Download Identification of a Chloroplast-encoded 9-kDa

Vol. 262, No. 26, Issue of September 15, pp. 12676-12684,1987
Printed in U.S.A.
THEJOURNAL
OF BIOLOGICAL
CHEMISTRY
0 1987 by The American Society for Biochemistry and Molecular Biology, Inc.
Identification of a Chloroplast-encoded9-kDa Polypeptide as a
2[4Fe-4S] Protein Carrying CentersA and B of Photosystem I*
(Received for publication, April 13, 1987)
Peter Bordier HejS, Ib Svendsed, HenrikVibe SchellerS, and Birger Lindberg
MellerS
From the $Department of Plant Physiology, Royal Veterinary and Agricultural University,
40 Thorvaldsemvej, DK-1871
Frederiksberg C and the §Department of Chemistry, Carlsberg Laboratory, 10 Gamk Carlsberg Vej, DK-2500 Valby, Denmark
An improved procedure is reported for large-scale
preparation of photosystem I (PS-I)vesicles from thylakoid membranes of barley (Hordeum uulgare L.).
The PS-I vesicles contain polypeptides of molecular
masses 82, 18, 16, 14, and9 kDa inan apparentmolar
ratio of 4:2:2:1:2. The 18-, 16-, and 9-kDa polypeptides were purified to homogeneity after exposure of
the PS-I vesicles to chaotropic agents. The isolated 9kDa polypeptide binds 65-70% of the zero-valence
sulfur of denatured PS-I vesicles, and the remaining
30-3570 is bound to P700-chlorophyll a-protein1.The
N-terminal amino acid sequence (29 residues) of the 9kDa polypeptide was determined. Comparison with the
nucleotide sequence of the chloroplast genome of Marchantiapolyrnorpha (Ohyama, K., Fukuzawa, H., Kohchi, T., Shirai, H., Sano, T., Sano, S., Umesono, K.,
Shiki, Y., Takeuchi, M., Chang, Z., Aota, S.-i., Inokuchi, H., and Ozeki, H. (1986) Nature 322, 572-574)
and of Nicotiana tabucum (Shinozaki, K., Ohme, M.,
Tanaka, M., Wakasugi, T., Hayashida, N., Matsubayashi, T., Zaita, W., Chunwongse, J., Obokata, J., Yamaguchi-Shinozaki, K., Ohto, C., Torazawa, K., Meng,
B. Y., Sugita, M., Deno, H., Kamogashira, T., Yamada,
K., Kusuda, J., Takaiwa, F., Kato, A., Tohdoh, N.,
Shimada, H., andSugiura, M. (1986) EMBO J. 5,
2043-2049) identified the chloroplast gene encoding
the 9-kDa polypeptide. We designate this gene psaC.
The complete amino acid sequence deduced from the
psaC gene identifies the 9-kDa PS-I polypeptide as a
2[4Fe-4S] protein. Since P700-chlorophyll a-protein1
carries center X, the 9-kDapolypeptide carries centers
A and B. A hydropathy plot permits specific identification of the cysteine residues which coordinate centers A and B, respectively. Except for the loss of the
N-terminal methionine residue, the primary translation product of the psaC gene is not proteolytically
processed.
P700-chlorophyll a-protein1binds 4 ironatoms and
4 molecules of acid-labile sulfide/molecule of P700.
Each of the two apoproteins of P700-chlorophyll aprotein 1 contains thesequence Phe-Pro-Cys-Asp-GlyPro-Gly-Arg-Gly-Gly-Thr-Cys(Fish, L. E., Kuck, U.,
and Bogorad, L. (1985) J. Biol. Chem. 260, 14131421). The stoichiometry of the component polypeptides of PS-I indicates the presence of four copies of
this sequence per molecule of P700. Center X may be
* This research was supported in part by grants from the Danish
Agricultural Research Council, the Danish Natural Science Research
Council, Dansk Investeringsfond, the Thomas B. Thriges Foundation,
the Carlsberg Foundation, the Tuborg Foundation, and Stiftelsen
Hofmansgave and by a Niels Bohr grant from the Royal Danish
Academy of Sciences and Letters. The costs of publication of this
article were defrayed in part by the payment of page charges. This
article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact.
composed of two [2Fe-2S] centersbound to the 8 cysteine residues contained in these foursegments.
The photochemical transferof electrons fromreduced plastocyanin to ferredoxin iscatalyzed by PS-I.’ The complex is
known to contain P700, the reaction center chlorophyll of
PS-I, and several electron acceptors which become reduced
upon illumination. The signals obtainedby a large number of
spectrophotophysicaltechniques (e.g. ESR and chemically
induced dynamicelectronpolarization)
haverevealed the
involvement of at least five different electron acceptors denoted A,, AI, X, B, and A (1-3). Whereas spectrophotophysical techniques have been very useful in detecting these acceptors in complex preparations, the same techniques are of
limited value in determining the identity
of the chemical
structures giving rise to the signals detected.As a result, the
chemical identity of P700 and centers A,, and AI still remains
unresolved (1-3). Mossbauer and ESR spectrometry indicated
that centers B and A are iron-sulfur centers (4-7). Based on
microwave power saturation studies, these two centers were
further assigned as [4Fe-4S] clusters (8).The ESR spectrum
of center X also had someresemblance to thoseof iron-sulfur
centers (9), and Mossbauer spectroscopy suggestedthat center
X could be a [4Fe-4S] center (10).However, microwave power
saturation studies indicated that centerX was not a typical
[4Fe-4S] or [2Fe-2S] center and that the spectrum possibly
represented a chlorophyll anion magnetically interacting with
iron (8).
An alternative approach to study PS-I is to characterize
the structural and functional
role of each of the PS-Ipolypeptides. Using this strategy, H0j and Merller (11)and Golbeck
(2) provided biochemical evidencedemonstrating that the82kDa polypeptides of P700-chlorophyll a-protein 1 bind an
iron-sulfur center, most likely center X. PS-I preparations
contain additional polypeptides of lower molecular mass (2,
3,11). Although the number reportedvaries,
four polypeptides
of approximate molecular masses 18,16,14, and9 kDa appear
to belong to the PS-I core (2,3, 11).The 18-kDapolypeptide
has beenclaimed to carry the iron-sulfur centers
A and B (12,
13). However, this identification was based solely on correlation between the gradualdepletion of the 18-kDapolypeptide
and the disappearance of centers A and B as monitored by
ESR spectroscopy. Malkin et al. (14) isolated an 8-kDa polypeptide from chloroplasts of spinach (Spinacia oleracea). The
The abbreviations used are: PS-I, photosystem I; DTT, 1,4dithiothreitol; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic
acid; Mes, 2-N-(morpholino)ethanesulfonicacid; ORF, open reading
frame; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; Tricine, N-[2-hydroxy-l,l-bis(hydroxymethyl)ethyl]glycine.
12676
A 9-kDa Protein of Photosystem I Carries Centers A and B
preparation
showed an ESR spectrum similar to that of a
ferrodoxin
but different from those of centers, A, B, and X.
It remains to be established
whether the reported
spectra
could have been derived from an artificial complex of similar
molecular mass and generated from cysteine, sulfide, and iron.
As demonstrated
by Hej and Meller (ll), the coincidental
comigration
of such artificially
formed oligomeric
complexes
with the 18 and 16-kDa polypeptides
was previously
interpreted to indicate that these two polypeptides
were binding
iron-sulfur
clusters (15, 16). Lagoutte et al. (17) used in uiuo
YS labeling and W-carboxymethylation
to demonstrate
that
a 9-kDa polypeptide
is the most cysteine-rich
component
in
PS-I preparations
obtained
from spinach and showed that
this polypeptide
contains approximately
8 cysteine residues.
Similar results pointing toward the 9-kDa polypeptide
as an
iron-sulfur
protein of PS-I have been obtained in labeling
studies with Anacystis
nidulans
(18). In contrast to these two
studies, Bonnerjea
et al. (12) found that a 30-40% depletion
of the 16-, 14-, and clO-kDa
polypeptides
from a spinach PSI preparation
did not result in a corresponding
decrease in
the ESR signals of centers A and B. These contradicting
results called for a more direct approach for the identification
of the apoprotein(s)
of centers A and B. In this study, we
have used acid-labile sulfide and zero-valence sulfur as specific
markers for intact and degraded iron-sulfur
clusters, respectively. The 18-, 16-, and 9-kDa polypeptides
were purified to
homogeneity
from PS-I vesicles of barley after exposure of
the vesicles to chaotropic agents. From its ability to bind zerovalence sulfur and from amino acid sequencing
data, the
purified
9-kDa polypeptide
is identified
as a chloroplastencoded, 2[4Fe-4S] protein that binds centers A and B.
MATERIALS
Unless
AND
METHODS
otherwise indicated, all procedures were carried out at 4 “C.
Preparation
of PS-I
the
9-kDa
polypeptide
but
12677
devoid
of P700-chlorophyll
a-protein
1
were combined and dialyzed against 5000 ml of 15 mM ammonium
acetate (pH 5.0) 0.05% (w/v) Triton X-100. After dialysis (conductivity, 1.4 millisiemens), the material was loaded onto a column (0.9 x
30 cm) of CM-Sepharose
(Pharmacia
Biotechnology
AB) equilibrated
in 25 mM ammonium
acetate
(pH 5.5), 0.05% (w/v)
Triton
X-100
and eluted with a linear gradient
of 25 mM ammonium
acetate
(pH
5.5), 0.05% (w/v) Triton
X-100 and 400 mM ammonium
acetate (pH
8.0), 0.05% (w/v) Triton
X-100
(2 X 120 ml) at a flow rate of 10 ml/
h. The pH of the two buffers
were adjusted
with HOAc
and NH3,
respectively.
Eluted fractions
were again assayed for their content
of
acid-labile
sulfide and zero-valence sulfur (20) and for their polypep-
tide content.
Fractions
rich in the 9-kDa polypeptide
were combined,
lyophilized,
dissolved
in a minimal
volume of AcA buffer, and finally
applied
to a column
(1.6 x 95 cm) of AcA 34 equilibrated
in AcA
buffer and eluted with the equilibration
buffer at a flow rate of 2 cm/
h. Fractions
judged by SDS-PAGE
to contain
pure 9-kDa polypeptide
were combined,
lyophilized,
and stored at -80 “C.
18- and 16-kDa Polypeptides-PS-I
vesicles (146 ml) obtained
from
the AcA column
were concentrated
to a volume of 10 ml, diluted with
120 ml of 20 mM imidazole
HCI (pH 7.6), and concentrated to 17 ml
by ultrafiltration.
Solid urea (7 g) was added to the concentrated
PSI vesicles
(16 ml, 1.20 mg of chlorophyll/ml),
and the material
was
applied
(25 ml/h) to a column
(0.9 x 46 cm) of Polvbuffer
Exchanger
94-(Pharmacia
Biotechnology
AB) equilibrated
in 20 mM imidazole
(pH 7.6), 5 M urea. Washing
of the column with 160 ml of equilibration
buffer
served
to elute the major
part of the 18-kDa
polypeptide
originally
bound to PSI. The column
was subsequently
washed-with
75 ml of 25 mM imidazole
HCl (oH 7.6) and then with 60 ml of 25
mM imidazole
HCl (pH 7.6), 0.1% (w/v) Triton
X-100,0.05%
Empigen BB (Albright
& Wilson
Ltd.,
Marchon
Works,
Whitehaven,
Cumbria.
United
Kinadorn)
which eluted some of the 16-kDa
DOIVpeptide
in a homogenous
form. P700-chlorophyll
a-protein
1, the 9and 14-kDa
polypeptides,
and some of the 16-kDa polypeptides were
eluted by applying
‘I-fold diluted
Polybuffer
74/HCl
(pH 4.5), 0.1%
(w/v)
Triton
X-100 to the column.
The fractions
containing
P700chlorophyll
a-protein
1 were concentrated
by ultrafiltration
and
treated
with NaSCN
as described
above for the 9-kDa polypeptide.
These manipulations
allowed
the isolation
of a particle
which
contained
only P700chlorophyll
a-protein
1 and the 14- and 9-kDa
polypeptides.
Vesicles
Chloroplasts
of barley
(Hordeurn
v&are
L.) were isolated
and
osmotically
lysed as described
in Ref. 19. The lamellar
systems
(200400 mg of chlorophyll)
were resuspended
in 20 mM Hepes (pH 6.3),
15 mM NaCl, 5 mhi MgCl, at 2.0 mg of chlorophyll/ml.
Triton
X-100
(25 mg of 100% Triton
X-lOO/mg
of chlorophyll)
was added, and
after stirring
for 30 min in the dark, a P700-enriched
supernatant
was obtained
by centrifugation
at 48,000 X g for 30 min. The super-
Polyacrylamide
Gel Electrophoresis
to a column
(5 x 20 cm) of DEAE-Sepharose
CL-6B
(Pharmacia
Biotechnology
AB, Uppsala,
Sweden)
equilibrated
in the same buffer.
The column
was washed with 1 column
volume of equilibration
buffer
containing
60 mM NaCl.
The PS-I vesicles
were-eluted using the
eauilibration
buffer fortified
with a 60-500
mM linear
NaCl aradient
Analytical
SDS-PAGE
was carried out in slab gels at 6 “C according
to Fling and Gregerson
(21) with a 5% stacking
gel (1.5 cm) and an
8-25 or 18% resolving
gel (18 cm), both containing
0.1% SDS.
Preparative
PAGE
was carried
out in the same system
except that
the SDS concentration
was lowered
to 0.033% and the length of the
stacking
and resolving
gel was 1 and 6 cm, respectively.
When stated,
the cysteine
residues
of the PS-I vesicle
and isolated
polypeptides
were labeled by S-carbamoylmethylation
using iodo[l-“Clacetamide
in the presence
of 2-mercaptoethanol
and 8 M urea (22) before
analysis
by SDS-PAGE
and autoradiography.
Apparent
molecular
masses were deduced
from the electrophoretic
mobility
of the following standards:
catalase,
aldolase,
bovine serum albumin,
ovalbumin,
and cytochrome
c. Determination
of the amount
of Coomassie
Bril-
(2
liant Blue R-250 bound to individual
natant was immediately frozen at -80 “C or diluted 5-fold with 20
mM Tricine
(pH 7.5), 0.2% (w/v) Triton X-100 and applied directly
x
2,000 ml) at a flow rate of 110 ml/h. The P700-containing
fractions
eluted
at an approximate
NaCl concentration
of 150 mM
and were combined
and concentrated
by ultrafiltration
to 30 ml in
an Amicon
cell (Amicon
Corp.)
fitted with a PM-30
membrane.
The
concentrated
PSI material
was applied
to a column
(5 X 95 cm) of
AcA 34 (LKB-Produkter
AB, Bromma,
Sweden)
equilibrated
in 25
mM Mes
(pH 6.5), 250 mM NaCl,
0.1% (w/v)
Triton
X-100
(AcA
buffer)
and eluted at a flow rate of 3 cm/h
(11). The vield of P700
was typically
300 nmol. When needed, the PSI vesicles were concentrated by ultrafiltration
using a PM-10
membrane.
Isolation
of Polypeptides
9-/Da
Polypeptide-Solid
NaSCN
(3.2 g) was added to a concentrated suspension
of PSI vesicles
(10 ml, 1.32 mg of chlorophyll/ml),
followed
by gentle shaking
for 30 min in the dark. The suspension
was diluted
to 15 ml with AcA buffer
and immediately
loaded onto a
column
(2.6 x 95 cm) of AcA 34 equilibrated
in AcA buffer and eluted
at a flow rate of 3 cm/h.
Fractions
(5.5 ml) were analyzed
for their
content
of acid-labile
sulfide and zero-valence
sulfur
(20), and their
polypeptide
content
was monitored
by SDS-PAGE,
followed
by Coomassie Brilliant
Blue R-250 staining.
Fractions
(70-85)
enriched
in
destaining
Ball (23).
of the
polyacrylamide
Amino
Acid
polypeptides
gels was
Analysis
carried
after staining and
out
according
to
and Sequencing
Samples
were hydrolyzed
with 6 N HCl at 110 “C in sealed, evacuated tubes for 24 h, and the hydrolysates
were analyzed
using a
Durrum
D500 amino acid analyzer.
Half-cystine
was determined
after
performic
acid oxidation
(24) or after S-carbamoylmethylation.
SCarbamoylmethylation
of isolated
polypeptides
was carried
out by
dialyzing
the purified
proteins
against
several hundredfold
excess of
20 mM ammonium
acetate
(pH 6.3), 0.05%
(w/v)
Triton
X-100,
followed
by lyophilization.
The lyophilized
protein
samnle was dissolved in 7 M guanidinium
chloride,
0.2 M Tris/HCl
(pH 8.0), 5 mM
EDTA,
and the solution
was thoroughly
flushed with nitrogen.
DTT
was added to give a final concentration
of 10 mM. After incubation
at 40 “C for 2 h, iodoacetamide
was added to the sample to give a
final concentration
of 200
mM.
After incubation
followed
by 24 h at 4 “C, the sample was extensively
20 mM ammonium
acetate
(pH 6.3), 0.05% (w/v)
lyophilized.
at 25 “C for 2 h,
dialyzed
against
Triton
X-100 and
12678
A 9-kDa
Protein
of Photosystem
Lyophilized polypeptide material was easily dissolved in 30% (v/
v) HOAc. Amino acid sequences were determined with both a Beckman 890C spinning cup sequenator (25) and an Applied Biosystems
gas-phase sequenator Model 479A usingthe program provided by the
company. Phenylthiohydantoins were identified by reverse-phase
high-pressure liquid chromatographyas described by Svendsen et al.
(26). C-terminal amino acid analysis (27) was carried out by carboxypeptidase Y (649 ng) digestion of the S-carbamoylmethylated 9-kDa
polypeptide (13 nmol) in a reaction mixture (150 pl) containing 40
mM Mes (pH6.5). 0.5% SDS, and 14 nmol of norleucine as an
internal standard. Aliquots withdrawn at different times were acidified (pH 2.2) to stop the enzymatic reaction and used directly for
amino acid analysis.
Acid-labile Sulfideand Zero-valence Sulfur
Acid-labilesulfide and zero-valencesulfurweredetermined
as
reported by H0j and Mdller (20) using the methylene blue procedure
with EtOAc extraction steps included to avoid interference from
of
chlorophyll. To secure accurate spectrophotometric determination
methylene blue, the absorption spectrum(500-750 nm) of each sample was recorded using an Aminco DW-2c spectrophotometer with a
typicalfull-scale setting of 0.05. The absorbance at 660 nmwas
determined from the spectra. Barley ferredoxin was isolated essentially as described (28) and used as a reference.
Additional Analytical Procedures
Thylakoids *S-labeledinvivowere
obtained as previously described (11). Chlorophyll was determined according to Arnon (29).
P700 wasquantitated from its ferricyanide-oxidized minus ascorbatereduced spectrum using an Aminco DW-2c spectrophotometer and
an extinction coefficientof 64 mM" cm" (30).
I Carries Centers A and B
rn
n
~
1
2
3
4
5
6
7
8
9
195-
105-
23-
9-
-
-
CBB-
Ag
CBB 35S
FIG. 1. Analysis of the polypeptide composition of purified
PS-Ivesicles and isolated polypeptides by SDS-PAGE. Electrophoresis was carried out overnight at 6 'C using an 8-25% high
RESULTS
Tris gradient gel. Unless otherwise indicated,the gel was stained with
Comparedtoour
previous study (ll),an ion-exchange Coomassie Brilliant Blue R-250( C B B ) .Lane 1 , purified PS-I vesicles;
chromatography step hasbeen introduced in the preparation lane 2, isolated 18-kDa polypeptide;lane 3, isolated 16-kDa polypepof the PS-I vesicles from barley. This allows processing of tide; lane 4, isolated 9-kDa polypeptideafter extensive handling; lane
5,isolated 9-kDa polypeptide;lane 6, thylakoids; lane 7,PS-I vesicles
large amounts of starting material and eliminates contamidevoid of the 18- and 16-kDa polypeptidesas obtained after NaSCN
nation with chloroplast coupling factor (Fig. 1, lane I ) . The treatment of PS-I vesicles originally treated with urea (the bands
isolation procedure here reported has routinely been used to werevisualizedby
alkalinesilver staining); lane 8, purified PS-I
isolatePS-I
vesicles fromthylakoidscontaining
several vesicles obtained from plants "S-labeled in uivo; lane 9, autoradioghundred milligrams of chlorophyll. The yield of P700 is ap- raphy of sample in lane 8.
proximately 1 nmol/mg of chlorophyll of the thylakoids. This
corresponds to a yield of 30%. The PS-I preparation used in
The Coomassie Brilliant Blue R-250bound to theindividual
Ref. 11 had a chlorophyll to P700 ratio of 110, whereas the polypeptide bands of the PS-I preparation was eluted from
ratiois 60in thepresentpreparation.This
difference is the gels and quantified spectrophotometrically(23). Normalexplained by the prolonged exposure time to Triton X-100 ization based on their apparent molecular masses and a uniform binding of Coomassie Brilliant Blue R-250 gave the
caused by inclusion of the ion-exchange step.
five main components:
InthePAGEsystems
previouslyused, the major PS-I following stoichiometryforthe
PSpolypeptides were assigned molecular masses of 70 (doublet), 2.01.2:1.2:0.5:1.1, indicating a stoichiometry in the native
polypeptide I complex of 42:2:1:2 for the apoproteinsof P700-chlorophyll
18, 15, 10, and 8 kDa (11). In this study, the
composition of the isolated PS-I vesicles was analyzed in the a-protein 1 and the la-, 16-, 14-, and 9-kDa polypeptides,
high Tris gel system of Fling and Gregerson (21). Although respectively.
devoid of urea, this system proved superior in focusing the
The relative distribution of sulfur amino acids among the
low molecular mass polypeptides of PS-I. Based on the elec- PS-I polypeptides was assessed by electrophoresis and autotrophoretic mobility of known standards in the high Tris radiography of a n 3sS-labeled PS-I preparation obtainedfrom
system, thecalculated apparent molecular masses of the major barley seedlings grown in the presenceof ["S]sulfate (Fig. 1,
PS-I polypeptides were 82 (doublet), 18, 16, 14, and 9 kDa,
lanes 8 and 9). In agreement with earlier observations (11,17,
respectively. Apparent molecular masses of 82 kDa for the 33), the 18- and 14-kDa polypeptides were found to incorpotwo apoproteins of P700-chlorophyll a-protein 1 are in close rate small amountsof "S label, whereas the 16-kDapolypepagreement with the molecular massespredicted from the tide was not labeled at all. P700-chlorophyll a-protein 1 and
nucleotide sequence of their genes (31). The band at105 kDa the 9-kDapolypeptide were both strongly labeled. The superrepresents P700-chlorophyll a-protein 1 which has not been imposition of the labeled band at 9 kDa with that obtained
converted into the apoprotein. The band 195
at kDa is prob- by Coomassie Brilliant Blue R-250 staining (Fig. 1, lanes 8
(17, 34, 35) inability to
ably a dimer of P700-chlorophyll a-protein 1. A minor com- and 9) shows thatthereported
ponent migrating just above the 9-kDa polypeptide was ob- visualize the 9-kDa polypeptide by Coomassie Brilliant Blue
served. In some preparations, an additional minor componentR-250 staining was not due to an intrinsic property of the
was observed inthe 23-kDaregion. This component is thoughtprotein, but merely reflected the poor characteristics of the
to representresidual amounts of light-harvesting chlorophyll- SDS-PAGE systems earlierused. Specific assessment of the
of the PS-I
protein I (32) and was not detectable in most
content of cysteine residues in the individual PS-I polypeppreparations.
tides was achieved by14C-S-carbamoylmethylationof the PS-
A 9-kDa Protein of Photosystem I Carries Centers A and B
0"
12679
-
I preparation from spinach(17), but is in agreementwith the
published nucleotide sequences of the genes encoding thetwo
apoproteins (36).
195An isolated intact iron-sulfur protein can
be detected by its
property to release acid-labile sulfide, whereas a denatured
105iron-sulfur protein may retain acid-labile sulfide in the form
02-0
of zero-valence sulfur (37). Incubation of the PS-I preparation
with 3.4 M NaSCN and subsequent gel filtration on AcA 34
allowed collection of fractions containingvarying amounts of
Y
the 18-, 16-, and 9-kDa polypeptides as monitored by SDS9 "
PAGE and Coomassie Brilliant Blue R-250 staining of each
;gz
individual fraction. The fractions also contained
zero-valence
14-=
sulfur, whereas no acid-labile sulfur was detectable. The elu9tion of zero-valence sulfur correlatedwith that of the cysteinerich 9-kDa polypeptide, but not with that of the 18- and 16kDa polypeptides. Fractions rich in zero-valence sulfur were
B: 1 4 pattern
~
A: CBB stain
combined and dialyzed. Subsequent chromatography on CMFIG.2. Polypeptide composition of purified PS-I vesicles Sepharose CL-GB followed by a final gel filtration step on
as monitored by Coomassie Brilliant
and isolated polypeptides
Blue R-250 staining ( A ) and autoradiography after I4C-S- AcA 34 resulted in a homogeneous preparation of the 9-kDa
carbamoylmethylation ( B ) .Electrophoresis was carried out over- polypeptide. From both columns, the elution of the 9-kDa
night at 6 "C using an 8-25% highTris gradient gel. Lane I , thylakoids polypeptide and zero-valence sulfur coincided. The homoge(not S-carbamoylmethylated) + I4C-labeled molecular mass stan- neity of the isolated 9-kDa polypeptide was assessed by SDSdards; lane 2, "C-labeled molecular mass standards; lane 3, purified PAGE, followed by Coomassie Brilliant Blue R-250 staining
PS-I vesicles; lane 4, isolated 18-kDa polypeptide; lane 5,isolated 16kDa polypeptide; lane 6, isolated 8-kDa polypeptide; lane 7, purified (Fig. 1, lane 5 ) and autoradiography after I4C-S-carbamoylmethylation (Fig. 2, lane 6).The zero-valence sulfur contained
PS-I vesicles.
in the isolated 9-kDa polypeptide was stable todialysis at pH
5 and tolyophilization and was therefore bound covalently to
the polypeptide backbone, most likely as a trisulfide (37). The
isolated polypeptide lacked chromophores absorbing around
420 nm. Such chromophores are presentin native ferredoxins
(37). Extensive handling of the isolated polypeptide, e.g. by
repeated lyophilizations, tended to generate tracesof a component of slightly faster electrophoreticmobility (Fig. 1, lane
1 2 3 4 5 6 7
x
-"
"
-
-
1 2 3 4 5 6 7
/I
4).
The procedure developed to purify the 9-kDa polypeptide
also provided homogeneous preparations of the 18- and 16kDa polypeptides. However, large amounts of these two poly0
5
10
1
20
0
5
10
15
20
pl P S I VESCLES
pl PS I VESICLES
peptides were more easilyobtained after treatmentof the PSFIG.3. Quantitation of the amount of zero-valence sulfur I preparation with6 M urea. When such an extract
was applied
associated with the homogeneous preparation of the 9-kDa to a column of Polybuffer Exchanger 94 equilibrated in 20
polypeptide as compared to the content of acid-labile sulfidemM imidazole HCl (pH7.6), 5 M urea, the 18-kDa
polypeptide
of t h e
in nativePS-I particles containing an identical amount
a
did
not
bind
to
the
column
and
could
be
collected
in
9-kDa polypeptide. A, standard curve indicating a linear relationship between the amount of PS-I vesicles (0.27 mgof chlorophyll/ homogeneous form (Fig. 1, lane 2). Upon subsequent washing
ml) assayed and the total amount of acid-labile sulfide and zero- of the column with 25 mM imidazole HCl (pH 7.6), 0.1%
valence sulfur detected; B, standard curve indicating the linear rela- Triton X-100, 0.05% Empigen BB, a proportion of the 16tionship between the amount of PS-I vesicles assayed and theamount kDa polypeptide was released in a homogeneous form (Fig. 1,
of Coomassie Brilliant Blue R-250 bound to the 9-kDa polypeptide
of the PS-I vesicle after SDS-PAGE. A and B, aliquots of a homoge- lane 3 ) . A preparation containing P700-chlorophyll a-protein
neous preparation of the 9-kDa polypeptide were subjected to analyses 1 and low molecular mass polypeptides was obtained by eluidentical to those described above. In one such experiment, the 9- tion with Polybuffer 74 (pH 4.5). Subsequent treatment of
kDa polypeptide was found to bind Coomassie Brilliant Blue R-250 this preparation with NaSCN followed by gel filtration recorresponding to an absorption of 0.033 a t 595 nm and to contain sulted in apreparation containingP700-chlorophyll a-protein
zero-valence sulfur producing an absorption of 0.0022 at 660 nm. As 1 and the 14- and 9-kDa polypeptides (Fig. 1, lane 7). A
illustrated on A and B, this particular experiment demonstrated that
the isolated 9-kDa polypeptide binds zero-valence sulfur correspond- specific association of acid-labile sulfideor zero-valence sulfur
ing to 32% of the amount of acid-labile sulfide present in PS-I vesicles with the 18- or 16-kDa polypeptides was not observed under
containing an identical amount of the 9-kDa polypeptide.
any of the isolation procedures tested. Even when purified,
the 16-kDa polypeptide was not reactive toward i ~ d o - l - [ ' ~ C ]
acetamide (Fig. 2, lune 5), whereasa weak labeling was
I preparationprior to electrophoresis (Fig. 2). The 9-kDa
18-kDapolypeptide (Fig. 2, lane 4 ) . A
polypeptide was by far the mostlabeled component, with less obtainedwiththe
condition for the successful application of the purification
label inP700-chlorophyll a-protein1,littleinthe18-kDa
homogeneous preparations
polypeptide, and none in the16- and 14-kDa polypeptides. It procedures reported here to obtain
should be noted that the denaturing conditions used for the of the 18-, 16-, and 9-kDapolypeptides of PS-I is theuse of a
reductive alkylation of PS-I resulted in a partial loss of the highly purified preparation of PS-I vesicles as the starting
material.
14-kDa polypeptidewhich seems to aggregatemoreeasily
The amountof zero-valence sulfur bound to theisolated 9than the rest of the PS-I components (Fig. 2, lane 3 ) . The
presence of I4C label inthe P700-chlorophyll a-protein 1 kDa polypeptide was quantitated with native PS-I as a stanpolypeptides is in contrast to the results obtained with
a PS- dard (Fig. 3). Increasing amounts of PS-I were subjected to
32x
,t
I
I
A 9-kDa Protein of Photosystem I Carries Centers A and B
12680
SDS-PAGE. After electrophoresis, the Coomassie Brilliant
*m
In
Blue R-250 specifically bound to the 9-kDa polypeptide band
of each gel lane was quantitated spectrophotometrically (Fig.
3B) (23). Identical amounts
of the PS-I preparation were
m
assayed for acid-labile sulfide and zero-valence sulfur (Fig.
3A), and two standard curves were constructed. In an analogous manner, the Coomassie Brilliant Blue R-250 bound to
3
the purified 9-kDa polypeptide afterSDS-PAGEandthe
correspondingcontent of zero-valance sulfur were deter-401s
mined. The yield of zero-valence sulfur obtained by assaying
the 9-kDa polypeptide from two different preparations was
32 and 37%of that obtained when assaying an amount of
-0,010
native PS-I containing exactly the same amount
of 9-kDa PSI. However, when the native PS-I vesicle was subjected to a
-0,005
NaSCN treatmentanalogous to that used to isolate the 9-kDa
PS-I, the recovery of acid-labile sulfide was less than 60%
-0
even after incubation with DTT. It is therefore evident that
the native 9-kDa polypeptide must bind a major part of the
GEL SLCE NUMBER
acid-labile sulfide of the native PS-I particle.
FIG.5. Relative distribution of zero-valence sulfur between
To determine therelative distribution of zero-valence sulfur the lowmolecular mass polypeptides. Electrophoresis was carried
between P700-chlorophyll a-protein 1and the 9-kDa
polypep- out for 5.5 h at 6 "C using an 18%high Tris gradient gel. After SDStide under identical experimental conditions, the
purified PS- PAGE, the gel was cut into 5-mm segments which were analyzed for
I vesicles were subjected to preparative SDS-PAGE. After their content of zero-valance sulfur (So,0).The polypeptide content
of each segment was analyzed by re-electrophoresis, and the amount
electrophoresis, the gel was cut into narrow horizontal seg- of
Coomassie Brilliant Blue R-250 bound to each polypeptide was
ments. The contentof acid-labile sulfide and of zero-valence quantitated. In this particular experiment, the recovery of the 9-kDa
sulfur was determined after D T T incubation as described polypeptide is low.
(20). Small,equally sized parts of each segment were used for
re-electrophoresis to establish the
polypeptide composition in peptides of P700-chlorophyll a-protein 1. The apoproteins of
the segments and to spectrophotometrically quantify
polythe P700-chlorophyll a-protein 1 barely enter the 18%resolving
peptides present (23).When an 8-25% gradient gel was used, gel, which is therefore not suitable for determination of the
between 30 and 35% of the recovered acid-labile sulfide was
relative distribution of zero-valence sulfur betweenP700associatedwith P700-chlorophyll a-protein 1, whereas 65- chlorophyll a-protein 1 and the low molecular weight poly70% was associated with polypeptides in the low molecular peptides.
mass region (Fig. 4). To separate the9-kDa polypeptide more
It was of interest to obtain sequence information on the
efficiently from those at 18, 16, and 14 kDa, a similar exper- isolated 9-kDa polypeptide. The S-carbamoylmethylated 9iment using an 18%resolving gel was performed (Fig. 5). This kDa polypeptide was therefore subjected to Edman degradaexperimentdemonstratedthatthe
acid-labilesulfide
re- tion in a liquid-phase spinning cup sequenator (-10 nmol, 29
covered after incubation with D T T was derived from the 9- cycles, repeated three times, 94% repetitive yield) as well as
kDa polypeptide. Thus, after separation of the polypeptides in a gas-phase sequenator (-1 nmol, 92% repetitive yield).
of the PS-Ivesicle by SDS-PAGE, 65-70% of the acid-labile The sequence for the 29 N-terminal residues of the isolated
sulfide recovered after DTT treatment is associated with the 9-kDa polypeptide is shown on Fig. 6. The spacing of the 4
9-kDa polypeptide, whereas 30-35% resides in thelarge poly- identified cysteine residues is strongly indicative of a [4Fe4S] protein(38).Amino acidanalysis of the 9-kDapolypeptide
I
I
after treatment with
performic acid (24) revealed the presence
0,008
'BAD"
of 8 cysteine residues/78 amino acids (Table I), indicating
8 2 a
16ADa
that the protein mightbe a 2[4Fe-4S] protein. Very recently,
142Da
the complete nucleotide sequences of chloroplast DNA from
the liverwort Marchantia polymorpha (39) and from tobacco
(Nicotiana tabacum) (40) have been determined. Ohyama e t
al. (39) pointed out that the M. polymorpha sequence contained two ORFs denotedfrxA and frxB inwhich the periodic
appearance of cysteine residues resembled that of [4Fe-4S]
ferredoxins. When the N-terminal sequences predicted from
these two ORFs were compared with the N-terminal amino
acid sequence of the 9-kDa polypeptide isolated from PS-I
vesicles of barley, the homology with frxA, but not with frxB,
was strikingand leaves nodoubtthattheisolatedPS-I
polypeptide is coded for by the corresponding ORF on the
barley chloroplast genome (Fig. 6). To accord with the no0
menclature used by Ohyama et al. (39) and Gray e t al. (41)
1
3
5
7
0
1
1
1
3
GEL SLICE NUMBER
for the previously identified chloroplast genes encoding memFIG. 4. Relative distribution of zero-valence sulfur between brane proteins catalyzing light reactions of photosynthesis,
P700-chlorophylla-protein 1 and the low molecular mass we designate this gene as psaC. The psaCof the M. polymorpolypeptides. Electrophoresis was carried out for 5.5 h at 6 "C using
an 8-25% high Tris gradient gel. The bars indicate the polypeptide pha chloroplast genome is locatedbetween the ndh41 and
distribution between the different gel slices (5 mm) as monitored by ndh4 genes (39). An initial search on the chloroplastgenome
of tobacco (40) for a similar ORF coding for an N-terminal
re-electrophoresis.
-i
;
-
l l i
I
i
A 9-kDa Protein of Photosystgm I Carries Centers A and B
FIG. 6. Partial amino acid sequence for the isolated 9-kDa
2[4Fe-4S] polypeptide of PS-I from barley compared with the
amino acid sequences deduced from the corresponding gene
psaC on the chloroplast genomes of tobacco (40) and M. polyrnorpha (39).Identity between codons and amino acid residues of
the 9-kDa polypeptide from different species is indicated with a plus.
Differences between the two nucleotide sequences causing amino acid
substitutions are indicated by dashed bores. The cysteine residues
chelating the [4Fe-4S] clusters are bored. The nucleotide inserted
into the tobacco sequence (40) to restore identity with the N-terminal
amino acid sequence of the 9-kDa polypeptide from barley is indicated
with a question mark.
TABLE
I
Amino acid composition of the 9-kDa photosystem Ipolypeptide
carrying the two I4Fe-4SI centers A and B
Amino acid
H. vulgare
N . tabacurn"
M . polymorphab
9
8
9
Cys'
8.1
7
6
5
Asx
5.6
7
6
7
Thr
6.1
5
7
6
Ser
7.0
6
6
7
Glx
7.2
4
4
4
Pro
3.6
5
4
6
GlY
6.2
6
6
6
Ala
6.3
6
6
5
Val
4.8
3
3
1
Met
1.2
4
4
4
Ile
3.5
4
4
4
Leu
4.1
3
3
3
TYr
2.5
1
1
P he
1.8
2
1
His
1.3
1
2
4
3
4
LYS
4.0
6
5
5
Arg
4.7
2
1
ND
Trp
NDd
12681
of codon 20 in the psaC ORF of M. polymorpha restored a
nucleotide sequence whichcoded for a protein identical to the
N-terminal part of the barley 9-kDa protein (Fig. 6). The
amino acid composition of the isolated 9-kDa PS-I polypeptide of barley resembles that predicted from the psaC genes
between
of M . polymorpha and tobacco (Table I). The identity
the partial sequence of the 9-kDa PS-I of barley and the
sequencededuced from the psaCgene of tobacco strongly
indicates that the
sequence of the remaining part
of the barley
protein will be very homologous to the correspondingsequences in tobacco and M . polymorpha.
The C-terminal amino acid of the 9-kDa polypeptide was
determined by carboxypeptidase Y digestion (27) and was
found tobe tyrosine. Thus, apartfrom the removal of the Nterminal methionine residue, the primary translation product
of the 9-kDa polypeptide is not proteolytically processed. Xray crystallographic analyses of the soluble 2[4Fe-4S] ferredoxin fromPeptococcus aerogenes had established the identity
of the cysteines bound to the
two [4Fe-4S] clusters (42). One
[4Fe-4S] cluster is bound to Cys-X-X-Cys-X-X-Cys (where
X represents amino acid)in the N-terminalhalf of the protein
and to Cys-Pro in the C-terminalhalf. The second cluster is
bound to Cys-X-X-Cys-X-X-Cys in the C-terminalhalf and
to Cys-Pro in the N-terminal part
of the protein. The partial
amino acid sequence of the isolated 9-kDa polypeptide and
the deduced amino acid sequence for the corresponding proteins in tobacco and M . polymorpha reveal identical cysteinecontaining segments, thereby identifying these proteins as
2[4Fe-4S] proteins.
A hydropathy plot according to Hopp and Woods (43) is
shown in Fig. 7. It is interesting to note that the 4 cysteine
residues at positions 10, 13, 16, and 57 anchoring one of the
[4Fe-4S] centers(42) are positioned in relatively hydrophobic
regions, whereas the cysteineresidues at positions 20, 47, 50,
and 53 coordinating the second center are located in more
hydrophilic stretches of the molecule. The positioning of the
cysteine residues in the bacterial 2[4Fe-4S] ferredoxins (38)
suggests that the cysteine
residue at position 33 is not involved
in anchoring an iron-sulfur center.
The aminoacid composition of the isolated 16- and 18-kDa
r
I l l
I
I
I l l
1
Total
78.0 - 79
81
81
Deduced from the nucleotide sequence identified on the genome
of N . tabacurn (40) after insertion of a missing nucleotide (see "Results").
* Deduced from the nucleotide sequence identified on the genome
of M.p o l y m o r p h (39).
Determined as cysteic acid after performic acid oxidation (24).
ND, not determined.
a
part of the polypeptide homologous to thebarley 9-kDa polypeptide was unsuccessful. Search at the nucleotide level, however, revealed extensive homology between the psaCregion of
M . polymorpha and the corresponding region in tobacco. It
turned out that the ORF found
in M . polymorpha was not
listed in the analysis
of the tobacco chloroplast genome, most
probably because a single base pair had been missed in the
sequencing of the tobacco chloroplast genome. Thus, insertion
of a thymine residue in the noncoding strand of the tobacco
sequence at a position corresponding to the wobble position
1
i0
20 40 30
50
60
AMINOACIDNUMBER
70
&J
FIG. 7. Hydropathy plots of the 9-kDa 2[4Fe-4S] polypeptides of tobacco and M. polyrnorpha carrying centers A and
B of photosystem I. The amino acid sequences werederived by
identification of the gene psaC for the two polypeptides on the
completely sequenced chloroplast DNA of tobacco (40) and M.polymorphs (39). The hydropathy plots were calculated according to Hopp
and Woods (43).
A 9-kDa Proteinof Photosystem I Carries CentersA and B
12682
SIWC.
P.
elrdenli
P . elldenti
Reriduer
11-281
iz9-541
FIG. 8. Comparison of the amino acid sequences of the 9kDa 2[4Fe-4S] PS-I polypeptides of barley and tobaccowith
that of a 2[4Fe-4S] ferredoxin of P. ehdenii (38).To indicate
the presumed gene duplication (47), the N-and C-terminal halves of
the proteins have been aligned with respect to the positioning of the
cysteine residues chelating the two [4Fe-4S] clusters. Identicalresidues are indicated with bores. Each amino acid is indicated by the
standard single-letter code.
polypeptides was also determined.' The 16-kDa polypeptide
contained neither methionine nor cysteine, in accordance
the earlier reported absenceof label in this protein afterI4Ccarbamoylmethylation and after
labeling with 35Sin vivo (11).
This polypeptide also lacks histidine. The 18-kDa
polypeptide
contained small but significant amountsof cysteine (-1 residue/molecule).2 Thus, the 16-and 18-kDa polypeptides are
not iron-sulfur apoproteins. Both the 18- and 16-kDa
polypeptides contained very high levels of alanine and proline.
This was reflected in the partial aminoacid sequences which
we have obtained for these two proteins.*
DISCUSSION
Partial amino acid sequencing of the 9-kDa polypeptide
isolated from PS-I vesicles of barley permitted identification
of its correspondinggene psaC on the chloroplast
genomes of
of the
tobacco (40) andM . polymorph (39) and identification
protein as a carrier of two [4Fe-4S] clusters (Fig. 6). The
spacing of the cysteine residues does not fit that of soluble
[ZFe-ZS] proteins which also lack the Cys-Pro segment (38,
44-46). The soluble2[4Fe-4S] ferredoxins reveala strong
internal homology presumably due toa gene duplication (47).
The deduced sequence for the 9-kDa polypeptide of tobacco
displays a similar internal homology between residues 3-25
and 40-62 (Fig. 8).
The availability of analytical procedures(20) permitting
fast andreliable determination of acid-labile sulfide and zerovalence sulfur was essential in the development
of the procedure which resulted in isolation
of the 9-kDapolypeptide from
barleyas a partiallydenatured 2[4Fe-4S] protein.Partial
denaturation was evidenced by the lack of absorption around
420 nm andby the inabilityof the polypeptide to release acidlabile sulfide without prior reduction. The amount of zerovalence sulfur bound to the isolated 9-kDa polypeptide was
quantified by two different procedures. After SDS-PAGE of
the purified PS-I vesicle, the amount of zero-valence sulfur
associated with the 9-kDapolypeptide was twice the amount
found tobe associated withP700-chlorophyll a-protein 1 (Fig.
4). In its purified state, the amount of zero-valence sulfur
associated with the 9-kDa polypeptide was one-third of the
amount of acid-labile sulfide present in a native PS-I preparation containing an identical amount of 9-kDa polypeptide
(Fig. 3). This is explainedby the less than 60% yield of acidlabilesulfide obtained from denaturedPS-I vesicles after
D T T reduction. Incomplete conversion of zero-valence sulfur
into acid-labile sulfide upon reduction may be one reason for
the lower recovery of acid-labile sulfide from the isolated 9kDa polypeptide. Thus, the yield obtained with cysteine trisulfide as a standard was 77% (37). In addition, Petering et
* H. V. Scheller, P. B. Hej, I. Svendsen, and B. L. Mller, manuscript in preparation.
al. (37) observed that the zero-valence sulfurs bound in the
oxidatively denatured bacterial 2[4Fe-4S] ferredoxins of Micrococcus lactylyticus, Clostridium pasteurianum, and Peptostreptococcus elsdenii were only 42, 48, and 63% of the acidlabile sulfide found in the native proteins, respectively. Of
these soluble ferredoxins, that of P. elsdenii showsthe highest
degree of structural homology with the psaC gene product
(Fig. 8) (38). The low recoveries led Petering etal. to conclude
that thezero-valence sulfur of these proteins is bound mainly
in a cysteine trisulfide structure
(37). Quantitative retainment
of the acid-labile sulfide originally present in native [4Fe-4S]
proteins would require the formationof a cysteine tetrasulfide
structure (37). However, the recovery of zero-valence sulfur
from oxidatively denatured [2Fe-2S] proteins wasalso low
(37). In contrast to the results of Petering et al. (37) and to
the
Golbeck and Kok (48) have
with results obtained in this study,
reported a 100% recovery of acid-labile sulfide following denaturation of PS-I particles and regeneration of acid-labile
sulfidefromzero-valence sulfur by D T T treatment. Using
recoveries of 63 and 77% for the formation of zero-valence
sulfur and the regeneration
of acid-labile sulfide, respectively,
the amount of acid-labile sulfur calculated to correspond to
the amount of zero-valence sulfur detected on the isolated
9kDa PS-I polypeptide corresponds t o 70% of the acid-labile
sulfide inthenativePS-Iparticle.This
value is inclose
agreement with the
relative distribution of zero-valence sulfur
polypepbetween P700-chlorophyll a-protein 1and the 9-kDa
tide as determined after the denaturing conditions of SDSPAGE (Fig. 4) by which acid-labile sulfide is converted into
zero-valence sulfur (11, 20, 34, 35).
Native PS-I vesicles are generally foundtocontain
12
molecules of acid-labile sulfide for each molecule of P700 (1,
2, 3, 11).From the results presentedhere, we can assign 8 of
the molecules of acid-labile sulfide to the 9-kDa polypeptide
and the remaining 4 to the apoproteins of P700-chlorophyll
a-protein 1. H0j and Maller (11)and Golbeck and Cornelius
(49) have recentlydemonstratedthat P700-chlorophyll aprotein 1 carries center X. Centers A and B are therefore
identified as the two [4Fe-4S]clusters of the chloroplastencoded 9-kDa 2[4Fe-4S] polypeptide. Experiments based on
ESR spectrometry have revealed a differential sensitivity of
centers A and B toward oxidative denaturation(48)and
toward reactivity with mercurials (50), with center B as the
most sensitive cluster. Similarly, center B has been demonpstrated tobe sensitive to the membrane-impermeant probe
diazonium benzene sulfonate (51). A hydropathy plot of the
9-kDa polypeptide reveals that the [4Fe-4S] cluster coordinated by cysteine residues 10, 13, 16, and 57 is buried in the
membrane, whereas the cluster coordinatedby cysteine residues 20,47,50, and53 is more external (Fig. 7). We therefore
conclude that these two [4Fe-4S] clusters represent centersA
and B, respectively.Selective destruction of one[4Fe-4S]
center had also been reported in the soluble 2[4Fe-4S] ferredoxinI of Azotobacter uinelandii(52). The localization of
centers A and B on the same
polypeptide chain is in agreement
with the strong interaction observed between these two centers by ESR spectroscopy ( 5 , 7, 54).
One important aspect regarding the biosynthesis of the
2[4Fe-4S]holoprotein is theformation of theiron-sulfur
cluster.Denatured soluble[2Fe-2S] ferredoxinsare easily
reconstituted (55). Although N- and C-terminal analyses of
the isolated 9-kDa polypeptide established that the primary
gene product of psaC is not post-translationally
cleaved except
for the loss of the N-terminal methionine residue, it has not
yet been possible by reconstitution experiments to regenerate
the ESR signals of the iron-sulfur centers
from denatured PS-
A 9-kDa
Protein
Photosystem
of
I Carries Centers A and B
12683
peptides were also purified to homogeneity. The function of
I vesicles (56). Takahashi et al. (57) have recently demonstrated that sulfur atomsof the iron-sulfur clusterof chloro- these two polypeptides remains unknown. However, partial
plast ferredoxin are derived from cysteine and thata soluble amino acid sequencing indicated that the two polypeptides
stroma enzyme isinvolved in the cluster formation. It
will be are related.*
interesting to test the activity
of this enzyme toward the
isolated 9-kDa apoprotein.
Acknowledgments-Hanne Linde Nielsen, Inga Olsen, Bodil Corneliussen, Lone Sbrensen, and Pia Breddam are thanked for skillful
P700-chlorophyll a-protein 1 was showninthepresent
technical assistance. Drs. Birte Svensson and David Simpson are
study to carry 4 of the 12 molecules of acid-labilesulfide
for helpful discussions. Professor Knud W. Henningsen and
associated with the native PS-Ivesicle per molecule of P700. thanked
Drs. Jan Lembeck, T. G . Petersen, and C.-E. Olsen are thanked for
We have previously reported that P700-chlorophyll a-protein performing the computer analyses.
1binds 4.3 iron atoms/molecule of P700 (11). Using experimentalconditions whereP700-chlorophyll a-protein 1 had
REFERENCES
been functionally detached from the lower molecular mass
1. Rutherford, A. W., and Heathcote, P. (1985) Photosynth. Res. 6 ,
polypeptides, Golbeck and Cornelius (49) obtained absorbance
295-316
transients at 698 nm, indicating that the iron-sulfur center
2. Golbeck, J. H. (1987) J. Membr. Sci., in press
associated with P700-chlorophyll a-protein 1 was center X.
3. Meller, B. L., H@j,P. B., Halkier, B. A., Olsen, I., Nielsen, H. L.,
and Madsen, A. (1987) Dan. J. Agron. 132,5-21
Mossbauer spectroscopyidentified center X as a[4Fe-4S]
4. Evans, E. H., Rush, J. D., Johnson, C. E., and Evans, M. C. W.
center (10). However, extended x-ray absorption fine-struc(1979) Biochem. J. 182,861-865
ture measurements (58) and core extrusion studies (59) indi5. Malkin, R., and Bearden, A. J. (1971) Proc. Natl. Acad. Sci.
cated that PS-I contains [2Fe-2S] clusterswell
as as [4Fe-4S]
U. S. A. 68, 16-19
clusters. In this study, centers A and B have been identified
6. Evans, E. H., Cammack, R., and Evans, M. C.W. (1976) Biochem.
as [4Fe-4S] centers. This points toward X
center
as a [2Fe-2S]
Biophys. Res. Commun. 6 8 , 1212-1218
7. Evans, M. C. W., Telfer, A., and Lord, A.V. (1972) Biochim.
center. With 4 iron and 4 sulfuratomspresentonP700Biophys. Acta 2 6 7 , 530-537
chlorophyll a-protein 1 per molecule P700, this would permit
8. Rupp, H., Rao, K. K., Hall, D. O., and Cammack, R. (1978)
the location of two [2Fe-2S] centers on this protein. Recently,
Biochim. Biophys. Acta 5 3 7 , 255-269
Bonnerjea and Evans (60) have provided evidence for a cor9. Evans, M. C. W., Sihra, C. K., and Cammack, R. (1976) Biochem.
responding heterogeneityof the signal associated with center
J. 1 5 8 , 71-77
10. Evans, E. H. Dickson, D. P. E., Johnson, C. E., Rush, J. D., and
X.
Evans, M. C. W. (1981) Eur. J . Biochem. l l 8 , 8 1 - 8 4
If center X is composed of two traditional [2Fe-2S] centers,
this would require the availability
of 8 cysteine residues. P700- 11. H0j, P. B., and Mbller, B. L. (1986) J. Biol. Chem. 2 6 1 , 1429214300
chlorophyll a-protein 1 is composed of two apoproteins with 12. Bonnerjea, J., Ortiz, W., and Malkin, R. (1985) Arch. Biochem.
approximate molecular masses of 83 kDaand which are
Biophy~.2 4 0 , 15-20
(31). Both apoproteins
present in near equimolar amounts
13. Golbeck, J. H., Parrett, K. G., and Root, L. L. (1987) in Progress
in Photosynthesis Research (Biggins, J., ed) Vol. I, pp. 253-256,
arechloroplast-encoded,andtheir
genes (psaA and psaB)
Martinus Nijhoff, Dordrecht, The Netherlands
have been sequenced inmaize (36), tobacco (40), spinach (61),
14.
Malkin,
R., Aparicio, P. J., and Arnon, D. I. (1974) Proc. Natl.
pea (62), M . polymorpha (39), Euglena (63), and SynechococAcad. Sci. U. S. A. 7 1 , 2362-2366
cus (64). In all these
species, both genes specifiedthe following 15. Hiller, R. G., Mdler, B. L., and H@yer-Hansen,G. (1980) Curlscompletely conserved stretch of 12 amino acids: Phe-Pro-Cysberg Res. Commun. 4 5 , 315-328
Asp-Gly-Pro-Gly-Arg-Gly-Gly-Thr-Cys.
Besides the 2cys16. Mdler, B. L., H@yer-Hansen,G., and Hiller, R. G. (1981) Proc.
Int. Congr. Photosynth. Res. 3,245-256
teines of this segment, no other cysteines were found in the
psaB gene product. ThepsaA gene of tobacco, maize, and pea 17. Lagoutte, B., Setif, P., and Duranton, J. (1984) FEBS Lett. 1 7 4 ,
24-29
specified 2 additional cysteines, whereas thatof Euglena and
18. Guikema, J., and Sherman, L. (1982) Biochim. Biophys. Acta
Synechococcus specified only 1 of these residues. Fish et al.
681,440-450
(53) speculated that the 2 cysteine residues of the conserved 19. Mdler, B. L., and Hplj, P. B. (1983) Carkberg Res. Commun. 4 8 ,
161-185
segment indicated above were connected by a disulfide bond
in the native protein. We propose, in agreement with specu- 20. H0j, P. B., and Mdler,B. L. (1987) Anal. Biochem. 2 4 0 , in press
lations of Golbeck et al. (59), that these conserved cysteine 21. Fling, S. P., and Gregerson, D. S. (1986) Anal. Biochem. 1 5 5 ,
83-88
residuesfromtwopsaApolypeptides
and fromtwo psaB 22. Crestfield, A. M., Moore, S., and Stein, W. H. (1963) J. Biol.
polypeptides constitute the 8 cysteine residues necessary to
Chem. 238,622-627
coordinate the two [2Fe-2S] clusters of P700-chlorophyll a- 23. Ball, E. H. (1986) Anal. Biochem. 155,23-27
protein 1. Thus, four large polypeptides would be required to 24. Hirs, C . H. W. (1967) Methods Enzymol. 1 1 , 197-199
bind 1 molecule of P700 and thetwo [2Fe-2S] clusters repre- 25. Johansen, J. T., Overballe-Petersen, C., Martin, B., Hasemann,
V., and Svendsen, I. (1979) Carlsberg Res. Commun. 4 4 , 201senting center X. Assuming a uniform binding of Coomassie
217
Brilliant Blue R-250, the stoichiometryof the polypeptides of 26. Svendsen, I., Martin, B., and Jonassen, I. (1980) Carlsberg Res.
the native PS-I vesicle was calculated t o be 42:2:1:2 for the
Commun. 45,79-85
apoproteins of P700-chlorophyll a-protein 1 and the 18-, 16-, 27. Martin, B., Svendsen, I., and Ottesen, M. (1977) Carlsberg Res.
Commun. 42,99-102
14-, and 9-kDapolypeptides, respectively. These ratios arein
close agreement with those obtained with I4C-labeled prepa- 28. Buchanan, B. B., and Arnon, D. I. (1971) Methods Enzymol. 2 3 ,
413-440
escaped our atten- 29. Arnon, D. I. (1949) Plant Physiol. 2 4 , 1-14
rations from Synechococcus (56). It has not
tionthattheseratiospredictthe
presence of two 9-kDa 30. Hiyama, T., and Ke, B. (1972) Biochim. Biophys. Acta 2 6 7 , 160171
polypeptides/molecule of P700, which would give a calculated
content of20 eq of iron and acid-labile sulfide/molecule of 31. Fish, L. E., and Bogorad, L. (1986) J. Biol. Chem. 2 6 1 , 81348139
P700. The experimentally found ratio is 12 (1, 2, 3, 11). At
32. Lam, E., Ortiz, W., and Malkin, R. (1984) FEBS Lett. 1 6 8 , 10present, we are not able to discriminatebetween the possible
14
explanations for this discrepancy.
33. Nechushtai, R., andNelson, N. (1981)J.Bid. Chem. 256,11624During the course of this work, the 18- and 16-kDa poly11628
12684
A 9-kDa
Protein
of Photosystem I Carries
Centers
34. Mdler, B. L., Halkier, B. A., and H#j, P. B. (1987) in Progress in
PhotosynthesisResearch (Biggins, J., ed) Vol. 11, pp. 49-52,
Martinus Nijhoff, Dordrecht, The Netherlands
35. H0j, P. B., Halkier, B. A., andMdler, B. L. (1987) in Progress in
(Biggins, J., ed) V O ~ .11, pp. 53-56,
~ h o t o ~ y n t h e s Research
is
Martinus Nijhoff, Dordrecht, The Netherlands
36. Fish, L. E., Kuck, U., and Bogorad, L. (1985) J . Biol. Chern. 260,
1413-1421
37. petering, D.,F ~ J.~A.,, and palmer, G. (1971) J , ~ i ~then.
l ,
246,643-653
38. Yasunobu, K. T., and Tanaka, M. (1980) Methods Enzyrnol. 6 9 ,
228-238
39. Ohyama, K., Fukuzawa, H., Kohchi, T., Shirai, H., Sano,T.,
S a m S., Umesono, K., Shiki, Y., Takeuchi, M., Chaw,
Aota, S.-i., Inokuchi, H., and Ozeki, H. (1986) Nature 3 2 2 ,
572-574
40. Shinozaki, K., Ohme, M., Tanaka, M., Wakasugi, T., Hayashida,
N., Matsubayashi, T., Zaita, N., Chunwongse, J., Obokata, J.,
Yamaguchi-Shinozaki, K., Ohto, C., Torazawa, K., Meng, B.
Y.,Sugita, M., Deno, H., Kamogashira, T., Yamada, K., Kusuda, J., Takaiwa, F., Kato, A., Tohdoh, N., Shimada, H., and
Sugiura, M. (1986) EMBO J. 5 , 2043-2049
41. Gray, J. C., Blyden, E. R., Eccles, C. J., Dunn, P. P. J., Hird, S.
M., Hoglund, A.-S., Kaethner, T. M., Smith, A. G . , Willey, D.
L., and Dyer, T. A. (1987) in Progress in Photosynthesis Research (Biggins, J., ed) Vol. IV, pp. 617-624, Martinus Nijhoff,
Dordrecht, The Netherlands
42. Adman, E. T., Sicker, L. C., and Jensen, L. H. (1973) J . Bioi.
Chem. 248,3987-3996
43, H
~ T. ~
p., and
~ Woods,
,
K, R. (1981) proc, Natl, ~
~ sei. ~
IJ. S. A. 78, 3824-3828
44. Bruschi, M. H., Guerlesquin, F. A., Bovier-Lapierre, G. E., Bonicel, J. J., and Couchoud, P. M. (1985) J. Biol. Chern. 260,
8292-8296
45. Takahashi, Y., Hase, T., Wada, K., and Matsubara, H. (1983)
Plant Cell Physiol. 2 4 , 189-198
46. Hase, T., Matsubara, H.,Koike, H., and Kat& s. (1983)
Biochirn. Biophys.
Acta
7 4 4 , 46-52
47. Tanaka, M., Nakashima, T., Benson, A., Mower,H., and Yasunobu, K. T. (1966) Biochemistry 5 , 1666-1681
z.,
A and B
48. Golbeck, J. H., and Kok, B. (1978) Arch. Biochern. Biophys. 188,
233-242
49. Golbeck, J. H., and Cornelius, J. M. (1986) Biochim. Biophys.
Acta 849,16-24
50. Kojima, y., Hiyama, T., and Sakurai, H. (1987) in Progress in
Photosynthesis
Research
(Biggins, J., ed) Vol. 11, pp. 57-60,
Martinus Nijhoff, Dordrecht, The Netherlands
51. Malkin, R. (1984) Biochirn.
Acta
Biophys.
7 6 4 , 63-69
52. Morgan, T. V., Stephens, P. J., Devlin, F., Burgess, B.K., and
Stout, C. D. (1985) FEBS Lett. 183, 206-210
53. Fish, L. E., Kuck, U., and Bogorad, L. (1985) in Molecular Biology
of the Photosynthetic Apparatus (Steinback, K. E., Bonitz, S.,
Arntzen, C. J., and Bogorad, L., eds) pp. 111-120, Cold Spring
Harbor Laboratory, Cold Spring Harbor, NY
54. A ~ R., ~~ ~ ~ , ~J., and
~ Vhnngird,
~
t T.~(1981)
o ~ ~i ~ , ~ h i ~ .
p h y ~Acta
.
637,118-123
55. Palmer, G. (1975) in The Enzymes (Boyer, P. D., ed) Vol. 12, pp.
1-56, Academic Press, New York
56, Lundell, D. J., Glazer, A.N., Melis, A., and Malkin, R. (1985) J .
Biol.
Chem. 260,646-654
57. Takahashi, Y., Mitsui, A., Hase, T., and Matsubara, H. (1986)
Proc. Natl.Acad.Sci. U. S. A. 83,2434-2437
58. McDermott, A. E., Yachandra, V. K., Guiles, R. D., Britt, R. D.,
Dexheimer, S. L., Sauer, K., and Klein, M. P. (1987) in Progress
in Photosynthesis Research(Biggins, J., ed) Vol. I, pp. 249-252,
Martinus Nijhoff, Dordrecht, The Netherlands
59. Golbeck, J. H., McDermott, A. E., Jones, w. K., and Kufiz, D.
M. (1987) Biochirn. Biophys. Acta 891,94-98
60. Bonneiea, J. R.7 and Evans, M. c. w. (1984) Biochim. BioPhYs.
Acta 767,153-159
61.d Kirsch,
,
W., Seyer, P., and Herrmann, R. G. (1986) Curr. Genet.
10,843-855
62. Lehmheck, J., Rasmussen, 0. F., Bookjans, G. B., Jepsen, B. R.,
Stummann, B. M., and Henningsen, K. W. (1986) Plant Mol.
Biol. 7,3-10
63. Cushman, J. C., Hallick, R. B., and Price, C. A. (1987) in Progress
in PhotosynthesisResearch (Biggins, J., ed) Vol. IV, pp,
667-670, Martinus Nijhoff, Dordrecht, The Netherlands
64. Bryant, D. A., Lorimier, R. D.,Guglielmi, G., Stirewalt, V. L.,
Cantrell, A,, and Stevens, S. E., Jr. (1987) in Progress in
PhotosynthesisResearch (Biggins, J.,ed) Vol. IV, pp. 749-755,
Martinus Nijhoff, Dordrecht, The Netherlands