synthase pathway in conditioning the amino acidmetabolism in bundle sheath and mesophyll cells of maize leaves Marie-He´le`ne Valadier1, Ayako Yoshida2, Olivier Grandjean3, Halima Morin3
Trang 1synthase pathway in conditioning the amino acid
metabolism in bundle sheath and mesophyll cells of
maize leaves
Marie-He´le`ne Valadier1, Ayako Yoshida2, Olivier Grandjean3, Halima Morin3, Jocelyne
Kronenberger3, Ste´phanie Boutet1, Adeline Raballand1, Toshiharu Hase2, Tadakatsu Yoneyama4 and Akira Suzuki1
1 Unite´ de Nutrition Azote´e des Plantes, Institut National de la Recherche Agronomique, Versailles, France
2 Institute for Protein Research, Osaka University, Japan
3 Laboratoire Commun de Cytologie, Institut National de la Recherche Agronomique, Versailles, France
4 Department of Applied Biological Chemistry, University of Tokyo, Japan
In the C4 plant maize, inorganic nitrate reduction to
ammonium and subsequent ammonium assimilation
into amino acids occur in two different
photosyn-thetic cells: bundle sheath cells (BSCs) and mesophyll cells (MCs) Nitrate taken up by roots moves in part, via the vascular bundle, to leaves for reduction
Keywords
amino acid translocation; compartmentation;
glutamine and glutamate synthesis; nitrogen
assimilation; Zea mays L
Correspondence
A Suzuki, Unite´ de Nutrition Azote´e des
Plantes, Institut National de la Recherche
Agronomique, Route de St-Cyr, 78026
Versailles cedex, France
Fax: +33 1 30 83 30 96
Tel: +33 1 30 83 30 87
E-mail: suzuki@versailles.inra.fr
(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 into glutamine and glutamate, and preconditions amino acid inter-conversion and remobilization
Abbreviations
AS, asparagine synthetase (EC 6.3.5.4); BSC, bundle sheath cells; DIG, digoxigenin; Fd, ferredoxin; Fd-NiR, ferredoxin-nitrite reductase (EC 1.6.6.4); FNR, ferredoxin:NADP + oxidoreductase (EC 1.18.1.2); GDH, glutamate dehydrogenase; GOGAT, glutamate synthase
(Fd-GOGAT, EC 1.4.7.1; GS, glutamine synthetase (EC 6.1.1.3); MC, mesophyll cells; NADH-GOGAT, EC 1.4.1.14); NR, nitrate reductase (EC 1.6.6.1); PS I (II), photosystem I (II).
Trang 2In leaves, nitrate is reduced to ammonium by
cyto-solic nitrate reductase (NR; EC 1.6.6.1), and then by
plastidial ferredoxin-nitrite reductase (Fd-NiR,
EC 1.6.6.4) [1] Ammonium, also derived from
pho-torespiration, is assimilated first into the
glutamine-amide group by glutamine synthetase (cytosolic
GS1 and plastidial GS2, EC 6.3.1.2) and then into
glutamate-amino group by glutamate synthase
(Fd-GOGAT, EC 1.4.7.1; NADH-(Fd-GOGAT, EC 1.4.1.14)
in vegetative organs GS1 is encoded by five genes in
maize, and the regulation and function of each gene
have been elucidated in part [2–4] Ammonium and
glutamine-amide group are also assimilated into
asparagine by asparagine synthetases (ASs; ammonia
ligase AS, EC 6.3.1.1; glutamine hydrolyase AS,
EC 6.3.5.4) Alternatively, mitochondrial
NADH-glu-tamate dehydrogenase (NADH-GDH, EC 1.4.1.2)
can incorporate high levels of ammonium into
gluta-mate under stress [5]
Nitrogen assimilation and amino acid synthesis
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
[7,15] We assessed the role of the GS⁄ GOGAT cycle
in the nitrogen assimilation between MCs and BSCs
in the amino acid synthesis and remobilization
dur-ing the early phase of grain filldur-ing in Zea mays L
Results
High levels of glutamate and glutamate derivatives in reproductive leaves Leaves above and below the ear act as sources to export nitrogen resources to sink organs via vascular bundles In order to examine the nitrogen status in the leaves, we determined the inorganic nitrogen and amino acid contents in the second leaf above the ear every 3 h during a 16 h light⁄ 8 h dark cycle The nitrate content was high during the second half of the light phase and remained at about 16 lmolÆ(g fresh weight))1 (Fig 1A) Ammonium accumulated in the middle of the light phase up to 6 lmolÆ(g fresh weight))1, indicating that a part of the ammonium was not assimilated in the light (Fig 1B) The major amino acids in the leaves were alanine (26–39%), glycine (26–40%), glutamic acid (6–14%), serine (8–12%) and aspartic acid (4–16%) in both the light and dark (Fig 1, Table 1) Following ammonium accumulation, glutamine increased about four-fold in the light (Fig 1C) In contrast, asparagine remained at a fairly constant level in the light and dark (Fig 1E, Table 1) 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
Trang 3GS1 genes (Gln1-1 to Gln1-4) were expressed in a
similar diurnal rhythm with an increase in the dark
to a maximum at 12 h, and then barely detectable
(Fig 2) The Gln2 mRNAs were low, peaking early
in the light phase and persisting longer than the
Gln1 mRNAs (Fig 2) The GLU mRNAs for
Fd-GOGAT were found at the highest level at 6 h,
similar to the Gln2 mRNAs (Fig 2) The
NADH-GOGAT mRNAs could not be detected in our assay
conditions As GDH genes are expressed at high levels in reproductive plant leaves [18], we also measured gdh expression Maize NADH-GDH is encoded by gdh1 and gdh2 for the b-subunit and a-subunit, respectively This gdh1 is phylogenetically related to tomato gdh1, whose b-homohexamer com-plexes appear to oxidize glutamate in transgenic tobacco [14] The gdh1 mRNAs accumulated on illumination, reaching a maximum level at 12 h, and
Gln
0.0 0.5 1.0 1.5 2.0 2.5
Glu
0
5
10
15
20
Asn
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)
–1 FW
–1 FW
Gly
0 10 20 30 40 50
Ser
0 5 10 15
Nitrate
0
5
10
15
20
25
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
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
Ammonium
0 2 4 6 8
Fig 1 Levels of nitrate (A), ammonium (B) and amino acids (C–I) in maize leaves collected every 3 h during a 16 h light ⁄ 8 h dark cycle Nitrate and ammonium contents represent the mean from five independent plants ± standard error Amino acid contents are expressed as a 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.
Trang 4a 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
0
50
100
3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24
3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24
3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24
3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24
0 50 100
0
50
100
0 50 100
0 50 100
0
50
100
0 50 100
0 50 100
0 50 100
0 50 100
0 50 100
0 50 100
0 50 100
0 50 100
0 50 100
Time of day (h)
Time of day (h)
Time of day (h)
0
50
100
Time of day (h)
Fig 2 Levels of the transcripts for NR (NR), cytosolic GS1 (Gln1-1, Gln1-2, Gln1-3, Gln1-4), chloroplast GS2 (Gln2), Fd-GOGAT (GLU), GDH (gdh1), AS (ASN), Fd (Fd I, Fd II, Fd III, Fd V, Fd VI) and leaf FNR (L-FNR 1 and L-FNR 2) in maize leaves The mRNAs were estimated by RT-PCR using an equal amount of total RNA from each sample, collected every 3 h during a 16 h light ⁄ 8 h dark cycle The time of day corre-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.
Trang 5In 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)1Æ(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)1Æ(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,
accounting for 90% of total GOGAT activity, with the
rest being NADH-GOGAT (Fig 3D,E) Both
Fd-GO-GAT and NADH-GOFd-GO-GAT were induced at the end of
the dark phase, peaked at the dark⁄ light transition,
and then became undetectable The patterns indicate
that there was no further nitrogen flux through the
GOGATs in the later light phase This contrasts with the diurnally active Fd-GOGAT and NADH-GOGAT
in developing maize seedlings, in which GOGATs cope with large amounts of primary and photorespiratory ammonium [22] Interestingly, there was a substantial increase in NADH-GOGAT activity at the end of the light phase (Fig 3E) Fairly high and constant activi-ties were detected for GDH in both the synthesis and deamination of glutamate (Fig 3F)
Localization of Fd-GOGAT Ammonium assimilation into glutamine by GSs (GS1 and GS2) occurs in the two cell types [23], but the localization of the major Fd-GOGAT between MCs and BSCs remains controversial [24] Therefore, we first determined the cellular and subcellular localization
of Fd-GOGAT Fd-GOGAT mRNAs were hybridized
in situ with the digoxigenin (DIG)-labelled antisense GLU mRNA probe Staining was found in the cyto-plasmic layers of BSCs (Fig 4A) No positive staining was detected in the BSCs using the control sense probe (Fig 4B) The localization of GLU mRNAs in the vicinity of the vascular bundle of BSCs suggests a role
of Fd-GOGAT in amino acid translocation Therefore,
NR
0
2
4
6
8
10
A B C
EDTA Mg2+
NiR
0 0.5
1 1.5
2 2.5
GS
0
10
20
30
40
Fd-GOGAT
0
2
4
6
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 (Mg2+) or 5 m M 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 bars represent the standard error from five independent plants Grey boxes indicate the dark phase.
Trang 6in situ mRNA hybridization was carried out for
cyto-solic AS, which provides asparagine for nitrogen
trans-port The signal was found in the cytoplasm of BSCs
in a ring around the vascular bundle with the antisense
ASN mRNA probe (Fig 4E) Staining was not
detected with the control probe (Fig 4F)
Further-more, the signal was also found on the surface of MC
chloroplasts in the cytoplasmic layers (Fig 4C) No
positive staining was observed with the control probe
in MCs, and chloroplasts appeared to be pink against
a pale background (Fig 4D)
The cellular and subcellular localization of
Fd-GO-GAT peptide was determined in leaf sections by the
indirect immunofluorescence method, as described in
[13] Using a confocal laser scanning microscope,
spe-cific immunofluorescence was found in the chloroplasts
of BSCs (Fig 5A) No fluorescence was detected using
nonimmune serum as a primary antibody (Fig 5B)
Higher magnification of MCs showed that the Fd-GO-GAT proteins were localized to the chloroplasts (Fig 5C) No signal was detected with nonimmune serum as the primary antibody (Fig 5D)
Expression of the genes involved in chloroplast electron transport
Although the GS⁄ Fd-GOGAT pathway was found to
be distributed between BSCs and MCs, the Fd-depen-dent electron donor to the enzyme in BSC chloroplasts
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]
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
probe BSC, bundle sheath cell; chl, chloroplast; MC, mesophyll
cell Bar: 10 lm.
C
BSC
MC
D
MC
BSC
B A
Fig 5 Immunocytochemical localization of Fd-GOGAT in thin sec-tions of 21-day-old maize leaves (A) Leaf bundle sheath cell and vascular bundle section using antibody against Fd-GOGAT as the primary antibody (B) Control leaf bundle sheath cell and vascular bundle section using nonimmune serum as the primary antibody (C) Leaf mesophyll cell section using antibody against Fd-GOGAT
as the primary antibody (D) Control leaf mesophyll cell section using nonimmune serum as the primary antibody BSC, bundle sheath cell; chl, chloroplast; MC, mesophyll cell Bar: 10 lm.
Trang 7Glutamate synthesis in the reconstituted system
using NADPH, FNRs, Fds and GOGAT
Glutamate synthesis depends on a subtle specialization
of FNRs and Fds We reconstituted a
NADPH-depen-dent glutamate synthesis system using recombinant
FNR, Fd and Fd-GOGAT proteins to assess whether
NADPH serves as the initial electron donor for the
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 lower Fd concentrations, compared with the L-FNR⁄ Fd I system (Fig 6C) The results indicate that nonphotosynthetic R-FNR and Fd III isoproteins promote efficient glutamate formation using NADPH
as reductant
Amino acid translocation in the vascular streams
In order to monitor the amino acids supplied for allo-cation by the source tissues, we analysed the amino acid contents in the phloem sap Phloem exudate was promoted and collected as described in [28] As the plants exuded at lower rates in the dark, the amino
A
C
B
Fig 6 Assay for Fd-GOGAT activity in the reconstituted electron
transfer system The complete reaction mixture contained 50 m M
Tris ⁄ HCl, pH 7.5, 100 m M NaCl, 0.2 m M NADPH, 5 m M
2-oxogluta-rate, 5 m M glutamine and maize recombinant proteins as follows:
0.2 l M L-FNR 1, 20 l M Fd I and 0.36 l M of either WT (A) or
Cys1Gly mutant (B) of Fd-GOGAT As a control, 2-oxoglutarate or
glutamine was omitted from the reaction mixture The kinetics of
Fd-GOGAT activity were assayed by increasing the concentrations
of Fd isoprotein as indicated in the figure (C) Photosynthetic (s)
and nonphotosynthetic (d) combinations contained L-FNR 1 ⁄ Fd I
and R-FNR ⁄ Fd III, respectively Oxidation of NADPH was followed
by monitoring the decrease in A340 nm.
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 m M NaCl, 0.2 m M NADPH, 5 m M
2-oxoglutarate, 5 m M glutamine and maize recombinant proteins as follows: 0.2 l M L-FNR or R-FNR, 20 l M Fd isoprotein and 0.36 l M
Fd-GOGAT Fd-GOGAT activity is expressed as the rate of NADPH oxidation [lmolÆmin)1Æ(mg Fd-GOGAT protein))1].
Photosynthetic isoproteins
Nonphotosynthetic isoproteins
Trang 8acid 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
aspartate: four carbon amide and amino acid) were
found to be the main nitrogen compounds in the
phloem exudates, amounting to 51% of the total
amino acids in the light and 56% in the dark
(Fig 7A) The total amount of alanine, serine and
gly-cine was reduced from 33% in the light to 25% in the
dark (Fig 7A) Glutamine and asparagine were the
major amino acids in the xylem sap in both the light
(63% of the total amino acids) and dark (60%)
(Fig 7B)
The amino acid percentage ratio in the leaves and
phloem distinguished three groups First, glutamine
and asparagine appeared to be preferentially
transported in the phloem, as indicated by high phloem⁄ leaf ratios (12–25) (Table 1) Translocation of glutamine and asparagine in the phloem seemed to be increased in the dark (Table 1) Second, glutamate, aspartate and serine were similarly distributed in leaves and phloem sap, yielding phloem⁄ leaf ratios of between 0.9 and 1.7 (Table 1) Third, alanine and gly-cine were poorly translocated, with phloem⁄ leaf ratios below 0.7 (Table 1) Finally, amino acids were selec-tively translocated in the xylem in the form of gluta-mine and asparagine, which showed a significantly high xylem⁄ leaf ratio of between 28 and 99 in the light and dark (Table 1)
Discussion
Glutamine is the main entry point of ammonium, which can be derived from nitrate reduction, protein turnover and, to a lesser extent, photorespiration in 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 [15N] label from [15N]glycine, fed to maize leaf, is recovered within
45 min exclusively in glutamine and glutamate [31], photorespiratory ammonium is primarily re-fixed via the vascular bundle-located GS1 [32], in concert with BSC-located Fd-GOGAT However, the physiological role of Fd-GOGAT in BSC chloroplasts is a matter of debate, because BSC chloroplasts contain only 20–30% of PS II polypeptides, and most of the capac-ity for noncyclic electron transport and concomitant
Fd reduction is localized to MC chloroplasts [25,33]
Phloem exudates
0
10
20
30
A
B
Gln Glu Asn Asp Ala Gly Ser
Light Dark
Xylem saps
0
10
20
30
40
Gln Glu Asn Asp Ala Gly Ser
Amino acids
Light Dark
Fig 7 Amino acid composition in phloem exudates (A) and xylem
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.
Trang 9In vitroFd-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
specific redox reaction in vivo for BSC-located
Fd-GO-GAT in ammonium assimilation Moreover, the
nonphotosynthetic R-FNR⁄ Fd III system yields a
2.5–3-fold higher GOGAT activity than all
photosyn-thetic FNR⁄ Fd systems This reflects a higher redox
potential of Fd III ()345 mV) than the other Fds [26],
leading to rapid thermodynamic electron transfer
Without light energy, the NADPH⁄ R-FNR ⁄ Fd III
sys-tem presumably substitutes for photoreduced Fd and
sustains the GS1⁄ Fd-GOGAT cycle to assimilate
ammonium in the dark (Fig 3) The reversible electron
transfer between NADPH and Fd via FNR has been
shown to occur in cyanobacteria and green algae
[38,39] However, the contribution of this system in the
light and dark to meet the needs for reductant supply
to Fd-GOGAT has not been elucidated In addition,
the reductant supply system from NADPH to
Fd-GO-GAT via L-FNR⁄ Fd II and R-FNR ⁄ Fd III is relevant
in BSC chloroplasts (Figs 4 and 5), because of the
internal light gradient within the translucent veins of
BSCs The light absorption and scattering attenuate
the photon fluence rate by 34% at 450⁄ 680 nm and
15% at 725 nm by the initial 50 lm across the maize
mesocotyl [40] As a result, Fd reduction deprived of
sufficient light at the core of vascular bundles depends
on the sensitive NADPH⁄ FNR ⁄ Fd system Glutamate
synthesis increases on addition of NADPH at a large
excess of stromal concentrations in the dark (0.3–
0.48 mm) [41] (data not shown) This provides evidence
that the supply of reduced Fd via NADPH limits
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_50II class, including plant mitochondrial NADH-GDHs, oxidize glutamate By contrast, mem-bers of the GDH S-50I 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-ing a possible alternative role of this isoform There-fore, NADH-GDH may provide the anaplerotic pathway with 2-oxoglutarate to regenerate NADH and 2-oxalacetate for further transaminations In addition
to GOGATs, glutamate can be produced by the amin-otransferases, which, in turn, consume the equivalent amount of glutamate in the reverse reactions to form aspartate and alanine for further amino acid intercon-versions Therefore, the net synthesis of glutamate through the GS1⁄ GOGAT cycle is a prerequisite for grain development This view is supported by the evi-dence that the overall glutamate level remains constant
in the source organs (stalks and cobs) [8,9], and the nitrate supply to roots after pollination reduces the loss of amino acids from these stalks and leaves for use in grain filling [8]
The amino acids were selectively remobilized in the phloem in the form of glutamine, asparagine, gluta-mate and aspartate, which had high phloem⁄ leaf ratios (Table 1) As a result, these amino acids make up the major components of the seed storage proteins [9] The abundance of glutamine, asparagine and glutamate in the phloem sap correlates with the spatial distribution
of GS1 [32], AS and Fd-GOGAT in BSCs, arranged in one or more layers adjacent to the sieve tubes (Figs 4 and 5) The phloem loading of glutamine, asparagine and glutamate from BSCs takes place via H+-coupled amino acid transporters into the vascular parenchyma
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
Trang 10acid 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
increased ratios of asparagine to glutamine in the
phloem sap in the dark could be attributed partly to
the conversion of glutamine to asparagine by AS in
the dark
Materials and methods
Plant growth
Seeds of maize (Zea mays L cv DEA) were germinated on
sand by supplying a complete nutrient solution, as described
in [22] Maize seedlings were grown for 21 days in a
controlled growth chamber under a regime of 16 h light
(photosynthetic photon flux density of 300 lmol
pho-tonsÆm)2Æs)1 at 25C) and 8 h dark (18 C) Plants were
then grown in a glasshouse for 2 months under natural light
with irrigation by complete nutrient solution as described
previously [22] Two weeks before harvest, plants were
trans-ferred to a controlled chamber and grown in a 16 h light⁄ 8 h
dark cycle under the conditions described above Leaves
were numbered from the bottom of the plant, and the
second leaves upward from the first ear were harvested for
analysis
Relative quantitative RT-PCR
Total RNA was extracted using a kit according to the
manu-facturer’s instructions (Qiagen GmbH, Hilden, Germany)
Relative RT-PCR was carried out using rRNA as an
endoge-nous standard, and the first cDNA strands were synthesized
from 2 lg of RNA using an Omniscript RT kit (Qiagen
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¢; reverse primer, 5¢-CACGACACACACACACACGT-3¢; Fd I (M73830): forward primer, 5¢-CTACAACGTGAAGCT GATCAC-3¢; reverse primer, 5¢-GATGGGCATGAATGAT TATGCGC-3¢; Fd II (AB016810): forward primer, 5¢-CCTG GCGGTGTATAGCTAAGCAG-3¢; reverse primer, 5¢-CTG AGCATGAGCATCCTCC-3¢; Fd III (M73831): forward primer, 5¢-CGAAGGTTCCAAGCCTGAAGACC-3¢; reverse primer, 5¢-CTAGCAGAACATAGAAGACAGC-3¢; Fd V (M73828): forward primer, 5¢-TCCAGCCATTACCCGCA GCTAGC-3¢; reverse primer, 5¢-GCTTAGGAGATAAG GTCGTCCTCC-3¢; Fd VI (AB001385): forward primer, 5¢-GACGGAGCACGAGTTCGAGGC-3¢; reverse primer, 5¢-CTCATATGCCATGATCTCATCG-3¢, L-FNR 1 (AB035644): forward primer, 5¢-ACAACACAAAATGTCAGCTGC AAAA-3¢; reverse primer, 5¢-AAGGCCAAGAAGGAGTC CAAGAAG-3¢; L-FNR 2 (AB035645): forward primer, 5¢-TTGCTTGAGCTGAACAATACAATGAA-3¢; reverse primer, 5¢-GAGCCGGTCAAGAAGCTGGAG-3¢ PCRs were carried out using 1 : 5, 1 : 10, 1 : 20 and 1 : 40 dilutions
of cDNA Reactions were hot started at 95C, and carried out for 32 cycles of 94C for 30 s, annealing temperature for
1 min and 30 s and 72C for 30–90 s Products were visual-ized by ethidium bromide in agarose gels, and bands were quantified by scanning with an FLA-5000 imaging system (Fujifilm SAS, St-Quentin, France)
In situ hybridization experiment Tissue inclusion
Leaf sections were harvested at 2–3 h into the light phase, and immediately fixed in 4% (v⁄ v) paraformaldehyde con-taining 0.1% Triton X-100 in NaCl⁄ Pi(10 mm sodium phos-phate, pH 7.0 and 130 mm NaCl) After dehydration in a