|
|
|
|||||||
|
|||||||||
(Received for publication, July 13, 1995; and in revised form, August 8, 1995) From the
In the green alga Chlamydomonas reinhardtii, the
copper-dependent accumulation of plastocyanin is effected via the
altered stability of the protein in copper-deficient versus copper-sufficient medium (t) < 20 min versus several hours). To understand the mechanism of plastocyanin
degradation in vivo, the purified apoprotein was characterized
relative to the holoprotein with respect to conformation and protease
susceptibility. Circular dichroism spectroscopy revealed that the
apoprotein in solution did not display the characteristic secondary
structure displayed by the native or reconstituted holoprotein. The
apoprotein was also susceptible to digestion in vitro by
chymotrypsin whereas the holoprotein was resistant. High ionic
conditions, which stabilize the folded structure of apoplastocyanin,
also inhibit its degradation by chymotrypsin. These results suggest
that one explanation for plastocyanin degradation in copper-deficient
cells in vivo might be the increased susceptibility of the apo
form to a lumenal protease. Since apoplastocyanin is a normal
biosynthetic intermediate for the formation of holoplastocyanin, the
increased susceptibility of apoplastocyanin to proteolysis implies that
degradative and biosynthetic activities would compete for the same
substrate. However, characterization of an apoplastocyanin-accumulating
mutant suggests that a plastocyanin-degrading protease is active only
in copper-deficient cells. Thus, apoplastocyanin is rapidly degraded in
copper-deficient cells, whereas its major fate in copper-supplemented
cells is holoplastocyanin formation.
Plastocyanin is a small (97-104 amino acids),
lumen-localized, copper-binding protein that functions in
photosynthesis to catalyze electron transfer from cytochrome f of the membrane-associated cyt ( In the case of green algae and cyanobacteria,
however, some species have the capacity to survive copper deficiency by
inducing a soluble c-type cytochrome to serve as a functional backup
for plastocyanin (Wood, 1978; Sandmann et al., 1983; reviewed
by Merchant, 1995). In these organisms, if copper is available in
amounts sufficient to satisfy the plastocyanin biosynthetic pathway,
this copper protein accumulates and is used for photosynthetic electron
transfer. Generally, cyt c Regulation of
plastocyanin accumulation has been examined in Chlamydomonas
reinhardtii, Scenedesmus obliquus, and Pediastrum
boryanum among the green algae, and Anabaena and Synechocystis spp. among the cyanobacteria (Merchant and
Bogorad, 1986a, 1986b; van der Plas et al., 1989; Briggs et al., 1990; Bovy et al., 1992; Li and Merchant,
1992; Nakamura et al., 1992; Zhang et al., 1992;
Ghassemian et al., 1994). Examination of mRNA and protein
abundance as a function of copper concentration provides evidence for
copper-responsive regulation at two stages in plastocyanin
biosynthesis: 1) at the level of template accumulation (mRNA abundance)
and 2) at the level of accumulation of mature protein. The relative
contribution of processes regulating mRNA abundance versus protein abundance differs in various organisms and can depend upon
the growth phase of the culture (reviewed by Merchant, 1996). In
general, when both levels of control are displayed, lower
concentrations of copper are required to induce the accumulation of
plastocyanin-encoding mRNAs than are required for the accumulation of
the protein. For the latter, the requirement for copper appears to be
stoichiometric (Merchant et al., 1991). This suggests that the
binding of copper to the polypeptide might be an important factor in
determining the steady state abundance of the protein. The
mechanisms underlying differential accumulation of the protein have
been investigated further in Chlamydomonas by
pulse-radiolabeling experiments. The results indicate that translation
of the message as well as import and processing of the translation
product occur normally and regardless of whether copper is present,
which is consistent with a model in which association of copper with
apoplastocyanin occurs after translocation of the protein into
the thylakoid lumen (Merchant and Bogorad, 1986b; Li et al.,
1990). However, in copper-deficient cells, mature plastocyanin is
degraded (t < 20 min), whereas in copper-supplemented
cells, the newly synthesized protein is extremely stable (t > several hours). If the apoform of the protein were a better
substrate for proteolysis than the holoform, the copper-dependent
differential degradation of plastocyanin could be explained. In
copper-deficient cells, the newly synthesized protein would remain in
the protease-susceptible apoform, whereas in copper-supplemented cells,
it would convert to the protease-resistant holoform. Crystallographic
studies reveal only minimal structural changes between apoplastocyanin
and holoplastocyanin (Garrett et al., 1984); however, in
solution, the apoprotein was noted to be much less stable (Koide et
al., 1993). Thus, it is reasonable to suggest that it might indeed
be more susceptible to one or more chloroplast proteases in vivo compared to the holoprotein, whose conformation is
metal-stabilized (McLendon and Radany, 1978; Parsell and Sauer, 1989;
Koide et al., 1993). There are many well-documented
examples of specific degradation of proteins in chloroplasts. In Chlamydomonas, the small subunit of Rubisco is selectively
degraded if it cannot assemble with large subunits to form the
multimeric holoenzyme (Schmidt and Mishkind, 1983). Likewise, in rye,
the nuclear-encoded subunits of the chloroplast coupling factor are
degraded in the absence of synthesis of the chloroplast-encoded
subunits (Biekmann and Feierabend, 1985). In the case of
chlorophyll-protein complexes, the apoproteins are degraded under
conditions where chlorophyll synthesis is reduced (e.g. Bennett, 1981; Slovin and Tobin, 1982; Kim et al., 1994).
And, for the electron transfer complexes, the pleiotropic effects of
mutations affecting a single subunit of the photosynthetic complexes
have been shown, in many cases, to result from the specific degradation
of the remaining subunits of the affected complex while other
(unaffected) complexes accumulate normally (Erickson et al.,
1986; Kuchka et al., 1989; Pakrasi et al., 1991;
Takahashi et al., 1991; Kuras and Wollman, 1994). Distinct
protease activities have been localized to chloroplasts, but their
relationships to specific degradation processes have not yet been fully
described (Liu and Jagendorf, 1984; Kuwabara and Hashimoto, 1990;
Hoober and Hughes, 1992; Bushnell et al., 1993). Questions of
interest with regard to proteolytic events in the chloroplast relate to
location, regulation, substrate recognition, and substrate specificity.
In the case of plastocyanin, the degradative activity is presumed to be
lumen-localized, or at least thylakoid membrane-associated, since the
fully processed form of the protein serves as the substrate for
degradation. We sought to compare the properties of purified apo-
and holoplastocyanin with respect to protease susceptibility in
vitro and in vivo with a view to increasing our
understanding of 1) proteolytic processes in chloroplasts and 2) the
mechanism of adaptation of the photosynthetic apparatus to
copper-deficiency. The data presented in this work support a model in
which apoplastocyanin is relatively unstable in solution. We also show
that apoplastocyanin is highly susceptible in vitro to
proteases. Nevertheless, the increased susceptibility of
apoplastocyanin to proteolysis may not be sufficient to explain the
regulation of its degradation in vivo. We propose that a
specific mechanism exists for activating an apoplastocyanin-degrading
protease in copper-deficient cells and that this additional factor is
an important determinant of the rate of apoplastocyanin degradation in vivo.
Figure 1:
Difference absorption
spectra of apo-, holo-, or reconstituted plastocyanin. Purified
plastocyanin (B), apoplastocyanin (A), and
reconstituted plastocyanin (C) at 60 µM in 25
mM Tris-Cl (pH 8.05) were used for the measurements. In order
to estimate the amount of copper-containing plastocyanin in different
samples, 10 µl of 100 mM ascorbic acid was added to 1 ml
of apo-, holo-, or reconstituted plastocyanin (prepared as under
``Experimental Procedures''), and the spectrum was recorded
as the reduced spectrum. A spectrum from the same sample to which was
added 10 µl of 100 mM potassium ferricyanide was measured
and recorded as the oxidized spectrum. The difference spectrum was
obtained by subtraction of the reduced spectrum from the oxidized
spectrum. The measurements were made in a 1-ml quartz cuvette (1-cm
pathlength). The peak at 597 nm is characteristic of copper-containing
plastocyanin. The amount of holoplastocyanin in each sample was
estimated from the
Figure 6:
Synthesis of mature plastocyanin in
strain pc235. C. reinhardtii strain pc235 was cultured in
copper-deficient medium (-Cu) or supplemented with
copper chloride (+Cu) prior to radiolabeling with
Na
The structure of the native protein in the crystal form revealed
that it is an eight-stranded, anti-parallel
Figure 2:
Far UV CD spectra of apo-, holo-, and
reconstituted plastocyanin. The plastocyanin content of each sample was
the same (0.56 mg/ml). Each sample was in a solution containing 25
mM Tris-Cl (pH 8.05) in a cuvette with a pathlength of 0.1 mm.
All measurements were made at room temperature. Thin solid
line, apoplastocyanin; thick solid line,
holoplastocyanin; dashed line, reconstituted
plastocyanin.
Analysis of apoplastocyanin preparations by
electrophoresis under non-denaturing conditions generally revealed
three species. In earlier work, we had identified the three species as
apoplastocyanin, a disulfide-linked dimer of apoplastocyanin, and
residual holoplastocyanin (Li and Merchant, 1992). Similar analysis of
spinach apoplastocyanin revealed the same population of species (Li et al., 1990). Bacterial amicyanin also readily dimerizes once
the copper is removed from the holoprotein (Kumar and Davidson, 1992).
To demonstrate more convincingly that the species identified as
apoplastocyanin indeed contained a reactive accessible thiol group (as
opposed to holoplastocyanin where the single cysteine provides a ligand
to the copper, or the dimer where the cysteine participates in the
disulfide bond), the preparation was treated with a fluorescent
maleimide (Fig. 3). As expected, only the band corresponding to
apoplastocyanin was found to react with the thiol reagent.
Figure 3:
Accessibility of the cysteinyl thiol in
apo- versus holoplastocyanin to reaction with maleimidylphenyl
methylcoumarin. Apo- and holoplastocyanin were treated with a
thiol-reactive reagent, maleimidylphenyl methylcoumarin, for 60 min.
Equivalent amounts of apo- and holoplastocyanin were analyzed as
follows after separation by electrophoresis through a 15%
polyacrylamide gel under either non-denaturing (A) or
denaturing (B) conditions: under UV-illumination to detect the
fluorescent reagent (lanes 3, 4, 7, and 8) or by staining with Coomassie Blue R-250 to detect protein (lanes 1, 2, 5, and 6). The arrows on the left point to apoplastocyanin (a),
disulfide-linked dimer of apoplastocyanin (d),
holoplastocyanin (h). The arrows and the numbers on the right indicate the size of the molecular weight
markers separated on the SDS-containing denaturing
gel.
Figure 4:
Protease susceptibility of apo- versus holoplastocyanin. 10 µg each of apo- or holoplastocyanin was
incubated with 0.02 units of chymotrypsin in 100 mM Tricine
(pH 8.0) in a total volume of 30 µl for the indicated amounts of
time. The reaction was terminated by the addition of
phenylmethylsulfonyl fluoride to >100 µg/ml. Portions of the
products (about 200 ng each lane) were loaded onto alternate lanes of a
12% polyacrylamide gel for electrophoresis under non-denaturing
conditions (Davis, 1964), which separates apo- and holoplastocyanin (Li
and Merchant, 1992). The residual protein was visualized by staining
with a silver reagent (Oakley et al., 1980). In panel
B, 0.5 M (NH
It is not
surprising that high salt (which stabilizes the native conformation)
only delays rather than prevents degradation. If the kinetic barrier to
unfolding is small and the two species (i.e. protease-resistant and protease-susceptible) are in equilibrium
(see Koide et al., 1993), the degradation of one form would
eventually lead to the degradation of the entire population of
molecules. Thus, the fact that the apoprotein can adopt a native
structure in the absence of the cofactor does not preclude its
recognition as a non-native structure by a protease. The identity of
the protease in vivo is not known, but the location of the
substrate demands a proteolytic mechanism for plastocyanin degradation
in the thylakoid lumen.
The characteristics of a weakly non-photosynthetic mutant strain of C. reinhardtii, pc235 (see ``Experimental
Procedures''), permitted us to distinguish between these
possibilities. This strain encodes plastocyanin with a wild-type
primary sequence but has reduced plastocyanin function owing to
decreased abundance of holoplastocyanin relative to a wild-type strain (Fig. 5, compare intensity of bands marked h). In fact,
apoplastocyanin accumulates at the expense of holoplastocyanin in
copper-supplemented cells of strain pc235. We conclude that holoprotein
formation is either inhibited in this strain or that the holoprotein,
once formed, is not stable. In either case, the strain provided us with
an experimental background where holoprotein formation is decoupled
from copper supplementation of the medium. The strain was tested for
its ability to synthesize plastocyanin during a brief 15 min labeling
period (Fig. 6). Label incorporation into various soluble
proteins (top panel) or into plastocyanin (bottom
panel) was essentially the same in copper-supplemented or
copper-deficient conditions. Thus, we conclude that strain pc235
synthesizes plastocyanin. Further, the immunoprecipitated species
co-migrates on SDS-containing polyacrylamide gels with mature
plastocyanin synthesized in wild-type cells. The accumulation of
apoplastocyanin in copper-supplemented cells of the mutant strain
therefore suggests regulated proteolysis of apoplastocyanin in the
thylakoid lumen. Apoplastocyanin accumulation in pc235 cannot be
attributed to a specific protease defect because apoplastocyanin fails
to accumulate in copper-deficient medium (Fig. 5) despite the
fact that the mutant is capable of synthesizing plastocyanin under
these conditions (Fig. 6). Thus, we conclude that a
protease-susceptible conformation may be a prerequisite for
apoplastocyanin degradation, but it is not sufficient in vivo.
Figure 5:
Differential accumulation of
apoplastocyanin under copper-supplemented versus copper-deficient conditions. C. reinhardtii strain CC-125 (WT) and plastocyanin-deficient strain (pc235) were cultured in copper-supplemented (+) or
copper-deficient(-) conditions as described under
``Experimental Procedures.'' Cells (usually 100-ml cultures)
were collected at stationary phase (about 1
At the present time, the identity of this protease is not
known. Various types of proteases have been identified in chloroplasts
including metalloproteases and ATP-dependent proteases (Liu and
Jagendorf, 1986; Gray et al., 1990; Hoober and Hughes, 1992;
Bushnell et al., 1993). Although we have noted an
ATP-dependent protease activity in chloroplast extracts that recognizes
apoplastocyanin preferentially over holoplastocyanin as a substrate
(data not shown), the absence of ATP in the thylakoid lumen (see Cline et al., 1992) makes it unlikely that this activity is the
putative lumenal protease responsible for apoplastocyanin degradation.
Our identification of the substrate for this protease (apoplastocyanin)
and our ability to prepare reasonable amounts of radiolabeled substrate
now opens the door for the assay and identification of a lumenal
protease and the copper-responsive system that might regulate it.
Toward this end, we have also characterized preparations of
apoplastocyanin by electrophoretic separation under non-denaturing
conditions. Independently (e.g. Li et al., 1990), a
similar analytical method had indicated that preparations of
apoplastocyanin contained multiple species that differed with respect
to electrophoretic mobility. In this work, we have identified the
species corresponding to apoplastocyanin on the basis of the chemical
reactivity of the cysteinyl thiol (Fig. 3), while in previous
work we identified the species corresponding to cysteine-linked
apoplastocyanin dimers (Li and Merchant, 1992).
Volume 270,
Number 40,
Issue of October 06, pp. 23504-23510, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
EVIDENCE FOR A PROTEASE-SUSCEPTIBLE CONFORMATION OF THE
APOPROTEINAND REGULATED PROTEOLYSIS (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)b
f complex to P700 in Photosystem I, and
in respiration (in cyanobacteria) to catalyze electron transfer from
the cytochrome b
f complex to the
terminal oxidase (reviewed recently by Morand et al., 1994;
Redinbo et al., 1994). It contains a single redox active
copper (E
370 mV) and is referred to
as a ``blue'' copper protein owing to the absorption
properties of the oxidized form of the protein. The catalytic activity
of plastocyanin is, of course, dependent on this copper center.
Accordingly, copper-deficient plants or chelator-treated thylakoid
membranes display loss of plastocyanin activity and impaired
photosynthesis.
is not detected in such
cultures. However, under conditions of copper deficiency (which would
limit or prevent the synthesis of a functional form of plastocyanin),
green algae and cyanobacteria remain photosynthetically competent by
inducing the synthesis of heme-containing cyt c.
These organisms thus serve as excellent model systems for the study of
the copper-dependent synthesis of plastocyanin.
Cell Growth and Culture
C. reinhardtii strain CC125 (wild-type) was grown at 22 °C under 125
µmol/m
/s illumination in either copper-deficient or
copper-supplemented Tris acetate-phosphate (TAP) medium (Merchant and
Bogorad, 1986a; Harris, 1988). A plastocyanin-deficient mutant, strain
pc235, was maintained at room temperature in TAP medium under
fluorescent house lights on the laboratory bench with occasional
swirling. The mutant strain was transferred to an incubator
(15-25 µmol/m/s, 22 °C, with continuous
agitation at 225 revolutions/min) for 1 or 2 days before collecting
cells for isolation of protein or other analysis.Analysis of Protein
Protein concentration was
determined using the Coomassie Blue G-250 dye binding method as
described by the manufacturer of the reagent (Pierce). Cell extracts
containing soluble proteins were separated by non-denaturing gel
electrophoresis (Li and Merchant, 1992), and proteins were visualized
by staining with Coomassie Blue R-250 or silver staining (Oakley et
al., 1980). Western blot analysis of plastocyanin levels was
performed as described previously (Li and Merchant, 1992). Bound
antibody was detected with an alkaline phosphatase-conjugated secondary
antibody (Bio-Rad). To ensure that plastocyanin was quantitatively
released into the supernatant fraction after freeze-thaw lysis of the
cells, the pellet fractions were occasionally analyzed (after detergent
solubilization) for immunoreactive plastocyanin species. In vivo radiolabeling and analysis of the radiolabeled products was
carried out as described previously (Merchant and Bogorad, 1986).Purification of Plastocyanin
Plastocyanin was
purified as described by Redinbo et al.(1993) except that the
final ultrafiltration step on Centricon-30 was eliminated. No
impurities were detected by silver staining of 1 µg of protein
eluted from the G-50 column and analyzed after electrophoresis on
SDS-containing polyacrylamide gels.Preparation of Apoplastocyanin and Reconstitution of
Holoplastocyanin
Apoplastocyanin was prepared by removal of
Cu(I) from purified plastocyanin under a nitrogen atmosphere (see
Tamilarasan and McMillin, 1986; Li and Merchant, 1992). In a typical
reaction, 60 µM plastocyanin in 5 ml of 10 mM sodium phosphate (pH 7.0) was reduced by the addition of sodium
ascorbate (5-fold molar excess) and dialyzed against 500 ml of a
solution containing 25 mM Tris-Cl (pH 8.05) and 10 mM KCN for 2 h at 0 °C. Excess cyanide was removed by dialysis
(also under a nitrogen atmosphere) against a solution containing 25
mM Tris-Cl (pH 8.05) for 3 h with one change. The yield of
apoplastocyanin was about 84-90%. To prepare apoplastocyanin in a
high salt solution, (NH
)SO was
added to 0.5 M into the dialysis buffers. Reconstitution of
holoplastocyanin was performed by direct addition of cupric ions
(10-fold molar excess) in the form of cupric acetate into the above
apoplastocyanin preparation (concentration 60 µM in
total volumes ranging from 1 to 5 ml). Excess copper was removed by
dialysis against 500 ml of a solution containing 25 mM Tris-Cl
(pH 8.05) with one change. Where indicated, the dialysis buffer
included 0.5 M (NH
)SO. All
protein samples (apoplastocyanin and reconstituted holoplastocyanin)
were stored at -20 °C for subsequent analysis including
determination of protein concentration, difference spectroscopy, and
protease susceptibility assays. About 50-85% of the apoprotein
was reconstituted with copper by this method. Copper-containing
plastocyanin was detected by a difference spectrum
(ferricyanide-oxidized minus ascorbate-reduced), and the amount of
holoplastocyanin was estimated using a difference extinction
coefficient of 49 mM cm
at 595 nm. The difference spectra representative of the samples
of plastocyanin used for measurement of CD spectra and protease
susceptibility are shown (Fig. 1). The visible and UV spectra
(not shown) were recorded on a Uvikon 930 spectrophotometer. The
estimation of holoplastocyanin content was not affected by salt
concentration.
A at this wavelength using a
of 49 mM
cm
.
In this particular experiment, the yield of apoplastocyanin was 84%,
and 97% of the apoplastocyanin was reconstituted to holoplastocyanin.
Thus, the total copper-containing plastocyanin in the reconstituted
sample was 97.6%.
Measurement of Circular Dichroic Spectra
The CD
spectra were measured in a Jasco 600 spectropolarimeter (Japan
Spectroscopic Co., Ltd. Japan) in a 0.1-mm cylindrical quartz cuvette
with samples containing 1) purified (apo)plastocyanin, 2) purified
holoplastocyanin, or 3) reconstituted holoplastocyanin (all at 0.56
mg/ml) in a solution buffered with 25 mM Tris-Cl (pH 8.05).
Measurements (at 2-nm intervals, measuring time 1 s, averaged over four
measuring cycles) were made at room temperature. The spectrum of the
solvent was subtracted from the protein spectra.Modification of the Cysteinyl Side Chain in
Apoplastocyanin
To apo- or holoplastocyanin (12 µg in 50
µl of a solution containing 25 mM Tris-Cl (pH 8.05)), 1
µl of 4 mM maleimidylphenyl methylcoumarin (Molecular
Probes Inc., Eugene, OR) in dimethyl formamide was added. After 60 min
at room temperature, the sample was diluted 20-fold with 10 mM sodium phosphate (pH 7.0), and 2 µl of dithiothreitol was
added from a 100 mM stock solution to a final concentration
representing a 50-fold molar excess with respect to the
maleimidylphenyl methylcoumarin concentration. Under these conditions,
the dissociation of copper from holoplastocyanin is avoided. The
protein in the diluted sample was subsequently concentrated in a
Centricon-3 unit (Amicon Corp., Beverly, MA) to remove excess dye and
dithiothreitol. One-third of the concentrated sample was analyzed by
electrophoretic separation in polyacrylamide gels under non-denaturing
conditions to resolve the various forms of plastocyanin (Li and
Merchant, 1992). Each of the samples was also analyzed on an
SDS-containing gel after reduction of disulfide linkages to confirm
that each form indeed comigrated after denaturation.Protease Susceptibility Assay
10 µg of protein
(holoplastocyanin or apoplastocyanin) was incubated at 25 °C with
0.02 units of chymotrypsin (diluted from a 1 mg/ml stock) in a solution
buffered with 100 mM sodium Tricine (pH 8.0) with or without
the addition of 0.5 M (NH)SO as indicated, in a total volume of 30 µl. The reaction was
timed from the addition of protease. At each time interval, a 5-µl
aliquot was removed and transferred to another tube containing
phenylmethylsulfonyl fluoride (1 µl of a 1 mg/ml stock solution in
100% isopropyl alcohol) to stop the reaction. The product was diluted
with H
O so that the final salt concentration was <50
mM. Each sample was analyzed by non-denaturing electrophoresis
in a polyacrylamide gel as described (Li and Merchant, 1992). The
protein was visualized by staining with a silver reagent (Oakley et
al., 1980). The specific activity of chymotrypsin at different
salt concentrations, assessed according to the manufacturer's
instructions using BTEE (N-benzoyl-L-tyrosine ethyl
ester) (Sigma) as a substrate, was found to be independent of salt
concentration within the range tested (0-0.5 M
(NH)SO).Strain pc235
The plastocyanin-deficient strain
pc235 was chosen from a collection of non-photosynthetic mutants
enriched by metronidazole treatment after UV mutagenesis (Harris,
1989). Survivors of the metronidazole enrichment step were tested on
minimal and TAP media to confirm their acetate-requiring phenotype.
Western blot analysis of cell extracts prepared from non-photosynthetic
mutants allowed us to identify strains displaying a plastocyanin-minus
(or plastocyanin-deficient) phenotype. Five such mutants (referred to
as pc strains) were identified. ()The defect in these
strains appeared to be localized to plastocyanin biosynthesis, since
the pc mutants displayed a conditional acetate-requiring phenotype.
Specifically, pc strains were acetate requiring in copper-containing
medium but not in copper-deficient medium (when plastocyanin function
is taken over by cyt c). One of the pc strains,
strain pc235, which exhibited a leaky non-photosynthetic phenotype, was
found to accumulate higher levels of apoplastocyanin, but lower levels
of holoplastocyanin, relative to wild-type strains, despite full copper
supplementation of the medium. Northern blot analysis of the mRNA
isolated from pc235 indicated that it contains normal amounts of
plastocyanin encoding messages. The mRNA encoded a preprotein that
could be precipitated with antibodies raised against plastocyanin and
that co-migrated with the wild-type pre-protein. (
)In
vivo radiolabeling experiments indicated that plastocyanin
synthesis and maturation in this strain was normal (see Fig. 6)
and comparable to a wild-type strain. Further, sequence analysis of the
plastocyanin-encoding genomic DNA in pc235 indicated a wild-type
sequence for the entire transcribed region, including the intron. The
phenotype therefore was concluded to result from a mutation lying
outside the structural gene for plastocyanin. The defect in pc235 is
thus attributed to a trans-acting factor that determines
either the formation of holoplastocyanin from apoplastocyanin or that
stabilizes the holoprotein in vivo after it is assembled. In
either case, most of the plastocyanin in strain pc235 must be
thermodynamically less stable (see ``Results'').
SO
for 15 min. The cells were
sampled after a 100-fold dilution of the radiolabel by the addition of
unlabeled NaSO to 10 mM, and after
addition of copper chloride to the copper-deficient culture to 6
µM. Label incorporation into total soluble proteins in
equivalent volumes of cell extract are displayed in the top
panel. Plastocyanin was quantitatively immunoprecipitated from
equal volumes of soluble cell extract and analyzed by electrophoresis
through a (12%) polyacrylamide gel under denaturing conditions (bottom panel).
Miscellaneous Materials
``Gold-label''
chemicals for preparation of copper-free medium were purchased from
Aldrich. All other chemicals were purchased from either Fisher
Scientific or Sigma, unless otherwise specified.
Preparation of Apoplastocyanin and Its Reconstitution
with Copper
In previous work (Merchant and Bogorad, 1986b), one
of us had noted that newly synthesized plastocyanin was degraded in
copper-deficient cells but was stable in copper-supplemented cells. A
simple explanation for differential degradation was that
copper-deficient cells contained a form of plastocyanin that was more
susceptible to degradation. Since copper-deficient cells were found to
synthesize and process plastocyanin precursors at approximately the
same rate as copper-supplemented cells, it was likely that mature
apoplastocyanin was the protease-susceptible species in vivo.
To test the properties of apoplastocyanin with respect to its structure
and susceptibility to proteases, it was necessary to prepare purified
apoplastocyanin, and this was accomplished by chelation of copper with
cyanide (Fig. 1). Visible spectroscopy confirmed that the
preparation of apoplastocyanin displayed very little absorbance
characteristic of the blue copper center. The yield of apoplastocyanin
in various preparations ranged from 84 to 90%. The ability of
apoplastocyanin to reconstitute with copper indicated that the
preparation was not irreversibly denatured (Fig. 1C). 
-sandwich (Colman et al., 1978; Guss and Freeman, 1983). Removal of the copper
from the crystal did not affect the structure very significantly, and
it was concluded that copper was not required for the folding of the
polypeptide into its native structure (Garrett et al., 1984).
Nevertheless, in vitro folding studies revealed that although
apoplastocyanin was capable of folding in the absence of copper, the
folded structure was not stable in solution in the absence of high
concentrations of salt (Koide et al., 1993). Our preparation
of apoplastocyanin was therefore characterized by CD spectroscopy for
the diagnostic -strand spectrum (Draheim et al., 1986) (Fig. 2). The spectra of the native holoprotein and the
reconstituted holoprotein were very similar. In fact, the estimated
secondary structure of the reconstituted holoprotein was not
significantly different from the secondary structure of the native
protein (Table 1). However, the spectrum of the apoprotein is
clearly altered. Specifically, the apoprotein has significantly reduced
-sheet content, which suggests that it does not display
substantial native structure under these conditions. The fact that the
apoprotein could be reconstituted with copper to yield a holoprotein
that displayed the characteristic spectroscopic properties ( Fig. 1and Fig. 2) indicates that our preparation was
indeed capable of adopting the native structure and was not
irreversibly denatured.
Protease Susceptibility of Apoplastocyanin in
Vitro
The CD spectra, electrophoretic mobility ( Fig. 2and Fig. 3), and increased immunoreactivity (Li and
Merchant, 1992) of preparations of apoplastocyanin are all supportive
of a different conformation of apoplastocyanin relative to
holoplastocyanin. To test whether the apoprotein is indeed more
susceptible to proteases compared to holoplastocyanin, equal amounts of
apo- and holoplastocyanin were separately incubated with proteases, and
the residual protein was analyzed by silver staining after
electrophoretic separation on a non-denaturing gel (Fig. 4).
Apoplastocyanin was found to be highly susceptible to digestion by
either trypsin (not shown) or chymotrypsin (Fig. 4A)
whereas holoplastocyanin was resistant for an extended period of time
during an identical incubation. Conditions (high salt) which stabilize
the folded conformation of apoplastocyanin (Koide et al.,
1993) also inhibit degradation (Fig. 4B). The protease
assays were highly reproducible; identical results were obtained when
the experiment was repeated under the same conditions. These data
suggest that the unfolded conformation is indeed more susceptible to
proteolysis. Since the same results are obtained with trypsin,
chymotrypsin, or (as yet uncharacterized) proteases found in extracts
of chloroplasts (data not shown), it is unlikely that there is a single
particularly susceptible site on the polypeptide.
)SO was
added to the samples before the incubation with protease. The arrows point to the positions of migration of apo- and
holoplastocyanin. The products of digestion are probably too small
(<15 amino acid residues) to be retained on the gel. In the absence
of added protease, apoplastocyanin is not
degraded.
Protease Susceptibility of Plastocyanin in Vivo
A
model for plastocyanin biosynthesis has been developed from
experimental work in numerous laboratories (Grossman et al.,
1982; Hageman et al., 1986; Smeekens et al., 1986;
Kirwin et al., 1988; Hageman et al., 1990; Li et
al., 1990; Bassham et al., 1991; Howe and Merchant,
1993). Pre-apoplastocyanin, synthesized in the cytosol, is translocated
across the envelope membranes of the chloroplast and processed by a
stromal peptidase into intermediate apoplastocyanin. The intermediate
is further transported across the thylakoid membrane into the lumen and
processed by a lumen-facing protease. The mature apoplastocyanin, which
is formed after the second translocation step, is a substrate for two
alternative pathways. Under copper-supplemented conditions,
apoplastocyanin is a substrate for the formation of holoplastocyanin
(Li et al., 1990). Under copper-deficient conditions,
apoplastocyanin is a substrate for degradation (Merchant and Bogorad,
1986b; this work). The inferred existence of a proteolytic mechanism in
the lumen and the susceptibility of apoplastocyanin to such a protease
suggests that degradation of apoplastocyanin could be in competition
for the biosynthetic reaction of holoplastocyanin formation, unless
holoprotein formation in copper-supplemented cells occurs much faster
than proteolysis. Alternatively, there is the possibility that the
proteolytic activity could be regulated by copper. We therefore sought
to determine whether the relative thermodynamic instability of
apoplastocyanin and its increased protease susceptibility in vitro was sufficient to account for its degradation in vivo in
copper-deficient cells, or whether the plastocyanin-degrading
proteolytic activity might actually be regulated in vivo.
10
cells/ml for the wild-type strain and 5 10 cells/ml for strain pc235) and resuspended in a minimal volume
(200 µl) of a solution containing 10 mM sodium
phosphate (pH 7.0). The concentrated cells (equivalent to
1-2 mg chlorophyll/ml) were lysed by two freeze-thaw cycles
(freeze at -80 °C and thaw at room temperature). The
supernatant, collected after centrifugation (15,850
g)
at 4 °C, was identified as the soluble cell extract and separated
by electrophoresis through a 15% polyacrylamide gel under
non-denaturing condition (Li and Merchant, 1992). Total soluble protein
equivalent to 1.0 OD in a 1-ml protein assay was loaded
onto each lane. After electrophoresis, the proteins were transferred
onto a polyvinylidene difluoride membrane which was subsequently
decorated with plastocyanin-specific antibodies. The arrows indicate apoplastocyanin (a), dimer of apoplastocyanin (d), and holoplastocyanin (h). Differences in the
abundance of various species of plastocyanin do not result from
differential release of the protein from the mutant versus the
wild-type cells, or from +copper versus -copper
cells, during preparation of cell extracts.
Degradation of Apoplastocyanin
The degradation
of apoplastocyanin in copper-deficient cells of C. reinhardtii and other algae has been assumed to result from
its decreased thermodynamic stability relative to the holoprotein.
However, the mechanism by which the apoprotein might be recognized as a
suitable substrate for proteolysis over the holoprotein was not well
understood. Unlike the light-harvesting chlorophyll a/b-binding protein, which requires its cofactor for folding
(Paulsen et al., 1993), copper is not required for the folding
of this polypeptide into its native structure. In fact, x-ray
crystallography and solution NMR studies indicate that the structures
of the apo and holoprotein are virtually identical (Garrett et
al., 1984; Koide et al., 1993). Nevertheless, Koide et al.(1993) noted that the folded conformation of the
apoprotein was only slightly more stable than the unfolded
conformation, and further, that the refolding of the unfolded
apoprotein was slow. The differential degradation of plastocyanin in
copper-deficient versus copper-supplemented cells can thus be
easily explained by the differential protease susceptibility of an
unfolded protein compared to a folded one (reviewed by Dice, 1987;
Gottesman and Maurizi, 1992). In this work, we have demonstrated that,
at low ionic strength, apoplastocyanin does not exhibit a native
structure (Fig. 2), and it is indeed highly protease susceptible (Fig. 4). It is not irreversibly denatured under these
conditions as evidenced by its ability to reconstitute with copper ( Fig. 1and Fig. 2); hence, the in vitro preparation may be a fair representation of the polypeptide in
vivo. However, the finding that apoplastocyanin in mutant strain
pc235 is degraded in copper-deficient cells but not in
copper-supplemented cells (Fig. 5) indicates that the increased
susceptibility of apoplastocyanin is only one component of the system.
The other is the protease, which our work suggests might be regulated
by copper.Protein Degradation in Chloroplasts
There are many
examples of specific degradation of proteins in chloroplasts (Mishkind et al., 1985; Vierstra, 1993). The substrates include the
unassembled subunits of multimeric proteins, multimeric complexes
lacking or containing a single mutated polypeptide and the
apopolypeptide of unassembled holoproteins (exemplified by the work of
Apel and Kloppstech, 1980; Bennett, 1981; Merchant and Bogorad, 1986b;
Howe and Merchant, 1992; Kimata and Theil, 1994). It is likely that in
each case the substrates are thermodynamically destabilized relative to
the native structures. However, in some cases their degradation appears
to be a species-specific phenomenon. For instance, in C. reinhardtii, apocyt f is rapidly degraded whereas in
etiolated wheat seedlings, apocyt f accumulates (Anderson and
Gray, 1991; Howe et al., 1994). The difference can probably be
attributed to the lack of a suitable protease in one system versus the other, since the substrate (apocyt f) is probably
equally destabilized in both systems. Likewise, in cyanobacteria, a
mutation in the psaC gene results in the accumulation of a
non-functional PSI complex in the photosynthetic membrane (Mannan et al., 1991) whereas in Chlamydomonas, the same
mutation results in the degradation of the entire complex (Takahashi et al., 1991). Once again, the conservation of PSI structure
between cyanobacteria and eukaryotic chloroplasts suggests that the
complex is perhaps equally destabilized (thermodynamically) in either
system, but might be degraded in one and not the other owing to the
occurrence of a suitable protease in one system but not the other. Our
results (Fig. 5) suggest that the same might be true for
plastocyanin, i.e. its degradation in vivo depends
upon the regulated activity of a lumenal protease.Summary
Two models were proposed to account for
differential degradation of plastocyanin in copper-deficient cells.
First, that copper-deficient cells contained a form of plastocyanin
that was more susceptible to proteolysis (i.e. apoplastocyanin) and second, that copper-deficient cells induced a
plastocyanin-specific protease activity in order to ensure that copper
might be available for other, perhaps indispensable, copper enzymes (e.g. cyt oxidase). In this work, we suggest that both models
are correct. We have demonstrated that apoplastocyanin is the preferred
substrate for proteolysis, but copper-deficient growth conditions are
additionally required for the degradation of apoplastocyanin (in mutant
strain pc235).
)))
We thank Prof. David McMillin and Dr. S. Koide for
helpful discussions concerning the preparation of apoplastocyanin,
Prof. Emil Reisler and his research group for assistance with CD
spectroscopy and the maleimide labeling experiments, Prof. Robert Cohen
for many helpful discussions, and Dr. Jeanette Quinn for performing the
experiment shown in Fig. 6.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. D. Page, J. Kropat, P. P. Hamel, and S. S. Merchant Two Chlamydomonas CTR Copper Transporters with a Novel Cys-Met Motif Are Localized to the Plasma Membrane and Function in Copper Assimilation PLANT CELL, March 1, 2009; 21(3): 928 - 943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Abdel-Ghany and M. Pilon MicroRNA-mediated Systemic Down-regulation of Copper Protein Expression in Response to Low Copper Availability in Arabidopsis J. Biol. Chem., June 6, 2008; 283(23): 15932 - 15945. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Yamasaki, S. E. Abdel-Ghany, C. M. Cohu, Y. Kobayashi, T. Shikanai, and M. Pilon Regulation of Copper Homeostasis by Micro-RNA in Arabidopsis J. Biol. Chem., June 1, 2007; 282(22): 16369 - 16378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Abdel-Ghany, P. Muller-Moule, K. K. Niyogi, M. Pilon, and T. Shikanai Two P-Type ATPases Are Required for Copper Delivery in Arabidopsis thaliana Chloroplasts PLANT CELL, April 1, 2005; 17(4): 1233 - 1251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Eriksson, J. L. Moseley, S. Tottey, J. A. del Campo, J. Quinn, Y. Kim, and S. Merchant Genetic Dissection of Nutritional Copper Signaling in Chlamydomonas Distinguishes Regulatory and Target Genes Genetics, October 1, 2004; 168(2): 795 - 807. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Quinn, J. Kropat, and S. Merchant Copper Response Element and Crr1-Dependent Ni2+-Responsive Promoter for Induced, Reversible Gene Expression in Chlamydomonas reinhardtii Eukaryot. Cell, October 1, 2003; 2(5): 995 - 1002. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Shikanai, P. Muller-Moule, Y. Munekage, K. K. Niyogi, and M. Pilon PAA1, a P-Type ATPase of Arabidopsis, Functions in Copper Transport in Chloroplasts PLANT CELL, June 1, 2003; 15(6): 1333 - 1346. [Abstract] [Full Text]
|
|
|
|
 |
|
 Y. Chassin, E. Kapri-Pardes, G. Sinvany, T. Arad, and Z. Adam Expression and Characterization of the Thylakoid Lumen Protease DegP1 from Arabidopsis Plant Physiology, October 1, 2002; 130(2): 857 - 864. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Quinn, M. Eriksson, J. L. Moseley, and S. Merchant Oxygen Deficiency Responsive Gene Expression in Chlamydomonas reinhardtii through a Copper-Sensing Signal Transduction Pathway Plant Physiology, February 1, 2002; 128(2): 463 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. H. Lu and M. Solioz Copper-induced Proteolysis of the CopZ Copper Chaperone of Enterococcus hirae J. Biol. Chem., December 14, 2001; 276(51): 47822 - 47827. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Quinn, P. Barraco, M. Eriksson, and S. Merchant Coordinate Copper- and Oxygen-responsive Cyc6 and Cpx1 Expression in Chlamydomonas Is Mediated by the Same Element J. Biol. Chem., February 25, 2000; 275(9): 6080 - 6089. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Quinn, S. S. Nakamoto, and S. Merchant Induction of Coproporphyrinogen Oxidase in Chlamydomonas Chloroplasts Occurs via Transcriptional Regulation of Cpx1 Mediated by Copper Response Elements and Increased Translation from a Copper Deficiency-specific Form of the Transcript J. Biol. Chem., May 14, 1999; 274(20): 14444 - 14454. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Itzhaki, L. Naveh, M. Lindahl, M. Cook, and Z. Adam Identification and Characterization of DegP, a Serine Protease Associated with the Luminal Side of the Thylakoid Membrane J. Biol. Chem., March 20, 1998; 273(12): 7094 - 7098. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Zhu, S. Labbe, Ma. M. O. Pena, and D. J. Thiele Copper Differentially Regulates the Activity and Degradation of Yeast Mac1 Transcription Factor J. Biol. Chem., January 16, 1998; 273(3): 1277 - 1280. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Labbe, Z. Zhu, and D. J. Thiele Copper-specific Transcriptional Repression of Yeast Genes Encoding Critical Components in the Copper Transport Pathway J. Biol. Chem., June 20, 1997; 272(25): 15951 - 15958. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. H. Li, J. Quinn, D. Culler, J. Girard-Bascou, and S. Merchant Molecular Genetic Analysis of Plastocyanin Biosynthesis in Chlamydomonas reinhardtii J. Biol. Chem., December 6, 1996; 271(49): 31283 - 31289. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |