By using blue native BN⁄ PAGE and Tricine ⁄ PAGE or colorless-native CN ⁄ PAGE, BN⁄ PAGE and Tricine ⁄ PAGE, combined with mass spectrometry, we attempted to obtain more information abou
Trang 1complex from mesophyll and bundle sheath chloroplasts
of Zea mays
Costel C Darie1, Martin L Biniossek2, Veronika Winter3, Bettina Mutschler4and
Wolfgang Haehnel4
1 Brookdale Department Molecular, Cell and Developmental Biology, Mount Sinai School of Medicine, New York, USA
2 Institut fuer Molekulare Medizin und Zellforschung, Albert-Ludwigs Universitaet, Freiburg, Germany
3 De´partement de Biologie Mole´culaire, Universite´ de Gene`ve, Switzerland
4 Institut fuer Biologie II⁄ Biochemie der Pflanzen, Albert-Ludwigs Universitaet, Freiburg, Germany
Complex I is a proton-pumping multisubunit-complex
involved in the respiratory electron transport chain,
which provides the proton motive force essential for
the synthesis of ATP Homologs of this complex exist
in bacteria, the mitochondria of eukaryotes, and the chloroplasts of plants The bacterial and mitochondrial
Keywords
chloroplast; maize; mass spectrometry;
native electrophoresis; Ndh complex
Correspondence
C C Darie, Brookdale Department of
Molecular, Cell and Developmental Biology,
Annenberg Building, Box1020, Mount Sinai
School of Medicine, One Gustave L Levy
Place, New York, NY 10029-6574, USA
Fax: +1 718 246 2616
Tel: +1 212 241 8620
E-mail: costel.darie@mssm.edu
(Received 31 January 2005, revised 23
March 2005, accepted 24 March 2005)
doi:10.1111/j.1742-4658.2005.04685.x
Complex I (NADH: ubiquinone oxidoreductase) is the first complex in the respiratory electron transport chain Homologs of this complex exist in bacteria, mitochondria and chloroplasts The minimal complex I from mitochondria and bacteria contains 14 different subunits grouped into three modules: membrane, connecting, and soluble subcomplexes The com-plex I homolog (NADH dehydrogenase or Ndh comcom-plex) from chloroplasts from higher plants contains genes for two out of three modules: the mem-brane and connecting subcomplexes However, there is not much informa-tion about the existence of the soluble subcomplex (which is the electron input device in bacterial complex I) in the composition of the Ndh com-plex Furthermore, there are contrasting reports regarding the subunit composition of the Ndh complex and its molecular mass By using blue native (BN)⁄ PAGE and Tricine ⁄ PAGE or colorless-native (CN) ⁄ PAGE,
BN⁄ PAGE and Tricine ⁄ PAGE, combined with mass spectrometry, we attempted to obtain more information about the plastidal Ndh complex from maize (Zea mays) Using antibodies, we detected the expression of a new ndh gene (ndhE) in mesophyll (MS) and bundle sheath (BS) chloro-plasts and in ethiochloro-plasts (ET) We determined the molecular mass of the Ndh complex (550 kDa) and observed that it splits into a 300 kDa mem-brane subcomplex (containing NdhE) and a 250 kDa subcomplex (contain-ing NdhH, -J and -K) The Ndh complex forms dimers at 1000–1100 kDa
in both MS and BS chloroplasts Native⁄ PAGE of the MS and BS chloro-plasts allowed us to determine that the Ndh complex contains at least 14 different subunits The native gel electrophoresis, western blotting and mass spectrometry allowed us to identify five of the Ndh subunits We also pro-vide a method that allows the purification of large amounts of Ndh com-plex for further structural, as well as functional studies
Abbreviations
BN⁄ PAGE, blue native ⁄ PAGE; BS, bundle sheath; CN ⁄ PAGE, colorless native ⁄ PAGE; ET, ethioplasts; MS, mesophyll; Ndh complex, NADH-dehydrogenase or NADH plastoquinone oxidoreductase; PS, photosystem.
Trang 2complexes function as NADH dehydrogenase (NADH:
ubiquinone oxidoreductase) [1]
The minimal complex I from mitochondria and
bacteria contains 14 polypeptides The mitochondrial
complex I contains additional subunits with no
counterparts in the bacterial complex In bacteria, the
14 subunits of the complex I are grouped into three
modules: seven subunits form the membrane
subcom-plex, four subunits form the connecting subcomplex
and the last three subunits form the soluble
subcom-plex The soluble subcomplex contains the
NADH-binding and -oxidizing site [1]
Genes for 11 of the 14 minimal subunits were also
found in the plastid genome of plants The 11 ndh
genes on the plastid genome that encode subunits
homologous to those of the NADH dehydrogenase or
complex I of mitochondria and bacteria are highly
conserved in most plants Their function as a
proton-pumping NADH, plastoquinone oxidoreductase
(NADH dehydrogenase or Ndh complex), has been
suggested [1]
A structural model indicates that the plastid ndhA–
ndhG gene products form the membrane subcomplex,
and the ndhH–ndhK gene products form a connecting
subcomplex that probably mediates the electron
trans-fer from NAD(P)H Subunits homologous to the three
peripheral subunits (from the soluble subcomplex) of
the NADH-oxidizing domain are likely encoded in the
nucleus, but have not been identified so far
Although most groups studying the Ndh complex
agree upon its molecular mass as 550–580 kDa [2–7],
other groups have reported detection of the Ndh
com-plex with a molecular mass between 800 and 1000 kDa
[8,9]
To date, not all 11 Ndh polypeptides encoded by
ndh genes have been identified in plastids Only the
polypeptides corresponding to seven out of the 11
genes have been identified in chloroplasts Four of
them are components of the connecting subcomplex:
NdhJ [6], NdhH [10], NdhK [11] and NdhI [12] From
the other seven polypeptides that form the membrane
subcomplex, only three have been identified: NdhA
[13], NdhB [6] and NdhF [14] Nothing is known
about the expression of ndhC, -D, -E and -G in
chloro-plast, nor in any plastid type
The C4 plants Sorghum bicolor and Zea mays have
mesophyll (MS) and bundle sheath (BS) chloroplasts
The chloroplasts of the MS cells contain grana, but
those in the BS cells have a variable degree of grana
development, depending on the species [15] The grana
from MS and BS chloroplasts exhibit normal
photosys-tem (PS) I activity, but the agranal BS thylakoids have
almost no PS II activity [16,17] However, transcription
of the ndh genes is much higher in BS chloroplasts, and elevated amounts of the Ndh complex have been found
in these plastids [18]
The function of the Ndh complex is still a matter of debate Some authors have proposed that in chloro-plasts the Ndh complex is involved in cyclic electron transport around PS I [18–22] Other authors have suggested a second role for this complex in chlorores-piration [3–5,23,24] However, controversial reports about the viability of ndh mutants [2,23,25] have clearly restarted the debate about the real function(s)
of the Ndh complex
To contribute to the structural and functional char-acterization of this large complex in chloroplasts, we produced antibodies against Ndh subunits from the membrane (NdhE) and connecting (NdhH, -J and -K) subcomplexes NdhE antibodies were used as markers for the presence of the membrane subcomplex, while NdhH, -J and -K antibodies were used to identify the connecting subcomplex
Using these antibodies, we detected the expression of
a new Ndh subunit (NdhE from the membrane subcom-plex) in maize MS, BS and ethioplast (ET) plastids By using (1) blue-native (BN)⁄ PAGE and (2) colorless-native (CN)⁄ PAGE and BN ⁄ PAGE, we separated the Ndh complex from both MS and BS chloroplasts and determined its monomeric and dimeric state We also demonstrated that the Ndh complex splits into a 300-kDa membrane subcomplex (containing NdhE) and
a 250-kDa subcomplex (containing NdhH, -J and -K)
By separating the Ndh complex that resulted from native electrophoresis in denaturing Tricine⁄ PAGE, we determined that the Ndh complex contains at least 14 different subunits, five of which were identified by Ndh antibodies and mass spectrometry
Results
Detection of a new expressed Ndh protein
In higher plants it has been demonstrated that seven (ndhA, -B, -F, -H, -I, -J, -K) of 11 ndh genes are expressed in plastids, but there is no information about the expression of ndhC, -D, -E and -G Here we report that a new Ndh protein (NdhE) is expressed in three different plastid types In Fig 1A, the protein pattern
of the MS, ET and BS plastids is shown In Fig 1B, the three plastid types were separated on SDS⁄ PAGE, electroblotted and immunodecorated with Ndh anti-bodies The Ndh antibodies detected polypeptides with a molecular mass of 46 (NdhH), 28 (NdhK),
18 (NdhJ) and 12 (NdhE) kDa, in agreement with their theoretical mass, calculated from their DNA
Trang 3sequences The NdhE antibodies recognized their
cor-responding antigens in three different plastid types:
MS and BS chloroplasts, as well as in ET Because of
high homology between NdhE from maize and rice,
this polypeptide was also identified by cross-reaction
of NdhE antibodies in rice chloroplasts, a C3 type
plant (data not shown) This is the first demonstration
that the ndhE gene produces a stable NdhE
polypep-tide in different types of plastids
The Ndh complex associates in homodimers and dissociates in membrane and soluble subcomplexes; isolation of the Ndh complex
by BN/ PAGE and Tricine/ PAGE Intact MS and BS chloroplasts were used as starting material for the separation of the Ndh complex
BN⁄ PAGE (which separates the protein complexes based on their molecular mass), of the BS chloroplasts revealed eight (five dominant and three minor) bands with molecular masses in the range of 750–200 kDa (Fig 2A left, the horizontal gel lane) To identify the Ndh complex, the gel strips that resulted from the first native dimension (1D) were separated in a second dimension (2D) under denaturing and reducing condi-tions (Tricine⁄ PAGE), electroblotted and immunodeco-rated with Ndh antibodies (Fig 2A) NdhH, -K and -J antibodies were markers for the Ndh connecting sub-complex, while NdhE antibodies were markers for the Ndh membrane subcomplex The Ndh antibodies recognized the 46, 28, 18 and 12 kDa polypeptides, cor-responding to the NdhH, -K, -J and -E subunits of the Ndh complex
The intact Ndh complex, which corresponded to the third band in BN⁄ PAGE, showed a molecular mass of 520–550 kDa (Fig 2A, left) and contained at least 14 visible subunits with a molecular mass in a range of 10–80 kDa (Fig 2B, left) Ndh antibodies also reacted with their antigens which were part of a 300–320-(NdhE) or 250-kDa complexes (NdhH, -K, and -J),
MS ET BS MS ET BS
kDa
97-
66-
45-
30-
20-
14-Ndh -H -K
-J -E
Fig 1 SDS ⁄ PAGE of the MS and BS chloroplasts and ET of maize.
After electrophoresis, the gels were stained with Coomassie blue
(A) or transferred onto membrane and probed with Ndh antibodies
(B) The polypeptide pattern of the maize plastids and the molecular
mass markers are shown in (A) The polypeptides detected by Ndh
antibodies are shown in (B) (right).
A
B
Fig 2 Separation of the MS and BS
chloro-plasts by BN ⁄ PAGE The horizontal gel lane
represents the first BN ⁄ PAGE dimension
(1D) After reduction and denaturation, the
gel lane was separated in the second
dimen-sion Tricine ⁄ PAGE (2D) The molecular
mass standards and direction of migration
are indicated for both the first (1D) and
sec-ond (2D) dimension (A) Tricine ⁄ PAGE gels
that resulted in 2D were electroblotted and
immunodecorated with Ndh antibodies The
position of the immune reaction for the Ndh
antibodies is indicated on the side of the
blots The experiments were performed in
identical conditions using BS chloroplasts,
except that the running time in the 1D right
was longer (B) Tricine ⁄ PAGE gels that
resu-lted in 2D were silver stained for the protein
pattern of the protein complexes The
mate-rial used was BS (left) and MS (right)
chloro-plasts The position of the Ndh complex in
both gels is indicated.
Trang 4suggesting that the Ndh complex splits into a
mem-brane (300 kDa) and a soluble (250 kDa) subcomplex
(Fig 2A, left)
Similar results were also obtained when MS
chloro-plasts were used as starting material The intact Ndh
complex had the same molecular mass (520–550 kDa)
(Fig 2A, right), and split into membrane (300 kDa)
and soluble (250 kDa) subcomplexes (data not shown)
In addition, Ndh antibodies detected the Ndh complex
with a molecular mass of 1000–1100 kDa, suggesting
that it exists in a dimeric form (Fig 2A, right)
How-ever, due to the low amounts of the Ndh complex in
MS chloroplasts, its polypeptide pattern could not be
observed in the second dimension of the BN⁄ PAGE
(Fig 2B, right)
The Ndh complex monomer (520–550 kDa), Ndh
com-plex dimer (1000–1100 kDa) and the 300- and 250-kDa
subcomplexes were also observed in sucrose gradient
and BN⁄ PAGE, as well as anion exchange and gel
fil-tration experiments (data not shown) Taken together,
these data suggest that the Ndh complex exists as a
monomer and dimer and splits into a membrane and
soluble subcomplexes
Separation of the Ndh complex by CN/ PAGE
(1D) and Tricine/ PAGE (2D) and CN/ PAGE (1D)
and BN/ PAGE (2D)
When the subunit composition of a protein complex is
investigated, one problem that can occur in BN⁄ PAGE
is that two protein complexes with identical molecular
mass may migrate together To further confirm the molecular mass and the number of subunits of the Ndh complex obtained by BN⁄ PAGE (at least 14 sub-units), colorless native PAGE (CN⁄ PAGE; in which separation of the protein complexes is based on their internal charge) was used as a prepurification step For location and isolation of the Ndh complex, BS chloroplasts were first separated on CN⁄ PAGE (1D) and Tricine⁄ PAGE (2D) (Fig 3A) The polypeptide pattern of the protein complexes from BS chloroplasts
is shown in Fig 3A (left)
To detect the Ndh complex, the gel that resulted in 2D was electroblotted and incubated with Ndh anti-bodies (Fig 3A, right) The polypeptides that reacted with Ndh antibodies were part of a protein complex, and corresponded to the second intense band in the 1D CN⁄ PAGE (Fig 3A, right) Similar results were obtained using MS chloroplasts (data not shown)
To further localize, isolate, and characterize the Ndh complex, the MS and BS chloroplasts were separated
on CN⁄ PAGE (1D, based on the internal charge of the protein complexes) and BN⁄ PAGE (2D, based on the external charge of the protein complexes and according to their molecular mass) (Fig 3B) Based on previous results, in BN⁄ PAGE (as a second dimension, 2D), the Ndh complex should correspond with the second intense band from CN⁄ PAGE; based on
BN⁄ PAGE (as a first dimension, 1D) results, it should have a molecular mass of 520–550 kDa It should be located in a square containing the three protein com-plexes marked a, b and c (Fig 3B) Indeed, western
1D
-Ndh H -Ndh K -Ndh J -Ndh E
kDa
14-1D
kDa
2D
-669
-440 -232
232-
440-
kDa
a
a
2D
A
B
Fig 3 Separation of the MS and BS chloro-plasts by two-dimensional CN ⁄ PAGE (A) The BS chloroplasts were separated in
CN ⁄ PAGE (1D, the horizontal bands) and then in denaturing and reducing conditions
by Tricine ⁄ PAGE (2D) The gel that resulted from 2D was silver stained (left) or electro-blotted and immunodecorated with Ndh antibodies (right) The polypeptides detected
by Ndh antibodies are shown on the right (B) The MS (left) and BS (right) chloroplasts were separated in CN ⁄ PAGE (1D, the hori-zontal bands) and then in nondenaturing conditions by BN ⁄ PAGE (2D) The molecular mass standards and direction of migration are indicated for both the first (1D) and second (2D) dimension The position of the Ndh complex was detected in the square containing the protein complexes marked (a) (b) and (c).
Trang 5blotting and silver staining of the in BN⁄ PAGE (as a
second dimension, 2D) gels confirmed again that the
molecular mass of the Ndh complex is 520–550 kDa
(data not shown) In order to increase the amount of
the Ndh complex for further analysis of its polypeptide
pattern, a three-dimensional preparative isolation of
the Ndh complex was performed
Three-dimensional preparative isolation of the
Ndh complex from MS chloroplasts: CN/ PAGE
(1D), BN/ PAGE (2D) and Tricine/ PAGE (3D)
Based on the results provided by CN⁄ PAGE,
BN⁄ PAGE and western blotting experiments, the
thyl-akoid membranes were separated on CN⁄ PAGE (1D)
and the second intense band containing the Ndh
com-plex was excised and further separated on BN⁄ PAGE
(2D) The BN⁄ PAGE band containing the Ndh
com-plex was excised, reduced, denatured and further
separ-ated on Tricine⁄ PAGE (3D) The resulting Tricine ⁄
PAGE (3D) gel was further divided into three pieces;
two of them were stained with silver or Coomassie
blue The third piece was electroblotted and
immunodecorated with Ndh antibodies A
computer-assisted reconstitution of the initial gel is shown in
Fig 4A Both the silver and CBB stained gel pieces
revealed at least 14 visible polypeptides with molecular
masses between 10 and 80 kDa, confirming the results obtained by BN⁄ PAGE (1D) and Tricine ⁄ PAGE (2D)
of the BS chloroplasts (Fig 2B) These experiments suggest that the Ndh complex contains at least 14 sub-units, four of them (NdhH, -K, -J and -E) identified by Ndh antibodies (Fig 4A) The tentative assignment of the Ndh subunits (based on their theoretical molecular mass and western blotting results) is shown in Fig 4B
Analysis of the Ndh subunits by mass spectrometry
To confirm the results obtained by Ndh antibodies, the gel bands that resulted from BN⁄ PAGE (1D) and Tri-cine⁄ PAGE (2D) or from CN ⁄ PAGE (1D), BN ⁄ PAGE (2D) and Tricine⁄ PAGE (3D), and which correspon-ded to Ndh subunits, were further analyzed by mass spectrometry (MALDI-TOF-MS) The mass spectro-metry measurements were submitted to the MASCOT database, as described in the Methods section To avoid obtaining false positive data, our search parame-ters were reduced to only one fixed modification (carbamidomethyl-cysteine), one variable modification (methionine-sulphoxide), a maximum of one missed cleavage and 100 p.p.m mass tolerance The mass spectra with the identified Ndh polypeptides are shown
in Fig 5
SS
-H
-K -J
-E
Ndh
kDa 97-
66-
45- 30-
20-
14-CBB
NdhB NdhD
NdhC
NdhA NdhF
NdhG NdhI
75 kDa
51 kDa
23 kDa
?
?
?
WB CBB
Fig 4 Three-dimensional isolation of the Ndh complex from MS chloroplasts (A) The MS chloroplasts were separated on CN⁄ PAGE (1D) and the band corresponding to the Ndh complex was excised and run in a second BN ⁄ PAGE (2D) The gel piece containing the Ndh complex was further separated on Tricine ⁄ PAGE (3D) The Tricine ⁄ PAGE gel was divided in three pieces and two of them were silver- (SS) and Coomassie- (CBB) stained The third gel piece was electroblotted (WB) and immunodecorated with Ndh antibodies (indicated on the right).
On the left, the apparent molecular mass is shown (kDa) (B) Tentative assignment of the Ndh subunits in the CBB stained Tricine ⁄ PAGE gel piece already shown in Fig 4A The theoretical molecular mass of the Ndh subunits is: NdhA (40 kDa) NdhB (56 kDa), NdhC (14 kDa) NdhD (56 kDa), NdhE (12 kDa), NdhF (83 kDa), NdhG (18 kDa), NdhH (45 kDa), NdhI (21 kDa), NdhJ (18 kDa), NdhK (29 kDa) The 75, 51 and 23 kDa subunits are assigned as the soluble subcomplex of the Ndh complex The unassigned polypeptides with a molecular mass of
10, 16 and 22 kDa are marked with question marks (?).
Trang 6The MASCOT database search of a trypsin-digested
gel band detected by NdhH antibodies identified the
46-kDa maize NdhH polypeptide with seven peptides
matched, two of them with one missed cleavage The
marked peaks with m⁄ z of 1352.75 (calculated, 1352.75),
1045.59 (calculated, 1045.54), 1832.88 (calculated, 1832.84), 1315.65 (calculated, 1315.66), 1568.69 (calcula-ted, 1568.71), 1937.03 (calcula(calcula-ted, 1936.97) and 1946.25 (calculated, 1946.19) from the mass spectrum from Fig 5A corresponded to peptides SIIQYLPYVTR,
A
B
C
Fig 5 Analysis of the Ndh subunits by MALDI-TOF-MS Spectra of NdhH (A), NdhI (B) and NdhJ (C) are shown The identified peptides in each spectrum are marked with an asterisk The monoisotopic peaks represent the mass ⁄ charge (m ⁄ z) ratio.
Trang 7ASGIQWDLR, KIDPYESYNQFDWK, IPGGPYEN
LEAR, AKNPEWNDFEYR (one missed cleavage),
GELGIYLVGDDSLFPWR and
IRPPGFINLQILP-QLVK (one missed cleavage) All of these peptides were
part of the maize NdhH subunit
Database analysis of a MALDI-TOF-MS
measure-ment of a 21-kDa trypsin-digested gel band (indicated
NdhI in Fig 4B) identified the maize NdhI
poly-peptide with a molecular mass of 21 kDa, with five
peptides matched, one of them with an oxidized
methi-onine The peaks with m⁄ z of 1749.83 (calculated,
1749.92), 1373.62 (calculated, 1373.74), 1482.66
(cal-culated, 1482.77), 1457.66 (cal(cal-culated, 1457.74) and
1736.82 (calculated, 1736.92) from the mass spectrum
from Fig 5B corresponded to peptides YIGQSFII
(cysteine modified by iodoacetamide:
carbamidometh-yl-cysteine), HELNYNQIALSR and LPISIMGDY
TIQTIR (methionine oxidized to
methionine-sulphox-ide) identified as part of maize NdhI (21 kDa)
Finally, a MASCOT database search of a
trypsin-digested gel band detected by NdhJ antibodies
identi-fied the 18-kDa maize NdhJ polypeptide, with four
matched peptides (one of them with one missed
clea-vage), which covered 28% of the protein The peaks
with m⁄ z of 1189.71 (calculated, 1189.64), 2616.25
(cal-culated, 2616.13), 1782.80 (cal(cal-culated, 1782.77) and
1725.86 (calculated, 1725.77) from the mass spectrum
from Fig 5C corresponded to peptides IPSVFWVWR,
cleavage), ESYDMVGISYDNHPR, DYITPNFYEIQ
DAH, which were part of the 18-kDa maize NdhJ
protein
By using these 2D (BN⁄ PAGE and Tricine ⁄ PAGE)
and 3D (CN⁄ PAGE, BN ⁄ PAGE and Tricine ⁄ PAGE)
experiments, we were able to determine that the Ndh
complex contains at least 14 subunits, some of them
identified by Ndh antibodies Moreover, by combining
the gel electrophoresis methods with mass
spectrome-try, we were able to identify five (NdhE-, H, -I, -J, and
-K) out of 14 Ndh subunits
Discussion
The minimal bacterial complex I homologous to the
chloroplast Ndh complex contains 14 subunits (Nuo
A–N), with a molecular mass of 550 kDa Seven
sub-units form a membrane subcomplex and four subsub-units
form a connecting subcomplex The remaining three
subunits form a soluble subcomplex, which harbors
the binding and oxidation site for NADH [1,26,27]
The maize chloroplast contains 11 ndh genes
enco-ding 11 polypeptides (NdhA–K) [28] Seven subunits
(NdhA–G) form the membrane subcomplex, while the remaining four subunits (NdhH–K) form the connect-ing subcomplex, both of them homologous to the bac-terial subcomplexes The theoretical molecular mass of the Ndh complex, calculated from its 11 subunits, is almost 400 kDa NdhA–G (the membrane subcom-plex) accounts for 290 kDa and NdhH–K (the con-necting subcomplex) accounts for 110 kDa
Previous reports regarding the molecular mass of the Ndh complex are ambiguous While some groups reported that the molecular mass of this complex is 550–580 kDa [2–7], other groups have reported detec-tion of the Ndh complex with a molecular mass between 800 and 1000 kDa [8,9]
By using BN⁄ PAGE and Tricine⁄ PAGE or
CN⁄ PAGE, BN ⁄ PAGE and Tricine ⁄ PAGE, we found that the molecular mass of the Ndh complex from both MS and BS chloroplasts is 550 kDa In addition,
we found that the Ndh complex (from both MS and
BS chloroplasts) may be in monomeric (550 kDa) as well as in dimeric form (1000–1100 kDa) Similar results were obtained with the semipurified Ndh com-plex isolated by anion exchange followed by gel filtra-tion, or by sucrose gradient combined with BN⁄ PAGE (data not shown) Our results confirm some previously reported results [2–7], but disagree with other reports [8,9], and suggest that the 1000-kDa Ndh complex des-cribed by these groups was actually a dimeric form The observation that the molecular mass of the Ndh complex monomer is similar in both MS and BS chloroplast types led us to hypothesize that the archi-tecture of the Ndh complex in other plastid types is similar
We also found that the Ndh complex splits into a 300-kDa subcomplex (corresponding to the membrane subcomplex, detected by NdhE antibodies) and a 250-kDa subcomplex (detected by NdhH, -J and -K antibodies) The 250-kDa subcomplex contains NdhH, -I, -J and -K subunits However, the theoretical mole-cular mass of these subunits is 110 kDa, suggesting that the difference up to 250 kDa may be the electron input module (the soluble subcomplex), which in the bacterial Ndh complex contains three polypeptides (75,
51 and 23-kDa subunits) Alternatively, the 250-kDa subcomplex contains two copies of NdhH, -I, -J, and -K, as suggested [29]
By using BN⁄ PAGE and Tricine ⁄ PAGE, we deter-mined that the Ndh complex contains a minimum of
14 subunits (with a molecular mass between 10 and
70 kDa) Some of them were detected by Ndh anti-bodies and some by mass spectrometry Similarly, using CN⁄ PAGE, BN ⁄ PAGE, and Tricine ⁄ PAGE, we also determined that the Ndh complex contains at
Trang 8least 14 subunits, partly detected by Ndh antibodies
and mass spectrometry Since there are only 11
plas-tidal ndh genes, this suggests that the detected extra
proteins are encoded in the nucleus, and may
repre-sent the electron input module, similar with the
bac-terial 75, 55 and 23-kDa homologue Indeed, Quiles
and his colleagues [7,9], reported that in oat and
bar-ley, the Ndh complex contains one nuclear-encoded
polypeptide homologous to bacterial 55-kDa protein
Later, the same group [30], compared the plastidal
Ndh complex and mitochondrial Complex I from the
same plant (barley) by western blotting, and reported
that both complexes contained the electron input
module (the soluble subcomplex) containing
polypep-tides homologous to bacterial 24, 51, and 75-kDa
pro-teins Quiles and colleagues [30] also suggested that
these nuclear gene products could contain a dual
targeting sequence, which allows them to be targeted
to both mitochondria and chloroplasts However,
these reports were based only on western blotting
experiments and to confirm this statement,
identifi-cation of these polypeptides in further studies will be
necessary
Based on the number and the molecular mass of
the polypeptides obtained by BN⁄ PAGE and Tricine ⁄
PAGE or CN⁄ PAGE, BN ⁄ PAGE and Tricine ⁄ PAGE,
we suggest that the 250-kDa subcomplex does not
contain contains two copies of NdhH, -I, -J, and -K
polypeptides, as previously reported [29], but contains
the soluble subcomplex (the electron input device)
Our results could explain why the theoretical
molecular mass of the Ndh complex is 400 kDa and its
determined mass is 520–550 kDa These data could
also explain why the experimental number of Ndh
polypeptides [determined by (1) BN⁄ PAGE and
Tri-cine⁄ PAGE and (2) CN ⁄ PAGE, BN ⁄ PAGE and
Tri-cine⁄ PAGE in both MS and BS chloroplasts] is at
least 14, despite the theoretical number of the encoding
genes Moreover, the number of Ndh subunits
observed in plastids from different plants is close to
our number: at least 15 polypeptides in oat [7], 16
polypeptides in pea [4] and 14 polypeptides in maize
[31] We also provide evidence that the number of the
polypeptides from the Ndh complex in MS and
C3-type chloroplasts is similar to Ndh proteins from
BS chloroplasts
Recently, Promeenade et al [29], found that the
cy-anobacterial Ndh complex contains two extra subunits
(slr1623 and sll1262), unrelated to the subunits of the
minimal bacterial complex I, but homologous to two
nuclear-encoded maize Ndh proteins, as detected in an
Ndh complex preparation by Funk et al [31] If we
calculate that the Ndh complex contains 11
plastidal-encoded subunits, three (mitochondrial-related) nuc-lear-encoded subunits (the soluble subcomplex), and two (cyanobacterial-related) nuclear-encoded subunits (slr1623 and sll1262), we should conclude that the minimal Ndh complex from higher plants contains at least 16 subunits
MALDI-TOF-MS [32,33] is a useful tool for the analysis and identification of proteins [34–36] Unfor-tunately, most groups that have tried to assign the polypeptide pattern of the plastidal Ndh complex failed because of the low protein yield obtained for further protein analysis Although we overcame this problem, we were still unsuccessful in the assignment
of all Ndh subunits, probably because of the technical difficulties inherent in assigning the highly hydropho-bic membrane subunits
It should be mentioned that the electron input device of the Ndh complex from the chloroplast could
be different than the corresponding one from cyano-bacteria, since the last one contains subunits with a molecular mass between 9 and 50 kDa [29], compared with the Ndh subunits, which have molecular masses between 10 and 70 kDa In addition, cyanobacteria contain the protein complexes for both the photosyn-thetic and respiratory functions on the same mem-brane
In conclusion, we demonstrated that one ndhE gene
is expressed in three different plastid types: MS and
BS chloroplasts and ET Furthermore, we were able to determine the molecular mass of the Ndh complex monomer (550 kDa) and dimer (100–1100 kDa) Also, the Ndh complex splits into a 300-kDa membrane sub-complex (containing NdhE) and a 250-kDa subcom-plex (containing NdhH, -J and -K) The 250-kDa subcomplex contains the connecting subcomplex (NdhH, -I, -J and -K) in a monomeric form and per-haps at least three nuclear-encoded subunits We were also able to determine that the Ndh complex contains
at least 14 different polypeptides (with a molecular mass between 10 and 70 kDa), five of them (NdhH, -I, -J, -K and -E) identified by western blotting and mass spectrometry The three-dimensional CN⁄ PAGE,
BN⁄ PAGE and Tricine ⁄ PAGE method we have des-cribed should allow the isolation of large amounts of pure Ndh complex from maize chloroplasts for further structural and functional studies This may include identification of further Ndh subunits, as well as deter-mination of the substrate specificity and function of the Ndh complex
We hope that further analysis of the Ndh subunits
by N-terminal sequence analysis or mass spectrometry (MALDI-TOF-MS and MS⁄ MS) will reveal the complete subunit composition of this incompletely
Trang 9characterized plastidal Ndh complex Providing more
sequence information will also give us more insights
about the function of the Ndh complex, and will be a
focus of future studies
Experimental procedures
Materials
Maize (Zea mays, L, Perceval, Deutsche Saatveredlung
pBluescript KS vector was from Stratagene The
pGEX-6P-2 vector and ECL immunoblotting kit were from
Amer-sham Pharmacia Biotech (Freiburg, Germany) The antisera
were raised in rabbits at Charles River Deutschland GmbH
(Sulzfeld, Germany) The Protean II cell for native PAGE
was from Bio-Rad (Munich, Germany) Poly(vinylidene
difluoride) membranes were from Immobilon-P (Millipore,
Billerica, MA, USA) Trypsin, horseradish peroxidase
con-jugated to secondary antibody (goat anti-rabbit IgG) and
all other chemicals were from Sigma-Aldrich (Munich,
Germany)
Plant material and isolation of maize chloroplasts
Maize seeds were grown in a green house at 24C during
an 18 h photoperiod of white light All experiments were
carried out with 14-day-old plants The leaves were
harves-ted 3–5 h after the beginning of the photoperiod Intact
MS and BS chloroplasts were isolated on a Percoll step
gradient as described [37] All procedures were performed
at 4C and all materials were also kept at this temperature
The leaves (13–18 g, second leaf, upper 5–10 cm) (after
excising the middle vascular system) were cut into small
pieces 3–5 mm and left for 2–4 h at 4C The leaves were
mixed with 100 mL buffer (350 mm sorbitol; 10 mm
EDTA; 1 mm MgCl2; 20 mm Hepes pH 8.0) and cut in a
mixer for 15 s at speed 3 and passed through a 600 lm
nylon mesh The solution contained MS chloroplasts and
was centrifuged for 5 min at 6000 r.p.m., at 4C, using a
GS3 rotor The resulting pellet was resuspended in the
same buffer, washed again and applied to a 40⁄ 80% Percoll
step gradient The lower band contained the intact MS
chloroplasts The procedure was similar for isolation of
ET, except that the maize was grown in complete darkness
The retained material was mixed two more times (the last
time only with 50 mL buffer and mixed 8 s) The
superna-tant from both steps was passed through a 100-lm nylon
mesh and the retained material was immersed in digestion
buffer [0.35 m sorbitol, 1 mm KH2PO4, 10 mm Mes⁄ KOH
pH 6.0, 0.3% (w⁄ v) macerozym, 2% (w ⁄ v) cellulase] and
shuttled for 45 min at 25C and 120 r.p.m The BS
chloro-plasts were released from the cells by mechanical treatment
and added to a preformed percoll gradient (30⁄ 80%),
followed by centrifugation for 10 min at 4500 r.p.m using
an HB4 rotor, at 4C The intact chloroplasts (lower band) were washed with 10 mL buffer for 2 min at 2000 r.p.m (HB4 rotor), and the pellet was used as a starting
SDS⁄ PAGE
Production of Ndh antibodies
Full-length cDNA was amplified by PCR and every ndh gene was cloned in the SmaI restriction site of pBluescript
KS, sequenced for PCR errors and transformed into XL1Blue competent cells After amplification of the vector containing the genes (Qiagen midi protocol), the ndh genes were excised with SalI and XhoI restriction enzymes, puri-fied and subcloned by T4 ligase into pGEX-6P-2 vector, already linearized with SalI and dephosphorylated with calf intestinal phosphatase The GST–Ndh fusion proteins were then overexpressed as inclusion bodies in Escherichia coli (BL21) The expression of the fusion proteins in bacteria was routinely monitored by western blotting using horse-radish peroxidase coupled to glutathione-S-transferase (GST) antibodies The expressed GST–Ndh proteins were then purified by preparative SDS⁄ PAGE, electroeluted from excised gel bands and used to raise antisera in rabbits
Native PAGE – CN/ PAGE and BN/ PAGE
CN⁄ BN ⁄ PAGE was carried out in the Protean II cell (Bio-Rad) following an in-house optimized protocol of the pub-lished method [38–40] Briefly, the starting material was run
on a separating gel (gradient 5–13% acrylamide) in both
CN⁄ PAGE and BN ⁄ PAGE Compared with CN ⁄ PAGE, in the cathode buffer of BN⁄ PAGE, Coomassie blue dye was used For transition from CN⁄ PAGE to BN ⁄ PAGE, the
CN⁄ PAGE band was excised, oriented horizontally and run in BN⁄ PAGE At first, a special cathode buffer was used (100 mm glycine, 20 mm bis⁄ Tris pH 8.1, 0.002% Coomassie blue) until the protein complexes ran out of the gel piece, followed by a substitution with the regular
BN⁄ PAGE cathode buffer
Tricine/ PAGE for the second/ third dimension
Gel lanes of the CN⁄ PAGE or BN ⁄ PAGE with separated protein complexes were excised and implanted horizontally
on denaturing gels for resolution of proteins in a second
dimension For the separation of proteins with low molecular mass, the Tricine⁄ SDS ⁄ PAGE was performed
as described [41] Before electrophoresis, the gel strips were incubated with denaturation solution [1% (w⁄ v) SDS⁄ 1% (v ⁄ v) 2-mercaptoethanol] for 2 h under moderate shaking
Trang 10SDS/ PAGE and immunoblot analysis
The isolated MS and BS chloroplasts were separated by
denaturing SDS⁄ PAGE [42] The gels were stained with
Coomassie dye or electroblotted onto poly(vinylidene
diflu-oride) membranes The immune reaction was performed by
incubation of the membranes with a mixture of primary
antibodies (NdhE, -H, -J and -K) in a dilution of 1 : 1000
Before a mixture of Ndh antibodies was used, their
individ-ual specificity was tested The horseradish peroxidase
conjugated to secondary antibody was used for
immunode-tection, performed by ECL immunoblotting kit according
to the manufacturer’s instructions
Enzymatic digestion of Ndh subunits
Digestion of gel pieces containing individual Ndh
polypep-tides with trypsin was carried as described [43] Gel pieces
containing Ndh subunits were incubated with 60% (v⁄ v)
acetonitrile for 20 min, dried completely in a SpeedVac
evaporator, and rehydrated for 10 min with digestion buffer
(25 mm ammonium bicarbonate, pH 8.0) This procedure
was repeated three times After drying, gel pieces were
again rehydrated in digestion buffer containing 10 mm
dithiothreitol and incubated for 1 h at 56C Following
reduction, cysteine residues were blocked by replacing the
dithiothreitol solution with 100 mm iodoacetamide in
25 mm ammonium bicarbonate pH 8.0, for 45 min at room
temperature with occasional vortexing Gel pieces were
dehydrated, dried, and rehydrated twice Dried gel pieces
were then digested overnight at 37C in digestion buffer
containing 15 ngÆlL )1 trypsin and 5 mm calcium chloride
When the digestion was complete, the peptides were
extrac-ted twice from gel pieces by addition of 300 lL of 60%
acetonitrile⁄ 5% formic acid (v ⁄ v ⁄ v) in 25 mm ammonium
bicarbonate, pH 8.0, and shaking for 60–90 min at room
temperature Solutions containing peptides of Ndh subunits
were dried and used for MALDI-TOF-MS in reflective
mode
MALDI-TOF-MS analysis of Ndh peptides
The measurements were carried out on a Reflex III TOF
system (Bruker Daltonics, Leipzig, Germany) reflectron,
equipped with a nitrogen laser (337 nm) The dry samples
were dissolved with 2% (v⁄ v) trifluoroacetic acid A
mix-ture of nitrocellulose and alpha cyano-4-hydroxycinnamic
acid was used as matrix After drying, samples were
ana-lyzed either undiluted or at suitable dilutions (1 : 10) After
ionization, the ions were measured in the mass⁄ charge
range of m⁄ z ¼ 700–3200 and time-to-mass conversion was
achieved by using external or internal calibration Peaks
detected by MALDI-TOF-MS corresponded to
monoiso-topic mass⁄ charge (m ⁄ z) peptides Calibration differences
for these measurements were generally under 100 p.p.m
Before analysis of the mass spectra, the peaks that resul-ted from trypsin autolysis (e.g m⁄ z 2163, 2185, 2273), as well as other peaks observed in the blank sample (peptides that resulted from trypsin digest of a gel piece containing
no protein) were subtracted
Analysis of the mass spectra was performed using pep-tidemass fingerprinting search algorithm from the MAS-COT search database (http://www.matrixscience.com) The parameters used in the MASCOT search were: database (NCBI, MSDB, OWL, Swissprot and EST), organism (all entries and Viridiplants), fixed modifications (carbamido-methyl), variable modifications (oxidation of methionine to methionine sulfoxide), missed cleavages (0 and 1) and mass tolerance (100 and 150 p.p.m.)
Acknowledgements
We thank Drs Alisa G Woods and Ike Woods (Padure Biomedical Consulting, Brooklyn, USA) for discussion and editing the manuscript and Dr Ch Peters and Patric Hoert (University of Freiburg, Ger-many) for providing the Reflex III for measurements and for the mass spectrometry measurements We also thank to the Deutsche Forschungsgemeinschaft for financial support, grant SFB388⁄ A1 Dr Costel C Darie received support from the Graduiertenkolleg
‘Biochemie der Enzyme’, Freiburg, Germany
References
1 Friedrich T, Steinmu¨ller K & Weiss H (1995) The pro-ton-pumping respiratory complex I of bacteria and mitochondria and its homologue in chloroplasts FEBS Lett 367, 107–111
2 Burrows PA, Sazanov LA, Svab Z, Maliga P & Nixon PJ (1998) Identification of a functional respiratory complex
in chloroplasts through analysis of tobacco mutants con-taining disrupted plastid ndh genes EMBO J 17, 868– 876
3 Guera A, De Nova PG & Sabater B (2000) Identifica-tion of the Ndh (NAD(P)H-plastoquinone-oxidoreduc-tase) complex in etioplast membranes of Barley: changes during photomorphogenesis of chloroplasts Plant Cell Physiol 41, 49–59
4 Sazanov LA, Burrows PA & Nixon PJ (1998) The plastid ndhgenes code for an NADH-specific dehydrogenase: isolation of a complex I analogue from pea thylakoid membranes Proc Natl Acad Sci USA 95, 1319–1324
5 Sazanov LA, Burrows PA & Nixon PJ (1998) The chloroplast Ndh complex mediates the dark reduction
of the plastoquinone pool in response to heat stress in tobacco leaves FEBS Lett 429, 115–118
6 Guedeney G, Corneille S, Cuine S & Peltier G (1996) Evidence for an association of ndhB, ndhJ gene products