Implication of the glutamine synthetase
⁄
glutamate
synthase pathway in conditioning the amino acid
metabolism in bundle sheath and mesophyll cells of
maize leaves
Marie-He
´
le
`
ne Valadier
1
, Ayako Yoshida
2
, Olivier Grandjean
3
, Halima Morin
3
, Jocelyne
Kronenberger
3
, Ste
´
phanie Boutet
1
, Adeline Raballand
1
, Toshiharu Hase
2
, Tadakatsu Yoneyama
4

Versailles cedex, France
Fax: +33 1 30 83 30 96
Tel: +33 1 30 83 30 87
E-mail:
(Received 20 February 2008, revised 16
April 2008, accepted 17 April 2008)
doi:10.1111/j.1742-4658.2008.06472.x
We investigated the role of glutamine synthetases (cytosolic GS1 and chlo-
roplast GS2) and glutamate synthases (ferredoxin-GOGAT and NADH-
GOGAT) in the inorganic nitrogen assimilation and reassimilation into
amino acids between bundle sheath cells and mesophyll cells for the remo-
bilization of amino acids during the early phase of grain filling in Zea mays
L. The plants responded to a light ⁄ dark cycle at the level of nitrate, ammo-
nium and amino acids in the second leaf, upward from the primary ear,
which acted as the source organ. The assimilation of ammonium issued
from distinct pathways and amino acid synthesis were evaluated from the
diurnal rhythms of the transcripts and the encoded enzyme activities of
nitrate reductase, nitrite reductase, GS1, GS2, ferredoxin-GOGAT,
NADH-GOGAT, NADH-glutamate dehydrogenase and asparagine synthe-
tase. We discerned the specific role of the isoproteins of ferredoxin and
ferredoxin:NADP
+
oxidoreductase in providing ferredoxin-GOGAT with
photoreduced or enzymatically reduced ferredoxin as the electron donor.
The spatial distribution of ferredoxin-GOGAT supported its role in the
nitrogen (re)assimilation and reallocation in bundle sheath cells and
mesophyll cells of the source leaf. The diurnal nitrogen recycling within the
plants took place via the specific amino acids in the phloem and xylem
exudates. Taken together, we conclude that the GS1 ⁄ ferredoxin-GOGAT
cycle is the main pathway of inorganic nitrogen assimilation and recycling

require reductants and ATP. Fd and Fd:NADP
+
oxi-
doreductase (FNR, EC 1.18.1.2) occupy a central posi-
tion to mediate chloroplast electron flow to yield
reducing equivalents [6]. The nitrogen metabolism
between BSCs and MCs depends on the efficient distri-
bution of energy between photosystem I (PS I) and
photosystem II (PS II) via the electron flow specific to
the two cell types. As a result, inorganic nitrogen
assimilation into amino acids is tightly correlated with
photosynthesis. Furthermore, light at low fluence
entrains circadian rhythms and plays an essential role
for molecular signalling in the expression of the genes
and encoded enzymes involved in nitrate assimilation
and amino acid synthesis [7].
The stalk and leaves below and above the ear act
as the source organs for nitrogen reallocation in the
reproductive stage of maize [8,9]. In the source
leaves, it has been postulated that the metabolic shift
from the GS2 ⁄ GOGAT cycle to the GS1 ⁄ GDH
pathway is responsible for ammonium assimilation
into glutamine and glutamate, as a result of a
decline in GS2 and the induction of the a-GDH
subunit (for a review, see [10]). However, the role of
GDH is controversial [11–14], and the regulation
and function of GOGATs in nitrogen remobilization
remain to be evaluated. In this study, we examined
the diurnal responses of the plants, which provide
valuable cues to nitrogen and carbon metabolism

The increase in ammonium was inversely correlated
with the decline in glutamic acid and aspartic acid in
the light (Fig. 1B,D,F).
Expression of the genes involved in nitrogen
assimilation
Light is a signal that regulates nitrogen metabolism,
and nitrogen assimilation into amino acids is tightly
correlated with the expression of the genes involved
[15]. Thus, we analysed the diurnal expression of the
genes encoding the enzymes of nitrogen assimilation in
the second leaf above the ear every 3 h during a 16 h
light ⁄ 8 h dark cycle. Total mRNAs were isolated and
estimated on the basis of equal total amounts of 18S
rRNA as the internal standard (data not shown). We
measured the NR transcripts as an additional control,
as the light regulation of NR expression has been
defined in maize and several plant species. The NR
mRNAs peaked at 6 h during the dark to light transi-
tion, and then decreased to undetectable levels (Fig. 2).
Similar diurnal patterns have been reported for other
plant NRs [15–17].
Gln1-1, encoding the main form of cytosolic GS1
in leaves [2], was strongly expressed, and slightly
smaller signals were detected for the Gln1-2 and
Gln1-3 mRNAs. In contrast, strong expression of
Gln1-4 was observed, as also reported in [3]. The
Amino acid metabolism and translocation M H. Valadier et al.
3194 FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS
GS1 genes (Gln1-1 to Gln1-4) were expressed in a
similar diurnal rhythm with an increase in the dark

0.0
0.2
0.4
0.6
Asp
0
5
10
15
20
Ala
0
10
20
30
40
50
Time of day (h) Time of day (h) Time of day (h)
Relative amount %
Relative amount %
Relative amount %
Amount µmol·g
–1
FW
Amount µmol·g
–1
FW
Gly
0
10

percentage relative to the total free amino acid contents, which represent the mean [lmolÆ(g fresh weight)
)1
] from five independent
plants ± standard error as follows: 17.9 ± 1.1 (3 h), 19.8 ± 1.2 (6 h), 25.4 ± 1.6 (9 h), 23.4 ± 1.4 (12 h), 18.9 ± 1.2 (15 h), 29.4 ± 1.9 (18 h),
36.0 ± 2.2 (21 h) and 28.0 ± 1.8 (24 h). The standard errors for individual amino acid contents are of the same order of magnitude as those
of the total amino acid contents for glutamine (C), glutamic acid (D), asparagine (E), aspartic acid (F), alanine (G), glycine (H) and serine (I).
Grey boxes indicate the dark phase.
M H. Valadier et al. Amino acid metabolism and translocation
FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS 3195
a second peak appeared at the beginning of the dark
phase (Fig. 2). The gdh2 mRNAs could not be
detected under our assay conditions. The ASN gene
for maize AS belongs to the light-inducible genes,
such as monocot rice ASN and Arabidopsis ASN2
and ASN3 [19]. The level of ASN mRNAs was
higher in the dark and decreased to about 70% in
the middle of the light phase (Fig. 2).
Table 1. Amino acid composition in leaves, and amino acid percentage ratio in the phloem exudates and leaves and xylem bleeding sap and
leaves in the light and dark. The amino acid composition in leaves is expressed as a percentage relative to the total amino acid contents,
which represent the mean [lmolÆ(g fresh weight)
)1
] from five independent plants ± standard error as follows: 25.48 ± 1.58 (light) and
21.90 ± 1.93 (dark). The standard errors for the individual amino acid contents are of the same order of magnitude as those of the total
amino acid contents. Phloem exudates and xylem sap were collected over a 16 h light ⁄ 8 h dark cycle.
Amino acid
Leaves (%)
% in phloem exudates ⁄ %
in leaves
% in xylem sap ⁄ %in
leaves

Gln1-2
Gln1-3
GLU
0
50
100
0
50
100
Fd I
Fd II
Fd III
gdh1
0
50
100
Gln2
0
50
100
0
50
100
0
50
100
0
50
100
0

sponds to the light phase (6–22 h) and dark phase (22–6 h). The values represent the mean from five independent plants mixed together
and expressed as a percentage relative to the maximum.
Amino acid metabolism and translocation M H. Valadier et al.
3196 FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS
In vitro activities of nitrogen assimilation into
amino acids
The in vitro activities of the key enzymes of nitrogen
assimilation were determined. NR displayed a delayed
light-induced activity compared with its mRNA abun-
dance (Fig. 3A). We detected a lower activity of NiR
than NR, and the primary nitrate reduction to nitrite
and then to ammonium took place potentially at rates
of 4–9 and 1.5–2 lmolÆh
)1
Æ(g fresh weight)
)1
, respec-
tively (Fig. 3A,B). There was a small change in NiR
activity, with a 25% decrease early in the light phase
(Fig. 3B), as observed in other plants [20]. The total
GS activity remained fairly constant at 30 lmolÆh
)1
Æ(g
fresh weight)
)1
(Fig. 3C). Because the activity ratio of
GS1 to GS2 reaches 20 in stalks at similar maturity
after anthesis [21], it is probably cytosolic GS1, which
assimilates ammonium during a day ⁄ night cycle.
Fd-GOGAT is the primary form in the source leaves,

0
2
4
6
8
10
A B C
DE F
0 3 6 9 12 15 18 21 24 0 3 6 9 12 15 18 21 24 0 3 6 9 12 15 18 21 24
0 3 6 9 12 15 18 21 24 0 3 6 9 12 15 18 21 24 0 3 6 9 12 15 18 21 24
Activity
µmol·h
–1
·g
–1
FW
Activity
µmol·h
–1
·g
–1
FW
µmol·h
–1
·g
–1
FW µmol·h
–1
·g
–1

8
10
12
Time of day (h)
NADH-GOGAT
0
0.5
1
1.5
Time of day (h)
GDH
0
5
10
15
20
25
Time of day (h)
Amination
Deamination
Fig. 3. Enzyme activities of NR (A), NiR (B), GS (C), Fd-GOGAT (D), NADH-GOGAT (E) and NADH-GDH and NAD-GDH (F) in maize leaves
collected every 3 h during a 16 h light ⁄ 8 h dark cycle. The NR assay was carried out in a reaction mixture in the presence of 10 m
M MgCl
2
(Mg
2+
)or5mM EDTA (EDTA) for the divalent cation-dependent activity and maximum catalytic activity, respectively. The GDH activity was
assayed for NADH-dependent glutamate synthetic activity (amination) and NAD
+
-dependent glutamate oxidation activity (deamination). Error

is not well understood. Therefore, we determined the
transcript levels of Fd and FNR, which are both
encoded by a small gene family [25]. We found constit-
utive mRNA levels of Fds and FNRs, except for
Fd VI, which gave two peaks at 3 and 15 h (Fig. 2).
Fd I, Fd II, Fd V, L-FNR 1 and L-FNR 2 are mainly
distributed in the leaves, whereas Fd III and Fd VI are
found in nonphotosynthetic tissues [26,27]. The lowest
mRNA level was detected for Fd I at 3 h at 80% of
the maximum (Fig. 2). Two-phase specific promoters
and ⁄ or mRNA stability could entrain two peaks of
Fd VI and gdh1 [7].
C
D
AB
E
F
BSC BSC
MC MC
BSC BSC
MC MC
chl chl MCMC
Fig. 4. In situ hybridization of GLU mRNA for Fd-GOGAT and ASN
mRNA for AS in thin sections of 21-day-old maize leaves. (A) Leaf
bundle sheath cell section using an antisense GLU mRNA probe.
(B) Control leaf bundle sheath cell section using a sense GLU
mRNA probe. (C) Leaf mesophyll cell section using an antisense
GLU mRNA probe. (D) Control leaf mesophyll cell section using a
sense GLU mRNA probe. (E) Leaf section using an antisense ASN
mRNA probe. (F) Control leaf section using a sense ASN mRNA

catalytic reaction of Fd-GOGAT through a redox cas-
cade of Fd and FNR. As shown in Fig. 6A, rapid
NADPH oxidation was observed when all protein
components and the substrates glutamine and 2-oxo-
glutarate were present in the assay mixture. In con-
trast, only a basal level of NADPH oxidation,
uncoupled to glutamate formation, was observed with-
out either substrate. The N-terminal cysteine of
Fd-GOGAT was essential for the amido transfer reac-
tion, and an Fd-GOGAT mutant with this cysteine
residue substituted by glycine, Cys1Gly, showed no
significant NADPH oxidation (Fig. 6B). NADPH
oxidation was correlated with glutamate formation
measured by HPLC (data not shown), confirming that
NADPH supported glutamate synthesis.
To further investigate the NADPH ⁄ FNR ⁄ Fd-GO-
GAT electron pathway in glutamate synthesis,
NADPH oxidation was assayed using several combina-
tions of photosynthetic isoproteins (L-FNR ⁄ Fd I,
L-FNR ⁄ Fd II or L-FNR ⁄ Fd V) or a combination of
nonphotosynthetic isoproteins (R-FNR ⁄ Fd III). The
rate of NADPH oxidation in the nonphotosynthetic
system was most efficient of all combinations of FNRs
and Fds (Table 2). The R-FNR ⁄ Fd III combination
gave an activity of about three-fold higher than those
of L-FNR ⁄ Fd I, L-FNR ⁄ Fd II and L-FNR ⁄ Fd V, all
of which showed a similar activity (Table 2). When
glutamate formation was determined as a function of
Fd concentration, the kinetics of NADPH oxidation in
the R-FNR ⁄ Fd III system were high, particularly at

.
Table 2. Comparison of Fd-GOGAT activity supported by different
combinations of Fds and FNRs. The reaction mixture contained
50 m
M Tris ⁄ HCl, pH 7.5, 100 mM NaCl, 0.2 mM NADPH, 5 mM
2-oxoglutarate, 5 mM glutamine and maize recombinant proteins as
follows: 0.2 l
M L-FNR or R-FNR, 20 lM Fd isoprotein and 0.36 lM
Fd-GOGAT. Fd-GOGAT activity is expressed as the rate of NADPH
oxidation [lmolÆmin
)1
Æ(mg Fd-GOGAT protein)
)1
].
Reaction Specific activity
Photosynthetic isoproteins
L-FNR ⁄ Fd I 0.508
L-FNR ⁄ Fd II 0.476
L-FNR ⁄ Fd V 0.405
Nonphotosynthetic isoproteins
R-FNR ⁄ Fd III 1.28
M H. Valadier et al. Amino acid metabolism and translocation
FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS 3199
acid analysis was carried out in exudates harvested
over a 16 h light phase and 8 h dark phase. The amino
acid composition in the phloem exudates was very dif-
ferent from that in xylem sap (Fig. 7), suggesting that
there was little contamination from xylem, and vice
versa. Glx (glutamine and glutamate: five carbon
amide and amino acid) and Asx (asparagine and

the post-flowering maize ear leaf. A large accumulation
of ammonium in the second half of the light period
revealed that ammonium assimilation was substantially
inhibited in response to ammonium formation. Despite
the low abundance of mRNA for four Gln1 genes,
active GS1 partially converted a high level of ammo-
nium into glutamine, which transiently increased
shortly after the ammonium peak. However, glutamine
could not be further metabolized because of glutamate
deficiency (Fig. 1). To obtain an insight into nitrogen
assimilation, we showed that Fd-GOGAT was located
in the chloroplasts of both BSCs and MCs (Fig. 5). To
our knowledge, this is the first demonstration of Fd-
GOGAT mRNAs in the cytoplasm on the periphery of
chloroplasts and of the enzyme protein in the chlorop-
lasts of the two cell types. This spatial distribution of
Fd-GOGAT contrasts with its exclusive localization in
BSCs of maize [29].
BSCs contain most of the photorespiratory
enzymes [23]. In the post-flowering maize ear leaf,
mitochondrial glycine decarboxylase complex
(EC 1.4.4.2 ⁄ 2.1.2.10) produces photorespiratory
ammonium [30] at rates between 25 and 50% of pri-
mary nitrate reduction (Fig. 3). As the [
15
N] label from
[
15
N]glycine, fed to maize leaf, is recovered within
45 min exclusively in glutamine and glutamate [31],

bleeding sap (B) collected from maize during a 16 h light phase
(grey bars, Light) and 8 h dark phase (black bars, Dark). Phloem
exudates were collected in tubes filled with 10 m
M Hepes buffer,
pH 7.5 containing 1 m
M EDTA. Xylem sap was obtained from cut
stumps of decapitated plants. The amino acid composition is
expressed as a percentage relative to the total amino acid con-
tents, which represent the mean (nmolÆ100 lL
)1
) from three inde-
pendent plants ± standard error as follows: 44.3 ± 2.9 (phloem,
Light), 19.8 ± 1.2 (phloem, Dark), 34.1 ± 2.0 (xylem, Light) and
24.6 ± 1.5 (xylem, Dark). The standard errors for individual amino
acid contents are of the same order of magnitude as those of the
total amino acid contents.
Amino acid metabolism and translocation M H. Valadier et al.
3200 FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS
In vitro Fd-GOGAT assay showed that Fds reduced
with NADPH via FNRs display a several-fold higher
ability to donate electrons to GOGAT (Table 2) than
does photoreduced Fd [34]. In spite of the low FNR
and Fd concentrations in BSCs (30 and 40 lm) [35,36],
a close location of FNRs, Fds and Fd-GOGAT on the
thylakoid membrane [25,37] allows protein–protein
complex formation essential for the GOGAT reaction
[24]. To our knowledge, our results provide evidence
for the first time that FNR couples Fd reduction with
NADPH oxidation in the GOGAT reaction. The data
indicate that the NADPH ⁄ FNR ⁄ Fd system drives a

nitrogen assimilation and presumably sulfur reduction
in the plastids, where the oxidative pentose phosphate
pathway produces NADPH [42,43].
Large amounts of ammonium are produced in the
ear leaf in response to the induction of proteolysis [44],
up to several fold higher than primary ammonium
(Fig. 3). Ammonium incorporation into glutamine and
glutamate occurs exclusively by GS, GOGAT and
GDH in a broad range of organisms [45]. The rapid
ammonium accumulation and contrasting shortage of
glutamate in the second half of the light phase provide
evidence that the impairment of ammonium assimila-
tion by the GS1 ⁄ GOGAT cycle is caused by the
decline in Fd-GOGAT and NADH-GOGAT (Figs 1
and 3). The active GDH does not contribute to allevi-
ate the excess ammonium into glutamate. This
contrasts with the proposed role of GDH in assimilat-
ing excess ammonium in the source leaves in which
GDH is induced after pollination (for a review, see
[9,10]). Genetic evidence indicates that members of the
GDH S_50
II
class, including plant mitochondrial
NADH-GDHs, oxidize glutamate. By contrast, mem-
bers of the GDH S-50
I
class, such as plastidial
NADPH-GDH (EC 1.4.1.4) of Chlorella, assimilate
ammonium into glutamate [46]. In fact, chloroplast
NADPH-GDH is found in higher plants [47], suggest-

at the border of BSCs ⁄ vascular parenchyma. The
amino acids are then apoplastically loaded into the
companion cell–sieve element complexes because of the
low abundance of plasmodesmata [48]. By contrast,
the phloem loading from MCs requires additional
H
+
-amino acid transporters across the MC–BSC inter-
face, and depends on the continuity of the electro-
chemical H
+
gradient between the two cell types. The
location of the GS1 ⁄ Fd-GOGAT cycle in the BSCs,
surrounding sieve element, meets the demand of amino
M H. Valadier et al. Amino acid metabolism and translocation
FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS 3201
acid synthesis in these cells, from which the amino
acids are loaded to the phloem for the grain. In the
phloem sap, glutamine is the preferred nitrogen carrier
rather than asparagine (Table 1). This can be attrib-
uted in part to the amide group on the d-carbon of
glutamine, which increases the binding affinity to the
transporter (AAP5) by at least three orders of magni-
tude compared with asparagine [49]. In Arabidopsis,
dark and sugar induce ASN1 and gln1-1, respectively,
and repress gln1-1 and ASN1, respectively. These
expression patterns correlate with the relative abun-
dance of asparagine and glutamine in the leaves [50].
The expression of gln1-3 and ASN in maize is inhibited
by light and sugar, respectively [51,52]. Therefore, the

GmbH). The abundance of initial cDNA strands between
samples was corrected using agarose gel electrophoresis and
Quantum RNA 18S internal standards (Ambion, Austin,
TX, USA). PCR was performed on a LightCycler Instru-
ment (Roche, Basle, Switzerland). For the genes of the multi-
gene family, the specific oligonucleotides were designed
along the nonconserved stretches of the genes in the same
gene family. The following specific primer sets were used for
each gene, indicated by the GenBank database accession
number: NR1 (accession number M27821): forward primer,
5¢-CTCAAGCGCATCATCGTCAC-3¢; reverse primer,
5¢-ATGATCTGGTACATGGGCGTG-3¢; GS1-1 (D14576,
X65929): forward primer, 5¢-CCCTCCTTCCTCCTTGG
GTT-3¢; reverse primer, 5¢-ATGGAATGGAAGTGGTGG
GAA-3¢; GS1-2 (D14577, X65928): forward primer,
5¢-TCTCGGACAACACCGAGAAGA-3¢; reverse primer,
5¢-CACAAGTGTGGTACGGCCATT-3¢; GS1-3 (D14578,
X65930): forward primer, 5¢-CAGCTCTTCTTGGGTTGC
CTA-3¢; reverse primer, 5¢-GTACCCAATAAACGGGA
AGCG-3¢; GS1-4 (D14579, X65926): forward primer,
5¢-CTTCTCGTCTGCCCGAGT-3¢; reverse primer, 5¢-CTG
GAAGCACAGCCAAACGTA-3¢; GS2 (X65931): forward
primer, 5¢-GACGGTTGGTTCGGGAATG-3¢; reverse
primer, 5¢-TCCGATGAATCAAAGACAGCC-3¢; Fd-GO
GAT (M59190): forward primer, 5¢-GCTGCTATGGGAG
CTGATGAA-3¢; reverse primer, 5¢-GCAACGGCCAAG
AATCATGTA-3¢; GDH1 (D49475): forward primer, 5¢-
TTGTTCCTTGGGAGGATAGAAAAA-3¢; reverse primer,
5¢-TTGCTTGCAGACAGCATCTCA-3¢; ASN (X82849):
forward primer, 5¢-AAAGCTTCATCGCAGCTCGT-3¢;

taining 0.1% Triton X-100 in NaCl ⁄ P
i
(10 mm sodium phos-
phate, pH 7.0 and 130 mm NaCl). After dehydration in a
Amino acid metabolism and translocation M H. Valadier et al.
3202 FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS
graded ethanol series (10%, 30%, 50%, 70% and 96%), tis-
sues were incubated in an ethanol ⁄ histoclear series (2 : 1,
1 : 1, 1 : 2, v ⁄ v), histoclear and histoclear ⁄ paraffin (1 : 1,
v ⁄ v), and embedded in paraffin at 59 °C.
Hybridization probe preparation
Total RNA was extracted from maize leaves using an RNA
isolation kit (Qiagen GmbH). First cDNA strands were
synthesized from 2 lg of RNA using an Omniscript RT kit
(Qiagen GmbH). Partial DNAs of Fd-GOGAT and AS
were amplified by PCR using the following primers. Fd-
GOGAT: sense probe: forward primer, 5¢-TGTAATTCGA
CTCACTATAGGGTACGCAGCCACCAGTCATGTA-3¢;
reverse primer, 5¢-TACGCAGCCACCAGTCATGTA-3¢;
antisense probe: forward primer, 5¢-CTTAGGGTGGACG
GTGGATTC-3¢; reverse primer, 5¢-TGTAATTCGACTC
ACTATAGGGTACGCAGCCACCAGTCATGTA-3¢; AS:
sense probe: forward primer, 5¢-TGTAATTCGACTCACT
ATAGGGCCTCCCTGCTAGCTTCTACCG-3¢; reverse
primer, 5¢-TCCAGACATACAGACACGGGC-3¢; antisense
probe: forward primer, 5¢-CCTCCCTGCTAGCTTC
TACCG-3¢; reverse primer, 5¢-TGTAATTCGACTCACTA
TAGGGCTCCAGACATACAGACACGGGC-3¢. Sense
and antisense DNAs (400 ng each) were labelled with DIG-
UTP using a transcription kit (Promega, Madison, WI,

solution (T1 including 1% BSA and 0.5% Triton X-100) at
room temperature. Slides were incubated with alkaline
phosphatase-conjugated anti-DIG IgG in T3. Alkaline phos-
phatase activity was developed with 5-bromo-4-chloro-3-
indolyl-phosphate (50 mgÆmL
)1
) and nitroblue tetrazolium
(75 mgÆmL
)1
). Slides were sealed with gel mount formol1
(Microm Microtech France, Francheville, France), and fluo-
rescence was observed using a Leica DMR microscope
(Leica Microsystems, Wetzlar, Germany).
Indirect immunofluorescence analysis
Leaf sections were fixed in 3.7% (w ⁄ v) formaldehyde in
50 mm PIPES buffer, pH 6.9, 5 mm MgSO
4
and 5 mm
EGTA (MTSB), and then in NaCl ⁄ P
i
(6.5 mm Na
2
HPO
4
,
1.5 mm KH
2
PO
4
, pH 7.3, 14 mm NaCl and 2.7 mm KCl).

2
or 5 mm EDTA for the divalent
cation-dependent activity and maximum catalytic activity,
respectively. NiR and GS were extracted and assayed
according to [17]. Fd-GOGAT and NADH-GOGAT were
extracted and assayed by measuring glutamate formation
by HPLC as described in [22]. GDH was extracted and
assayed for reductive glutamate synthetic activity and glu-
tamate oxidation activity according to [17].
Reconstituted electron transfer system to
Fd-GOGAT
Fd-GOGAT was assayed by reconstituting the electron
transfer pathway from NADPH to Fd via FNR as
M H. Valadier et al. Amino acid metabolism and translocation
FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS 3203
described essentially in [36]. The recombinant maize
proteins of FNRs and Fds (L-FNR, R-FNR, Fd I, Fd II,
Fd III and Fd V) were prepared in the Escherichia coli
expression system [25,35]. Maize Fd-GOGAT was also pre-
pared in a similar system (T. Hase, unpublished work). The
complete reaction mixture contained 50 mm Tris ⁄ HCl,
pH 7.5, 100 mm NaCl, 0.2 mm NADPH, 5 mm 2-oxogluta-
rate, 5 mm glutamine and maize recombinant proteins as
follows: 0.2 lm L-FNR 1 or R-FNR, 20 lm of Fd I, Fd II,
Fd III or Fd V and 0.36 lm of Fd-GOGAT. The oxidation
of NADPH was followed by monitoring the decrease in
A
340 nm
. The formation of glutamate was also analysed with
an equivalent assay system. The reaction mixture contained

ysed as described in [17]. Free ammonium contents were
determined by the phenol hypochlorite assay (Berthelot
assay). Soluble protein contents were determined by the
Coomassie blue dye-binding assay (Bio-Rad Laboratories,
Hercules, CA, USA).
Acknowledgements
We thank Dr David Tepfer for proofreading the
manuscript. We also thank Franc¸ ois Gosse for culture
and maintenance of the plants.
References
1 Moore R & Black C Jr (1979) Nitrogen assimilation
pathways in leaf mesophyll and bundle sheath cells of
C
4
photosynthesis plants formulated from comparative
studies with Digitaria sanguinalis (L.) Scop. Plant Phys-
iol 64, 309–313.
2 Sakakibara H, Kawabata S, Takahashi H, Hase T &
Sugiyama T (1992) Molecular cloning of the family of
glutamine synthetase genes from maize: expression of
genes for glutamine synthetase and ferredoxin-depen-
dent glutamate synthase in photosynthetic and non-
photosynthetic tissues. Plant Cell Physiol 33, 49–58.
3 Li M-G, Villemur R, Hussey PJ, Silflow CD, Gantt JS
& Snustad DP (1993) Differential expression of six glu-
tamine synthetase genes in Zea mays. Plant Mol Biol
23, 401–407.
4 Muhitch MJ (2003) Distribution of the glutamine syn-
thetase isoenzyme GS
p1

synthetase and glutamate dehydrogenase in nitrogen
assimilation and possibilities for improvement in the
nitrogen utilization of crops. J Exp Bot 53, 979–987.
Amino acid metabolism and translocation M H. Valadier et al.
3204 FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS
11 Magalha
˜
es JR, Ju GC, Rich PJ & Rhodes D (1990)
Kinetics of
15
NH
4
+
assimilation in Zea mays. Prelimin-
ary studies with a glutamate dehydrogenase (GDH1)
null mutant. Plant Physiol 94, 647–656.
12 Stewart GR, Shatilov VR, Turnbull MH, Robinson SA
& Goodall R (1995) Evidence that glutamate dehydro-
genase plays a role in the oxidative deamination of glu-
tamate in seedlings of Zea mays. Aust J Plant Physiol
22, 805–809.
13 Masclaux-Daubresse C, Reisdorf-Cren M, Pageau K,
Lelandais M, Grandjean O, Kronenberger L, Valadier
M-H, Feraud M, Jouglet T & Suzuki A (2006) Gluta-
mine synthetase ⁄ glutamate synthase pathway and gluta-
mate dehydrogenase play distinct roles for sink ⁄ source
nitrogen cycle in tobacco. Plant Physiol 140, 444–456.
14 Purnell MP & Botella JR (2007) Tobacco isoenzyme 1
of NAD(H)-dependent glutamate dehydrogenase catab-
olizes glutamate in vivo. Plant Physiol 143, 530–539.

21 Ta CT (1991) Nitrogen metabolism in the stalk tissue of
maize. Plant Physiol 97, 1375–1380.
22 Suzuki A, Rioual S, Lemarchand S, Godfloy N, Roux
Y, Boutin JP & Rothstein S (2001) Regulation by light
and metabolite of ferredoxin-dependent glutamate syn-
thase in maize. Physiol Plant 112, 524–530.
23 Majeran W, Cai Y, Sun Q & van Wijk KJ (2005) Func-
tional differentiation of bundle sheath and mesophyll
maize chloroplasts determined by comparative proteo-
mics. Plant Cell 17, 3111–3140.
24 Suzuki A & Knaff DB (2005) Glutamate synthase:
structural, mechanistic and regulatory properties, and
role in amino acid metabolism. Photosynth Res 83,
191–217.
25 Okutani S, Hanke GT, Satomi Y, Takao T, Kurisu G,
Suzuki A & Hase T (2005) Three maize leaf ferre-
doxin:NADPH oxidoreductases vary in subchloroplast
location, expression, and interaction with ferredoxin.
Plant Physiol 139, 1451–1459.
26 Onda Y, Matsumura T, Kimata-Ariga Y, Sakakibara
H, Sugiyama T & Hase T (2000) Differential interaction
of maize root ferredoxin:NADP
+
oxidoreductase with
photosynthetic and nonphotosynthetic ferredoxin isoen-
zymes. Plant Physiol 123, 1037–1045.
27 Sakakibara H (2003) Differential response of genes for
ferredoxin and ferredoxin:NADP
+
oxidoreductase to

34 Anderson JW & Done J (1977) Polarographic study of
ammonia assimilation by isolated chloroplasts. Plant
Physiol 60, 504–508.
35 Matsumura T, Kimata-Ariga Y, Sakakibara H, Sugiy-
ama T, Murata H, Takao T, Shimonishi Y & Hase T
(1999) Complementary DNA cloning and characteriza-
tion of ferredoxin localized in bundle-sheath cells of
maize leaves. Plant Physiol 119, 481–488.
M H. Valadier et al. Amino acid metabolism and translocation
FEBS Journal 275 (2008) 3193–3206 ª 2008 The Authors Journal compilation ª 2008 FEBS 3205
36 Yonekura-Sakakibara K, Onda Y, Ashikari T, Tanaka
Y, Kusumi T & Hase T (2000) Analysis of reductant
supply systems for ferredoxin-dependent sulfite reduc-
tase in photosynthetic and non-photosynthetic organs
of maize. Plant Physiol 122, 887–894.
37 Bottela JR, Verbelen JP & Valpuesta V (1988) Immuno-
cytolocalization of ferredoxin-GOGAT in the cells of
green leaves and cotyledons of Lycopersicon esculentum.
Plant Physiol 87, 255–257.
38 Jin T, Huppe HC & Turpin DH (1998) In vitro
reconstitution of electron transport from glucose-6-
phosphate and NADPH to nitrite. Plant Physiol 117,
303–309.
39 Thomas J-C, Ughy B, Lagoutte B & Ajliani G (2006) A
second isoform of the ferredoxin:NADP oxidoreductase
generated by an in-frame initiation of translation. Proc
Natl Acad Sci USA 103, 18368–18373.
40 Kunzelmann P & Scha
¨
fer E (1985) Phytochrome-medi-

Plant Cell Environ 25, 125–149.
49 Boorer KJ & Fischer W-N (1997) Specificity and stoi-
chiometry of the Arabidopsis H
+
⁄ amino acid trans-
porter AAP5. J Biol Chem 272, 13040–13046.
50 Thum KE, Shasha DE, Lajay LV & Coruzzi GM
(2003) Light and carbon signalling pathways.
Modeling circuits of interactions. Plant Physiol 132 ,
440–452.
51 Sakakibara H, Kawabata S, Hase T & Sugiyama T
(1992) Differential effects of nitrate and light on the
expression of glutamine synthetase and ferredoxin-
dependent glutamate synthase in maize. Plant Cell
Physiol 33, 1193–1198.
52 Chevalier C, Bourgeois E, Just D & Raymond P (1996)
Metabolic regulation of asparagine synthetase gene
expression in maize (Zea mays L.) root tips. Plant J 9,
1–11.
53 Rosen H (1957) A modified ninhydrin colorimetric anal-
ysis for amino acids. Arch Biochem Biophys 67, 10–15.
54 Feraud M, Masclaux-Daubresse C, Ferrario-Me
´
ry S,
Pageau K, Lelandais M, Ziegler C, Leboeuf E, Jouglet
T, Viret L, Spampinato A et al. (2005) Expression of a
ferredoxin-dependent glutamate synthase gene in meso-
phyll and vascular cells and functions of the enzyme in
ammonium assimilation in Nicotiana tabacum (L.).
Planta 222, 667–677.