enzymes AK and to a lesser extent HSDH
discussed previously. The way this pleiotropic regulation
can take place may be different according to the mutation. It
has been shown in previous studies in which the LOR and
SDH enzymes were isolated and characterized, that the
LOR has an essential role in the regulation of lysine
catabolism, as this enzyme is modulated by Ca2+, ionic
strength and protein phosphorylation/dephosphorylation in
several plant species [19,29,44,45], however, such modulation effects do not appear to influence SDH activity.
Pleiotropic regulation is also supported by the effect of the
mutations on the storage proteins analyzed by 2D-PAGE.
In parallel with their considerable effect on LOR and SDH
activity, both o2 and fl2 induced large alterations in the
synthesis pattern of a22 kDa and c27 kDa zeins. Furthermore, the Oh43o2 and Oh43fl2 mutants also exhibited
higher concentrations of soluble lysine in the endosperm,
not only based on its concentration, but accompanying the
effect of each mutation on the concentration of the total
pool of SAA.
Curiously, the Oh43o1 mutant, which has been classified as high-lysine, did not exhibit major effects on the
catabolism of lysine in the endosperm, which suggests that
the high lysine concentration cannot be explained by an
altered lysine catabolism in this mutant. Although there
was a slight increase in the concentration of soluble lysine

in the Oh43o1 mutant when compared with the other
mutants, the lysine degradation enzyme pattern as well as
the AK activity, was shown to be at the same level of the
wild-type counterpart. Furthermore, Hunter et al. [31]
who also analyzed this mutant, but in a different genetic
background, could not find any important effect of this
mutation. The mutants Oh43o1 and Oh43fl1 exhibited
little effect either on the zein polypeptides or on LOR and
SDH activities.
The analysis of other mutations in the same phenotypic
class as the o2 gene indicates that the mutations may
strongly influence lysine metabolism and storage protein
synthesis and accumulation in maize. Many of the zein
polypeptides have been shown to vary in these mutants and


Ó FEBS 2003

Lysine metabolism in maize mutants (Eur. J. Biochem. 270) 4907

a new range of studies must be carried out to determine the
precise molecular regulation of the synthesis of these
polypeptides by such mutations. It is also clear that future
studies on the effect of these mutations should also be
carried out on the activity of the DHDPS enzyme, which has
been shown to be a key regulatory step in lysine biosynthesis
[5,10], but has only been tested in the o2 mutant so far [43].

Acknowledgements
This work was financed by grants to RAA from Fundacao de Amparo

¸ ˜
`
a Pesquisa do Estado de Sao Paulo, Brazil (FAPESP 98/12461–0 and
˜
01/13904–8) and the British Council (RAA and PJL). The authors also
wish to thank the Conselho Nacional de Desenvolvimento Cientı´ fico e
´
Tecnologico (CNPq, Brazil) and FAPESP for the scholarships and
fellowships received, Professor L. Sodek (UNICAMP) for the critical
reading of the manuscript, J. Carmezzini for the growth of the
mutants, A. Karime, M. Garcia and F. Mestrinelli for technical
assistance.

References
1. Vasal, S.K. (1994) High quality protein corn. In: Specialty Corns
(Hallauer, A.R., ed.), pp. 79–121. CRC Press, Boca Raton, FL.
2. Sodek, L. & Wilson, C.M. (1971) Amino acid composition of
proteins isolated from normal, opaque-2, and floury-2 corn
endosperms by a modified Osborne procedure. J. Agric. Food
Chem. 19, 1144–1149.
3. Wilson, C.M. (1991) Multiple zeins from maize endosperms
characterized by reverse- phase high performance liquid chromatography. Plant Physiol. 95, 777–786.
4. Azevedo, R.A. & Lea, P.J. (2001) Lysine metabolism in higher
plants. Amino Acids 20, 261–279.
5. Azevedo, R.A. (2002) Analysis of the aspartic acid metabolic
pathway using mutant genes. Amino Acids 22, 217–230.
6. Mertz, E.T., Bates, L.S. & Nelson, O.E. (1964) Mutant gene that
changes protein composition and increase lysine content of maize
endosperm. Science 145, 279–280.
7. Pixley, K.V. & Bjarnason, M.S. (1993) Combining ability for yield

and protein quality among modified-endosperm opaque-2 tropical
maize inbreds. Crop Sci. 33, 1229–1234.
8. Moro, G.L., Lopes, M.A., Habben, J.E., Hamaker, B.R. &
Larkins, B.A. (1995) Phenotypic effects of opaque2 modifier genes
in normal maize endosperm. Cereal Chem. 72, 94–99.
9. Gaziola, S.A., Alessi, E.S., Guimaraes, P.E.O., Damerval, C. &
˜
Azevedo, R.A. (1999) Quality protein maize: a biochemical study
of enzymes involved in lysine metabolism. J. Agric. Food Chem.
47, 1268–1275.
10. Azevedo, R.A., Arruda, P., Turner, W.L. & Lea, P.J. (1997) The
biosynthesis and metabolism of the aspartate derived amino acids
in higher plants. Phytochemistry 46, 395–419.
11. Azevedo, R.A., Smith, R.J. & Lea, P.J. (1992) Aspects of aspartate
kinase regulation in maize: co-purification of aspartate kinase and
homoserine dehydrogenase sensitive to threonine. Phytochemistry
31, 3731–3734.
12. Galili, G. (2002) New insights into the regulation and functional
significance of lysine metabolism in plants. Annu. Rev. Plant Biol.
53, 27–43.
13. Azevedo, R.A., Arana, J.L. & Arruda, P. (1990) Biochemical
genetics of the interaction of the lysine plus threonine resistant
mutant Ltr*19 with opaque-2 maize mutant. Plant Sci. 70, 81–90.
14. Muehlbauer, G.J., Gengenbach, B.G., Somers, D.A. & Donovan,
C.M. (1994) Genetic and amino acid analysis of two maize
threonine-overproducing, lysine-insensitive aspartate kinase
mutants. Theor. Appl. Genet. 89, 767–774.

15. BrinchPedersen, H., Galili, G., Knudsen, S. & Holm, P.B. (1996)
Engineering of the aspartate family biosynthetic pathway in barley

(Hordeum vulgare L.) by transformation with heterologous genes
encoding feed-back-insensitive aspartate kinase and dihydrodipicolinate synthase. Plant Mol. Biol. 32, 611–620.
16. Arruda, P., Kemper, E.L., Papes, F. & Leite, A. (2000) Regulation
of lysine catabolism in higher plants. Trends Plant Sci. 5, 324–330.
17. Goncalves-Butruille, M., Szajner, P., Torigoi, E., Leite, A. &
¸
Arruda, P. (1996) Purification and characterization of the
bifunctional enzyme lysine-ketoglutarate reductase-saccharopine
dehydrogenase from maize. Plant Physiol. 110, 765–771.
18. Gaziola, S.A., Teixeira, C.M.G., Lugli, J., Sodek, L. & Azevedo,
R.A. (1997) The enzymology of lysine catabolism in rice seeds.
Isolation, characterization, and regulatory properties of a lysine
2-oxoglutarate reductase/saccharopine dehydrogenase bifunctional
polypeptide. Eur. J. Biochem. 247, 364–371.
19. Miron, D., Ben-Yaacov, S., Karchi, H. & Galili, G. (1997) In vitro
dephosphorylation inhibits the activity of soybean lysineoxoglutarate reductase in a lysine-regulated manner. Plant J. 12,
1453–1458.
20. Cunha-Lima, S.T., Azevedo, R.A., Santoro, L.G., Gaziola, S.A.
& Lea, P.J. (2003) Isolation of the bifunctional enzyme lysine
2-oxoglutarate reductase-saccharopine dehydrogenase from
Phaseolus vulgaris. Amino Acids 24, 179–186.
21. Cunha-Lima,S.T.,Azevedo,R.A.&Santoro,L.G.(2003)Improved
procedures for extraction of lysine 2-oxoglutarate reductase/saccharopine dehydrogenase (LOR/SDH) enzyme from Phaseolus
vulgaris cultivars. New Zeal. J. Crop Hort. Sci. 31, 261–268.
22. Schmidt, R.J., Ketudat, M., Aukerman, M.J. & Hoschek, G.
(1992) Opaque-2 is a transcriptional activator that recognizes a
specific target site in 22-kD zein genes. Plant Cell 4, 689–700.
23. Neto, G.C., Yunes, J.A., DaSilva, J., Vettore, A.L., Arruda, P. &
Leite, A. (1995) The involvement of opaque-2 on beta-prolamin
gene regulation in maize and coix suggests a more general role for

this transcriptional activator. Plant Mol. Biol. 27, 1015–1029.
24. Lohmer, S., Maddaloni, M., Motto, M., DiFonzo, N., Hartings,
H., Salamini, F. & Thompson, R.D. (1991) The maize regulatory
locus opaque-2 encodes a DNA binding protein which activates
the transcription of the B-32 gene. EMBO J. 10, 617–624.
25. Maddaloni, M., Donini, G., Balconi, C., Rizzi, E., Gallusci., P.,
Forlani, F., Lohmer, S., Thompson, R., Salamini, F. & Motto, M.
(1996) The transcriptional activator Opaque-2 controls the
expression of a cytosolic form of pyruvate orthophosphate
dikinase-1 in maize endosperms. Mol. Gen. Genet. 250, 647–654.
26. Habben, J.E., Moro, G.L., Hunter, B.G., Hamaker, B.R. &
Larkins, B.A. (1995) Elongation factor 1 alpha concentration is
highly correlated with the lysine content of maize endosperm.
Proc. Natl Acad. Sci. USA 92, 8640–8644.
27. Damerval, C. & Le Guilloux, M. (1998) Characterization of novel
target proteins of the o2 mutation expressed during maize
endosperm development. Mol. Gen. Genet. 257, 354–361.
28. Brennecke, K., Souza-Neto, A.J., Lugli, J., Lea, P.J. & Azevedo,
R.A. (1996) Aspartate kinase in the maize mutants Ask1-LT19
and opaque-2. Phytochemistry 41, 707–712.
29. Kemper, E.L., Cord-Neto, G., Papes, F., Martine Z., Moraes,
K.C., Leite, A. & Arruda, P. (1999) The role of opaque-2 on
the control of lysine degrading activities in developing maize
endosperm. Plant Cell 11, 1981–1994.
30. Wang, X. & Larkins, B.A. (2001) Genetic analysis of amino acid
accumulation in opaque-2 maize endosperm. Plant Physiol. 125,
1766–1777.
31. Hunter, B.G., Beatty, M.K., Singletary, G.W., Hamaker, B.R.,
Dilkes, B.P., Larkins, B.A. & Jung, R. (2002) Maize opaque
endosperm mutations create extensive changes in patters of gene

expression. Plant Cell 14, 2591–2612.


Ó FEBS 2003

4908 R. A. Azevedo et al. (Eur. J. Biochem. 270)
32. Landry, J., Delhaye, S. & Damerval, C. (2000) Improved method
for isolating and quantitating a-amino nitrogen as nonprotein,
true protein, salt-soluble proteins, zeins, and true glutelins in maize
endosperm. Cereal Chem. 77, 620–626.
33. Landry, J. & Delhaye, S. (1996) A simple and rapid procedure for
hydrolysing minute amounts of proteins with alkali. Anal. Biochem. 243, 191–194.
34. Consoli, L. & Damerval, C. (2001) Quantification of individual
zein isoforms resolved by two-dimensional electrophoresis: genetic
variability in 45 maize inbred lines. Electrophoresis 22, 2983–2989.
35. Neuhoff, V., Arold, N., Taube, D. & Ehrhardy, W. (1988)
Improved staining of proteins in polyacrylamide gels including
isoelectric-focusing gels with clear background at nanogram sensitivity using coomassie brilliant blue G-250 and R.-250. Electrophoresis 9, 255–262.
36. Bradford, M.M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilising the
principle of protein-dye binding. Anal. Biochem. 72, 248–254.
37. Di Fonzo, N., Gentinetta, E., Salamini, F. & Soave, C. (1979)
Action of the opaque-7 mutation on the accumulation of storage
products in maize endosperm. Plant Sci. Lett. 14, 345–354.
38. Misra, P.S., Mertz, E.T. & Glover, D.V. (1975) Studies on corn
proteins. VIII. Free amino acid content of opaque-2 and double
mutants. Cereal Chem. 52, 844–848.
39. Balconi, C., Berardo, N., Reali, A. & Moro, M. (1998) Variation
in protein fractions and nitrogen metabolism of developing

40.
41.

42.

43.

44.

45.

normal and opaque endosperm mutants of maize. Maydica 43,
195–203.
Righetti, P.G., Gianazza, E., Viotti, A. & Soave, C. (1977) Heterogeneity of storage proteins in maize. Planta 136, 115–123.
Lugli, J., Campbell, A., Gaziola, S.A., Lea, P.J. & Azevedo, R.A.
(2002) Enzymes of lysine metabolism from Coix lacryma-jobi
seeds. Plant Physiol. Biochem. 40, 25–32.
Zhu, X. & Galili, G. (2003) Increased lysine synthesis coupled with
a knockout of its catabolism synergistically boosts lysine content
and also transregulates the metabolism of other amino acids in
arabidopsis seeds. Plant Cell 15, 845–853.
Wang, X., Stumpf, D.K. & Larkins, B.A. (2001) Asparate kinase 2.
A candidate gene of a quantitative trait locus influencing free
amino acid content in maize endosperm. Plant Physiol. 125, 1778–
1787.
Kemper, E.L., Cord-Neto, G., Capella, A.N., Goncalves¸
Butruille, M., Azevedo, R.A. & Arruda, P. (1998) Structure
and regulation of the bifunctional enzyme lysine-oxoglutarate
reductase-saccharopine dehydrogenase in maize. Eur. J. Biochem.
253, 720–729.

Gaziola, S.A., Sodek, L., Arruda, P., Lea, P.J. & Azevedo, R.A.
(2000) Degradation of lysine in rice seeds: effect of calcium, ionic
strength, S-adenosylmethionine and S-2-aminoethyl-L-cysteine on
the lysine 2-oxoglutarate reductase-saccharopine dehydrogenase
bifunctional enzyme. Physiol. Plant. 110, 164–171.