Báo cáo Y học: Isolation and characterization of a thioredoxin-dependent peroxidase from Chlamydomonas reinhardtii doc

Isolation and characterization of a thioredoxin-dependent
peroxidase from
Chlamydomonas reinhardtii
Aymeric Goyer
1
, Camilla Hasleka
Ê
s
2
, Myroslawa Miginiac-Maslow
1
, Uwe Klein
2
, Pierre Le Marechal
3
,
Jean-Pierre Jacquot
4
and Paulette Decottignies
3
1
Institut de Biotechnologie des Plantes, Universite
Â
Paris-Sud, Orsay Cedex, France;
2
Department of Biology,
Division of Molecular Biology, University of Oslo, Blindern, Oslo, Norway;
3
IBBMC, Universite
Â
Paris-Sud, Orsay Cedex,

France;
4
Interaction Arbres)Microorganismes INRA/Universite
Â
Nancy, Vandoeuvre Cedex, France
All living o rganisms contain redox systems i nvolving thior-
edoxins (Trx), proteins featuring an extremely conserved and
reactive active site that perform thiol-disul®de interch anges
with disul®de bridges o f target p roteins. In photosynthetic
organisms, numerous isoforms of Trx coexist, as revealed b y
sequencing of Arabidopsis genome. The speci®c functions of
many o f them are s till unknown. In an attempt to ®nd new
molecular targets of Trx in Chlamydomonas reinhardtii,an
anity column carrying a cytosolic Trx h mutated at the less
reactive cysteine of its active site was used to trap
Chlamydomonas proteins that form mixed disul®des with
Trx. T he major protein bound to the column w as identi®ed
by amino-acid sequencing and mass spectrometry as a
thioredoxin-dependent 2Cys peroxidase. Isolation and
sequencing of its gene revealed t hat this peroxidase is most
likely a c hloroplast protein with a high homology to plant
2Cys peroxiredoxins. It is shown that t he Chlamydomonas
peroxiredoxin ( Ch-Prx1) is active with various thioredoxin
isoforms, functions as an antioxidant toward r eactive oxy-
gen species (ROS), and protects DNA against ROS-induced
degradation. Expression of the p eroxidase gene i n
Chlamydomonas was found to be regulated b y light, oxygen
concentration, and redox state. The data suggest a role
for t he Chlamydomonas P rx in R OS detoxi®cation i n the
chloroplast.

Keywords: Chlamydomonas; peroxiredoxin; thioredoxin;
redox signaling; oxidative stress.
1
Peroxiredoxins (Prx) form a ubiquitous group of peroxid-
ases found in bacteria [1], yeast [2,3], animals [4], and, more
recently, in higher plants [5±7]. Prx can be classi®ed
according to the number of conserved cysteine residues:
the 2Cys-Prx subgroup, and 1Cys-Prx subgroup contain
two and one conserved cysteines, respectively. 2Cys-Prx
proteins are reduced by the AhpF protein in bacteria, and
by the thioredoxin/thioredoxin reductase system in yeast
and animals, w hile 1Cys-Prx may be r educed by a small
thiol molecule such as glutathione. R ecently, 1Cys-Prx has
been identi®ed i n y east and Arabi dopsis, and has been
shown to be thioredoxin-dependent and function in a
similar manne r to 2 Cys-Prx [ 8,9].
2Cys-Prx catalyzes, in vitro, the red uction of alkyl
hydroperoxide and h ydrogen peroxide. These enzymes exist
as homodimers. Each subunit contains the two conserved
cysteines that are essential residues for the reduction of
peroxides. The N-terminal cysteine is ®rst oxidized by a
peroxide to sulfenic acid (Cys-SOH), which rapidly reacts
with the C-terminal cysteine of the other s ubunit to form an
intermolecular disul®de [10]. In animals, yeast, a nd plants,
the disul®de is reduced via a thiol/disul®de redox inter-
change with reduced thioredoxin (Trx), thus regenerating an
active peroxidase.
The present study was aimed at setting up an af®nity
chromatography column for speci®c trapping of proteins
that react w ith Trx, b ased on their a bility to f orm mixed

disul®de-linked adducts with a single cysteine mutant
thioredoxin. The system allowed u s to purify and identify
for the ®rst time a 2Cys-Prx p rotein (Ch- Prx1 ) from t he
green a lga Chlamydomonas reinhardtii. The puri®ed protein
was characterized by its antioxidant properties towards
reactive oxygen species (ROS), p rotection of DNA against
degradation, its p eroxidase a ctivity, and its ability t o u se
different thioredoxin isoforms as hydrogen donors. To
better understand the function of Ch-Prx1 in vivo, we
isolated the cDNA and the Ch-Prx1 gene, and examined the
regulation of its expression by different cultur e conditions.
EXPERIMENTAL PROCEDURES
Algal strains and culture conditions
The C. reinhardtii strain CW15 (137c, mt+, cw15, no cell
wall present) and strain CC 125 were obtained from the
Chlamydomonas Genetics Center at Duke University, NC,
USA. Cells were grown in a pho toautotrophic minimal
medium (HSM
2
; [ 11]). CW15 cultures were grow n in ¯asks at
25 °C under c ontinuous stirring and bub bling with 5 % CO
2
enriched air. Light intensity was 3 00 lmolám
)2
ás
)1
at the
level of the culture ¯asks. CC125 cells were grown in 200-mL
Correspondence to M. Miginiac-Maslow, Institut de Bio t echnologie
des P lantes, UMR CNRS 8618, Baà t. 630, Universite

Â
Paris-Sud, 91405
Orsay Cedex, France. Fax: + 33 1 69 15 34 23,
E-mail:
Abbreviations:Nbs
2
,5,5¢-dithiobis(2-nitrobenzoic acid); Prx, perox-
iredoxin; t-BOOH, tertiobutyl hydroperoxide; Trx, thioredoxin; ROS,
reactive oxygen species; NTR, NADPH-dependent thioredoxin
reductase; DCMU, 3-(3¢,4¢-dichlorophenyl)-1,1-dimethyl urea;
DBMIB, 3-methyl-6-isopropyl-p-benzoquinone.
(Received 13 July 2001, revised 29 October 2001, accepted 31 October
2001)
Eur. J. Biochem. 269, 272±282 (2002) Ó FEBS 2002
tubes at 32 °C. Cultures were kept in a 12-h light/dark
regime (light intensity 150 lmolám
)2
ás
)1
) and bubbled
with 2% CO
2
-enriched air. Cell density was adjusted daily
at the onset of light with fresh medium to  2 ´ 10
6
cellsámL
)1
.
Puri®cation of Ch-Prx1
Puri®cation of the native protein from Chlamydomonas

cells. Chlamydomonas CW15 ce lls were grown to a cell
density > 5 ´ 10
6
cells pe r mL. Ce lls were pelleted,
resuspended in 30 m
M
Tris/HCl pH 7.9, and broken by
two cycles o f freeze-thawing i n liquid n itrogen. Broken cells
were centrifuged for 30 min at 15 000 g and the supernatant
was adjusted to 2% (w/v) streptomycin sulfate b y addition
of a 20% solution. After incubation for 20 min at 4 °C, the
suspension was centrifuged at 15 000 g for 30 m in to pellet
precipitated nucleic acids. The supernatant was adjusted to
95% (w/v) a mmonium sulfate a nd incubated for 20 min at
4 °C. The suspension was centrifuged at 15 000 g for
30minandthepelletwasresuspendedin10mL30m
M
Tris/HCl, pH 7.9, and dialyzed against 5 L 30 m
M
Tris/
HCl, pH 7.9. The protein solution was loaded onto an
af®nity column m ade of a mutated Chlamydomonas
cytosolic h-type thioredoxin (C39S mutant) grafted on a
CNBr activated sepharose support and equilibrated with
30 m
M
Tris/HCl, p H 7.9. Covalent coupling of this thio-
redoxin to activated Sepharose was carried out essentially as
recommended b y the supplier of the Sepharo se (Amersham
Pharmacia). The C39S mutant thioredoxin lacks one

cysteine residue and has only its most reactive a ctive-site
cysteine left (Cys36). Site-directed mutagenesis of Trx h and
recombinant protein puri®cation was carried out as
described previously [12]. T o avoid the formation of thio-
redoxin dimers, the single-cysteine mutant thioredoxin was
pretreated with a 40-fold excess of 5,5¢-dithiobis(2-nitro-
benzoic acid) (Nbs
2
; Pierce) before coupling. The deriv-
atized thioredoxin was treated with dithiothreitol to
eliminate t he 5-mercapto-2-nitrobenzoate
3
adduct, and was
washed extensively to remove the dithiothreitol. After
loading on the column, the proteins were eluted with
15 mL of 1 m
M
dithiothreitol in 30 m
M
Tris/HCl, pH 7.9.
The e luted p roteins w ere d ialyzed against 30 m
M
Tris/HCl,
pH 7.9 and re-applied on the af®nity column. P roteins were
eluted with 15 mL of 1 m
M
dithiothreitol in 30 m
M
Tris/
HCl, pH 7.9, and analyzed by SDS/PAGE.

Puri®cation of the recombinant Ch-Prx1 expressed in
Escherichia coli. BL21 (DE3) E. coli strain was trans-
formed with a pET-8c vector (Stratagene) containing a
600-bp fragment obtained by RT-PCR (see below)
corresponding to the coding sequence of C h-Prx1 without
the chloroplast transit peptide. Cells were grown in 5 L of
M9 medium up to D
600
 0.5, and recombinant protein
expression was induced by the addition of 100 l
M
isopropyl t hio-b-
D
-galactoside. A fter induction, cells were
grown for 18 h at 37 °C, pelleted, and r esuspended in
10 mL 30 m
M
Tris/HCl, pH 7.9, 1m
M
EDTA, 500 l
M
phenylmethanesulfonyl ¯uoride, and 14 m
M
2-mercapto-
ethanol. Cells were broken in a French press a t 60 MPa
4
.
Broken cells were centrifuged for 30 min at 20 000 g and
the supernatant was adjusted to 2% (w/v) streptomycin
sulfate by a ddition of a 20% solution. After incubation for

20 min at 4 °C, the s uspension w as centrifuged at 20 000 g
for 3 0 min to precipitate the nucleic acids. The supernatant
was subjected to 35±80% (w/v) ammonium sulfate frac-
tionation. After centrifugation at 20 000 g for 30 min, the
pellet w as re su spend ed i n 10 mL 3 0 m
M
Tris/HCl, pH 7.9,
dialyzed against 5 L o f 30 m
M
Tris/HCl, pH 7 .9, and the
Ch-Prx1 protein was puri®ed as described above on a C39S
Trx h af®nity column equilibrated with 3 0 m
M
Tris/HCl,
pH 7.9.
Polyacrylamide gel electrophoresis
Denaturating [4% ( w/v) SDS] electrophoresis was carried
out on 10% polyacrylamide gels. G els were st ained w ith
CoomassieBlue(2.5gáL
)1
).
Tryptic digestion, separation of the tryptic peptides,
analysis by sequencing and MALDI-TOF mass
spectrometry
Further puri®cation of Ch-Prx1 was achieved, prior to
digestion, by RP-HPLC o n a 4.6 ´ 25 cm Vydac C 4 (30 A
Ê
)
column. C h-Prx1 was eluted with a linear gradient from 2 8
to 70% acetonitrile in 0.1% tri¯uoroacetic acid over 30 min

at a ¯ow rate of 1 mLámin
)1
. Tryptic digestion was
performed for 20 h at 3 7 °Cin0.1
M
NH
4
HCO
3
with
sequence grade trypsin (Boehringer). The peptides were
separated b y RP-HPLC on a 0 .21 ´ 25 cm Vydac C18
(300 A
Ê
) column. Peptides were eluted with a linear g radient
from 0 to 70% acetonitrile in 0.1% tri¯uoroacetic acid, over
90 min, at a ¯ow rate of 0.3 m Lámin
)1
. A bsorbance w as
recorded at 215 nm and 280 nm. Peptides were sequenced
using an Applied Biosystems model 476 A sequencer
equipped with an online phenylthiohydantoin amino-acid
analyser. F or MALDI-TOF analysis, 1 lL of tryptic digest
or HPLC-puri®ed peptides was mixed with 1 lLof
saturated solution of a-cyano-4-hydroxycinnamic acid in
50% acetonitrile, 0.3% tri¯uoroacetic a cid. Samples were
loaded into MALDI-TOF spectrometer (Perseptive B io-
systems, Voyager STR-DE) equipped with a nitrogen laser
(337 nm). Spectr a were obtained in re¯ectron mode using
delayed extraction.

RT-PCR
Ch-Prx1 cDNA was isolated by RT-PCR. In this approach,
primers used in cDNA synthesis w ere designed based on the
amino-acid sequence of N- and C-terminal peptides, a s
determined by the p rocedure described above. First-strand
cDNA was ®rst synthesized from total RNA with
M-MuLV reverse t ranscriptase (Life T echnologies). The
reaction mixture c ontained i n a volume of 20 lL: 0.9 lg
heat-denatured Chlamydomonas total RNA, 1 ´ RT buffer,
1m
M
dNTPs, 20 m
M
dithiothreitol, 100 U reverse tran-
scriptase (RT), and 1 l
M
of the degenerated reverse primer
5¢-GCGGATCCTTA(G/C)ACGGCGGCGAAGTAC
TCC-3¢. After reverse transcription for 30 min at 42 °C, the
®rst-strand cDNA was ampli®ed in a PCR performed under
the following conditions: 5 min at 96 °C; 39 cycles (94 °C
for 1 min, 64 °C for 2 min, and 72 °Cfor2min).In
addition to the above degenerated reverse primer two
direct primers 5¢-AACCATGGCCTCCCACGCCGAGA
AGCC(G/C)CTG-3¢ and 5¢-AACCATGGCCAGCCAC
Ó FEBS 2002 Peroxiredoxin from Chlamydomonas (Eur. J. Biochem. 269) 273
GCCGAGAAGCC(G/C)CTG-3¢ were used. PCR products
were separated b y electrophoresis on a 0.8% agarose g el.
A 600-bp fragment was puri®ed by using the nucleospin
extract k it (Macherey±N agel), then digested by NcoIand

BamHI, and cloned i nto the pET-8c vector.
Screening of a cDNA library
A kgt11 cDNA library of Chlamydomonas was a gift from
M. Goldschmidt-Clermont (Universite
Â
de Gene
Á
ve, S witzer-
land). The Ch-Prx1 coding sequence obtained b y R T-PCR
was used to screen 200 000 plaque forming units of the
kgt11 library. Colony hybridization was performed at 64 °C
in buffer containing 0.5
M
NaHPO
4
,pH7.2,1m
M
Na
2
EDTA, 7% SDS, and 1% BSA. After p uri®cation of
positive clones, k DNA was extracted and puri®ed as
described in [ 13]. cDNA inserts were excised by a digestion
with EcoRIandclonedintotheEcoR I site of the SK
+
Bluescript plasmid (Stratagene).
Screening of a gene library
A B AC library of genomic C. reinhardtii DNA (Genome
Systems Inc., St Louis, USA, product FBAC-8417)
spotted at high density on a nylon ® lter was used to
isolate a clone containing the c omplete Ch-Prx1 gene. Two

heterologous primers, (5 ¢-GACTTCACCTTCGTGTGCC
CCACCGAG-3¢ and 5¢-GGGGTCGATGATGAACAG
GCCGCG-3¢), designed from the conserved 5¢ and
3¢ sequences of known 2Cys peroxiredoxin genes and
optimized for the codon usage of Chlamydomonas,were
used to amplify by PCR from genomic Chlamydomonas
DNA a fragment (  780 bp) of the Chlamydomonas
peroxiredoxin gene. The PCR fragment was cloned,
sequenced, and used as a probe to identify a Chlamydo-
monas BAC clone that contains the Ch -Prx1 gene sequence.
Two BAC clones that strongly hybridized to the probe on
the ® lter were am pli®ed and used to i solate the complete
Ch-Prx1 gene sequence using conventional S outhern ana-
lyses, subcloning, and sequencing techniques [ 13].
Sequencing of cDNA and genomic DNA
The BigDye T erminator Cycle Sequencing
5
Kit (Perkin-
Elmer) or the Thermo Sequenase Radiolab eled Terminator
Cycle Seq uencing Kit (United States Biochemicals) were
used to sequence the Ch-Prx1 cDNA and t he Ch-Prx1 gene,
respectiv ely.
RNA isolations
RNA for RT-PCR and for northern a nalyses was islolated
by alternative m ethods. I n t he ®rst method,  30 million
cells were pelleted by centrifugation (3000 g,5min).The
pellet was immediately resuspended in 1 mL TRIzol reagent
(Gibco BRL) and polysaccharides, membranes, and unlysed
cells were eliminated by centrifugation (12 000 g,10min).
The supernatant was treated as instructed by the supplier.

The dried pellet was resuspended in 20 lL of milliQ
(Millipore) puri®ed sterile water. The second method
followed e ssentially a protocol described previously [14,15].
PolyA
+
RNA was isolated with magnetic oligo d(T) b eads
(Dynal) f ollowing the protocol of t he supplier.
Southern and Northern blot analysis
For Southern analyses, genomic DNA was prepared
following the protocol described previously [16]. DNA
was digeste d w ith a ppropriate restri ction e nzymes, s ize-
fractionated on a 0.8% a garose g el, and transferred t o a
Hybond N
+
(Amersham Pharmacia) o r Z etaProbe (Bio-
Rad) nylon membrane. Hybridization with t he
32
P-radio-
labeled cprx probe (4 ´ 10
6
c.p.m.) and washin g of t he gel
blot at 65 °C was carried out as described previo usly [ 17].
The membrane was exposed to X-ray ® lm at )20 °Cor,
when using an intensifying s creen, at )80 °C.
RNA for Northern analyses was isolated as described
above and separated in a 1.3% agarose/formaldehyde gel.
The RNA was b lotted to a nylon membrane (ZetaProbe,
Bio-Rad) by alkaline t ransfer, U V-crosslinked, and hybrid-
ized to the random primer radiolabeled Ch-Prx1 cDNA
probe for 2 4 h. T he cDNA p robe was g enerated using

biotinylated single stranded template bound to magnetic
streptavidin-coated beads (Dynal) in a speci®c priming
reaction [18]. The speci®c primer used for the reaction was
the downstream primer used in the RT-PCR reaction. After
washing [17], the membrane was exposed t o X-ray ®lm with
an intensifying screen at )80 °Cfor 2days.
Antioxidant activity of the Ch-Prx1 protein
The antioxidant activity of the Ch-Prx1 protein was tested in
a DNA-cleavage assay modi®ed after [19,20]. Bluescript
plasmid DNA (2 lg) was exposed to a mixed function
oxidation system containing 0.32 m
M
dithiothreitol and
3 l
M
FeCl
3
. The reaction contained various amounts of
concentrated Ch-Prx1 p rotein (5±20 l
M
), and was initiated
30 min before addition of the DNA. Control reactions were
performed without Ch-Prx1 protein and with BSA
(400 lgámL
)1
). The reactions were stopped by adding
3.3 m
M
EDTA and analyzed on an agarose gel.
Thioredoxin-dependent peroxidase activity of Ch-Prx1

Peroxidase activity assays were initiated by the addition of
H
2
O
2
(500 l
M
)ort-b utyl hydroper oxide (t-BOOH)
(500 l
M
)to1mL30m
M
Tris/HCl, pH 7.9, reaction
medium containing 197 n
M
A. thaliana NADPH Trx
reductase (NTR), 180 l
M
NADPH, 5.7 l
M
Trx and
2.4 l
M
Prx. The reaction was monitored s pectrophotomet-
rically by following the decrease in absorbance at 340 nm at
30 °C. Recombinant A. thaliana NTR was puri®ed as
described p reviously [21]. C hloroplastic thioredoxins m and
ffromspinachwereakindgiftofP.Schu
È
rmann (University

of Neuchaà t el, Switzerland). Thioredoxins m and h from
Chlamydomonas were puri®ed as described p reviously [22].
An alternative assay, avoiding t he need of a t hioredoxin
reductase, was also used, based on colorimetric determina-
tion of hydrogen peroxides or alkyl hydroperoxides with the
PeroXOquant kit (Pierce) following the supplier's recom-
mendations. Ch-Prx1 (43.8 l
M
) was incubated w ith 400 l
M
dithiothreitol, 16.6 l
M
Trx a nd 500 l
M
t-butyl hydroper-
oxide in 50 lLof30m
M
Tris/HCl, pH 7.9, buffer. The
quantity o f t-BOOH was measured on 5 lL a liquots added
to a spectrophotometer cuvette containing 500 lLof
PeroXOquant medium. The activity was estimated from
the decrease in absorbance at 595 nm.
274 A. Goyer et al. (Eur. J. Biochem. 269) Ó FEBS 2002
RESULTS
Isolation of a 2Cys-Prx by using a single cysteine
mutant of
Chlamydomonas
Trx h
In an attempt to isolate new T rx targets in Chlamydomonas,
a s trategy w as used based o n the formation o f stable mixed

disul®des between a Trx, mutated in its less reactive active-
site cysteine, and its potential targets. The approach has
previously been used successfully to characterize in vivo
complexes o f thioredoxins w ith s ome of its target proteins
[8,23] and to identify in vitro the most reactive internal
cysteine (Cys207) of Sorghum NADP-MDH [12]. The
cytosolic Trx h of Chlamydomonas has only two cysteines in
its primary sequence, both belonging to the active site
disul®de. T herefore, t his excludes the poss ibility of Trx
forming artefactual disul®des
6
with th e a dditional cysteines
present in t he chloroplastic Trx m a nd f sequences. Many
Trx targets, s uch a s NADP-MDH o r phosphoribulokinase
can be a ctivated in vitro by various thioredoxin isoforms:
chloroplastic Trx f or m [24,25] but also cytosolic Trx h
[22,26]. We took advantage of this lack of speci®city t o try
to isolate various putative Trx targets.
AmutatedChlamydomonas cytosolic Trx h, in which
only the most reactive cysteine (Cys36) remained, was
coupled to an activated CNBr Sepharose column (see
Materials and methods for details on preparation of the Trx
af®nity c olumn). Among the Chlamydomonas proteins that
were retained on the column after loading of a protein
extract and extensive washing, was a major protein of
 21 kDa (Fig. 1). This p rotein was further puri®ed to
homogeneity by HPLC and digested with t rypsin. The
tryptic peptides were separated by HPLC and some of them
were totally or partially analyzed by Edman sequencing
and/or by MALDI-TOF mass spectrometry ( Fig. 2).

Computer database searches based on the amino-acid
sequences of sequenced peptides revealed 75% identity
with a thioredoxin-dependent peroxidase (TPx), also named
peroxiredoxin, of barley (the BAS1 protein) and Arab idop-
sis. These proteins belong to the 2Cys-Prx group, because of
the p resence of two conserved cysteines [5,6]. Arabidopsis
BAS1 was s hown to be a chloroplastic protein. The identity
of our peptides with BAS1 and the presence of two cysteines
in alignment with the conserved cysteines of barley and
Arabidopsis BAS1 suggested that our 21-kDa protein also
belongs to this protein family, and could b e chlo roplastic.
We called the Chlamydomonas 21-kDa protein Ch-Prx1.
Cloning and sequences of
Ch-Prx1
cDNA
and the
Ch-Prx1
gene
In order to complete the sequence data for this new protein,
and to be able to make a thorough characterization of its
biochemical properties, we isolated the cDNA and
expressed it in E. coli to produce a pure recombinant
protein. We also isolated and sequenced the gene e ncod-
ing the 21-kDa polypeptide. Degenerate oligonucleotides
designed from the s equences of N- and C-terminal peptides
(P1 and P10) were us ed as primers to s ynthesize t he cDNA
of the coding sequence of t he mature Prx. The direct P CR
primers were synthesized with a 6-bp extension at their
5¢ ends (ATGGCC, encoding methionine and alanine) for
translation i nitiation and in frame cloning. The ampli®ed

product, cloned into the pET-8c vector, was sequenced
showing that the 597-bp product encod ed the putative
Ch-Prx1 (accession number AJ304857).
To isolate t he full-length cDN A, the RT-PCR product
was u sed to s creen a cDNA library. A clone carrying a 1244-
bp fragment was isolated and sequenced. Unfortunately, the
cDNA sequence in this clone was not complete. It contained
the 591 bp sequence c oding for the mature protein, a 647-bp
sequence corresponding to the 3¢ region including the polyA
tail, plus 6 bp in the 5 ¢ region. Screening of a Chlamydo-
monas genomic BAC library (Genome Systems Inc., St
Louis, USA) resulted in the isolation of a clone that
contained a 1946-bp sequence corresponding to the g ene
coding for t he 21-kDa protein of Chlamydomonas.We
named this gene Ch-Prx1 (accession number AJ304856).
The 3¢ end of the gene was not complete but could be
deduced from the sequence of t he cDNA. The gene contains
two introns and three exons. A 12-bp exon separates both
introns (data not shown). When the amino-acid sequence
deduced from the gene s equence w as compared to the
sequence of the N-terminal peptide of the mature protein,
codons for 3 8 a dditional amino acids, which were not
present i n the mature protein, were discovered at the 5 ¢ end
LMW
Elution product
s
kDa
1
4
.

4
2
0
.
1
3
0
4
3
6
7
9
7
21 kDa
Fig. 1. An alysis of elution products f rom the thioredoxin anity column
by red ucing SDS/PAGE. Protein extracts o f Chlamydomonas cu ltures
were app lied on a cytosolic Trx h C39S mutant anity column. The
elution was performed with d ithioth reitol.
Ó FEBS 2002 Peroxiredoxin from Chlamydomonas (Eur. J. Biochem. 269) 275
of the c oding region (Fig. 2). A chloroplast transit peptide
prediction program (
CHLOROP
) [27], predicted a putative
cleavage site between arginine 37 an d alanine 38, but the
mature protein starts at Ser39, as indicated by the peptide
sequencing (Fig. 2). It is possible that the protein c leaved
between positions 37 and 38 is further processed in the
chloroplast to the native form.
Southern blot analys is on genomic DNA digested with
ApaI ( an enzyme known t o c leave w ithin the Ch-Prx1 gene

sequence), or EcoRI (an enzyme t hat does not cleave within
Ch-Prx1 gene), produced four and two fragments, respec-
tively, that h ybridized to the radiolabeled Ch-Prx1 coding
sequence (Fig. 3). These results indicate that an additional
gene homologous to the Ch-Prx1 gene exists in C. rein-
hardtii genome. In this respect, it c an be noted t hat a
sequence o f a putative cytosolic Chla mydomonas Prx is
available in d ata b anks.
Amino-acid sequence comparisons
We compared the C h-Prx1 amino-acid se quence t o other
similar proteins. Three groups could be distinguished
(Fig. 4 ). The ® rst group contains Prx with one conserved
cysteine but these p roteins s eem to be functionally closer to
2Cys-Prx than to 1Cys-Prx. Members of the second group
contain t wo con served c ysteines. C h-Prx1 belongs to this
group and i s close to the higher p lant 2Cys-Prx pr oteins,
which have been described as nuclear-encoded chloroplastic
proteins. This suggests, in addition to the presence of a
transit peptide, that our 2Cys-Prx is a chloroplastic protein.
The third group contains Prx with one conserved c ysteine.
Biochemical characterization of recombinant Ch-Prx1
Under o xidizing conditions, 2Cys-Prx e xists p redominantly
as dimers linked by two identical d isul®de bonds between
the ®rst Cys of one subunit and the s econd Cys of the other
subunit [10]. Ch-Prx1 also shares this feature: under
reducing conditions it is a monomer o f 21 kDa, which is
converted to a dimer under nonreducing c onditions (Fig. 5).
The antioxidant activity of Ch-Prx1 was characterized in a
mixed function oxidation system ( Fig. 6). Prx proteins are
Fig. 2.

BLAST
of peptide sequences of 21 kDa proteins in databases. Tryptic peptides were puri®ed by RP-HPLC and some of them were to tally or
partially analyzed by Edman sequencing (ááá) and/or by MALDI-TOF mass spectrometry (±±). The experimental masses (M + H)
+
were compared
with the c alculated masses ( indicated in brackets): P1, 1 681.71 (1681.89); P2 , 1853.42 (1853.93); P4, 297 2.26 (2972.51); P5, 163 1.07 (1630.87); P 6,
1393.66 (1393.73); P7, 2512.52 (2512.36). Accession numbers: barley BAS1, Z34917; Arabidopsis BAS1, X97910. The missing amino-acid stretches
and the transit peptide sequence were deduced from the cloned c DNA and gene sequences (accession numbe rs: AJ304856 for t he ge ne and
AJ304857 f or the cDNA). T he conserved r esidues are shade d in grey and the sequence of the putative t ransit peptide i s in i talics.
ApaI
EcoRI
12
10
4
1.7
1.4
kb
Fig. 3. Sou thern blot analysis of the Ch-prx1 gene. Ge nomic DNA
from CW15 Chlamydomonas strain was digested with ApaIorEcoRI,
size-fractionated on a 0.8% agar ose gel, and transferred to a Hybond
N
+
nylon m embrane. Th e ® lter was hybridized with the
32
P-radiola-
beled Ch-Prx1 c oding reg ion probe. M olecular siz e markers a re indi-
cated on t he left.
276 A. Goyer et al. (Eur. J. Biochem. 269) Ó FEBS 2002
known t o p revent damage of DNA against R OS. ROS can
be produced by incubating dithiothreitol with Fe

3+
,which
catalyzes the reduction of O
2
to H
2
O
2
. The latte r is further
converted by the Fenton reaction to hydroxyl radicals
[19, 20]. The radicals produced by incubating dithiothreitol
with Fe
3+
caused complete degradation of 2 lg pBluescript
DNA within 1 h . Ch-Prx1 protected the pBluescript DNA
against degradation, while BSA, even a t a concentration of
Fig. 4. Phy logenetic tree of peroxiredoxins.
CLUSTAL X
was used t o generate the tree . Accession numbers: Trypanosoma-mpx, AJ006226; Try-
panosoma, u26666; yeast: Type I T Px, N P013684, type II TPx, u53878, 1Cys-Prx, ybl0524; Chinese cabbage C2C-Prx, AF052 202; human: A OE37-2,
u25182, AOP1, P30048, pag, q06830, NKEFB, l19185; barley: BAS1, Z34917, PER1, X96551; Arabidopsis: BAS1, X97910, TPx2, AF121356,
MHF15.19, AF326871, Per1, O04005; Chlamydomonas Ch-Prx1, AJ304857; spinach B AS1, ´ 94219; mouse TPx, u20611; Bromus pbs128, p52571;
Tortula, u40818; Dro sophila DPx-2540±2, AF311880; Rattus Prdx3, NM022540; Oryza, AF203879; Phaseolus, AJ288896; Thermus aquaticus,
AF276071; Br assica Per1, A F139817, BAS1, A F311863.
Ó FEBS 2002 Peroxiredoxin from Chlamydomonas (Eur. J. Biochem. 269) 277
20 l
M
, h ad no effect (not shown). T he degree of protection
correlated with the a mount of Ch-Prx1 a dded to the a ssay.
Thioredoxin-dependent peroxidase activity of Ch-Prx1

towards H
2
O
2
or t-BOOH was e xamined indirectly by
measuring the oxidation r ate of NADPH (followe d by the
decrease in A
340
) in t he presence of NTR and thioredoxin.
Pure recombinant proteins expressed i n Escherichia coli
were used in this test: NTR from A. thaliana,Trxhfrom
Chlamydomonas and C h-Prx1. Figure 7A shows a tim e-
course of NADPH oxidation with either H
2
O
2
or t-BOOH
as substrates. C learly, C h-Prx1 was equally ef®cient with
both, and the reaction required all three protein components
(Ch-Prx1, Trx h and NTR). The speci®c activity of the
recombinant e nzyme w as identical to the speci®c activity of
the native protein puri®ed fro m Chlamydomonas (dat a not
shown). The rate of NADPH oxidation was very weak
when Trx m from Chlamydomonas was used (data not
shown), probably b ecause o f t he weaker af® n ity of Ar abid-
opsis NTR for Trx m [21]. Therefore, the ability of Trx m
from Chla mydomonas to donate protons to Ch-Prx1 w as
measured directly by following the degradation of t-BOOH,
A
A340nm

0
0,2
0,4
0,6
0,8
1
1,2
0 5 10 15 20 25
Time (min
)
B
0
20
40
60
80
100
120
0 2 4 6 8 10
Time (min)
% t-BOOH initial
Fig. 7. Per oxidase activity of recombinant Ch-Prx1 toward H
2
O
2
or
t-butyl hydr operoxide. Dependence o f the re ac tion on re duce d thiore-
doxin. (A) The r eaction was followed spectr ophotometrically by
NADPH oxidation in a cou pled assay with N TR from Ar ab idopsis and
thioredoxin h from Chlamydomonas. C omplete assays with H

2
O
2
(e)
or t-BOOH (j). Controls: m in us T RX ( m), minus PRX (r), minus
peroxide (s), minus NTR (q). (B) Peroxidase activity w ith various
thioredoxins. T he concentration of t-BOOH was measured colori-
metrically and expressed a s a percentage of the i nitial c oncentration.
Thioredoxins were reduced with dithiothreitol. Chlamydomonas Trx h
(m)orTrxm(j), spinach T rx f (d)orTrxm(r). Control w ithout
Trx (n).
SK DNA
20
µ
M
Ch
-Prx1
5
µ
M
Ch
-Prx1
C
OC
S
Fig. 6. Prote ction of DNA against free r adical attac k by the recombi-
nant Ch-Prx1 protein. Plasmid Bluescript SK DNA was incub ated for
30 min in a thiol-MFO system containing 3.0 l
M
Fe

3+
.Lane1is
untreated SK DNA, open circle (OC) and superc oiled (S) plasmid
DNA are indicated; lanes 2±3 containing 20 or 5 l
M
Ch-Prx1 protein
show that protection of DNA against nic king increases with increasing
amounts of prote in; lane 4 ( C ) contro l without pro tein s hows maxi-
mum DNA degradat ion.
LMW
+
ββ
ββ
-
Mercapto
-
ββ
ββ
-
Mercapto
1
4
.
4
2
0
.
1
3
0

4
3
6
7
9
7
k
D
a
Fig. 5. Ana lysis by SDS/PAGE of the monomer/dimer shift of recom-
binant Ch-Prx1. The protein was e ith er r educed with 2-mercapto-
ethanol, or not, a s indicated.
278 A. Goyer et al. (Eur. J. Biochem. 269) Ó FEBS 2002
using the peroXOquant kit (see Experimental procedures),
in the presence of dithiothreitol as an electron donor to Trx.
The r ate of d isappearance of t-BOOH was i dentical with
TrxmandTrxhfromChlamydomonas used at the same
concentration ( Fig. 7B). When Trx was omitted, the r ate of
disappearance of t-BOOH was negligible, proving that
dithiothreitol alone, used at low concentration, cannot
signi®cantly act ivate Ch-Prx1. In order to d etermine
whether d ifferent chloroplastic T rx differ in their abilities
to function with Ch-Prx1, we compared the e f®ciencies of
two Trx isoforms f and m from spinach, b ecause no T rx f
from Chlamydomonas has been isolated until now. Both
were active with Ch-Prx1, but while Trx f was as ef®cient as
Trx h and m from Chlam ydomonas, T rx m from spinach
showed a lower ef®ciency.
Regulation of
Ch-Prx1

expression
To study possi ble roles of Ch-Prx1 in vivo , the expression of
the Ch-Prx1 gene was monitored under different culture
conditions.
A number of genes have been reported to be regulated by
light, among them the Trx m and Trx h genes from
Chlamydomonas [28]. As Prx are Trx-dependent proteins,
we investigated a possible regulation of the Ch-Prx1 gene
expression by light. Under dark conditions, levels of
Ch-Prx1 gene transcripts are relatively low (Fig. 8A).
Ch-Prx1 gene transcript levels increased in illuminated cells
reaching a maximum after  6 h of light. In the dark,
Ch-Prx1 mRNAs returned to the basal level in less than 2 h.
These results show that the pool size of Ch-Prx1 transcripts
is affected by light/dark conditions.
The role o f noncyclic photosynthetic electron transport in
regulating the expression of the Ch-Prx1 gene was s tudied
by blocking electron transfer w ith 3-(3¢,4¢-dichlorophenyl)-
1,1,-dimethyl urea ( DCMU) or 2 ,5-dibromo-3-methyl-6-
isopropyl-p-benzoquinone (DBMIB)
8
. Light induction of
Ch-Prx1 gene expression was a ffected by bo th inhibitors
(Fig. 8B, lanes 3 and 4), indicating that photosynthetic
electron transport is required for upregulating Ch-Prx1 gene
expression. This regulation does not imply the redox state of
plastoquinone, suggesting that another element situated
downstream the plastoquinone is responsible for this
regulation.
The highest levels of Ch-Prx1 transcripts were f ound in

cells bubbled with pure o xygen for 90 min in the dark
(Fig. 8C, lane 2). This shows that oxidative s tress, directly or
indirectly, affects Ch-Prx1 gene expression.
Taken together, the results support the notion that the
redox state and/or the concentration of reactive oxygen
species in the chloroplast play a role in regulating the level of
transcripts of the Ch-Prx1 gene in Chlam ydomonas.
DISCUSSION
The mixed disul®de approach as a tool to isolate new
thioredoxin targets
The formation of stable mixed-disul®de cross -linked com-
plexes has been used p reviously to determine interactions
between target enzymes, such as phosphoribulokinase or
NADP-malate dehydrogenase, and thioredoxins [12,29].
This approach also provid ed evidence for conformational
changes occurring in the structure of thioredoxin reductase
upon interaction with its substrate thioredoxin [30]. U sing
the mixed disul®de approach for isolation of Trx target
proteins in vivo is dif®cult because of the limited stability o f
Trx-target complexes in c ells in which reductants t hat split
disul®des, are abundant. To o vercome this dif®culty,
thioredoxin-de®cient yeast or E. coli mutant strains have
been used to express a single-cysteine m utant thioredoxin
allowing the i solation of thioredoxin targets in y east [8] and
Fig. 8. Ch -Prx1 gene expression. (A) Levels of Ch-Prx1 transcripts in cells growing in 12-h light/dark cycles. Total RNA was extracted in two hours
intervals from division-synchronized Chlamydomonas cells kept in a 12- h light/dark regime. RNA samples were processed as described in Materials
and methods. RNA gel blots were hybridized to the radiolabeled Ch-Prx1 cDNA probe. Numbers ab ove the lan es indicate the time a t which the
RNA samples were take n. (B) I nduction of Ch-Prx1 gene expression in Chlamydomonas by light in the abse nce an d in th e prese nce of DCMU or
DBMIB. Cultures were grown in continuous light and R NA samples were analyzed for Ch-Prx1 transcript levels by northern analysis. Lane 1, cells
taken after 16 h in the dark; lane 2, cells taken 3 h a fter the start of the light period; lane 3, cells taken a fter 3 h in the light in the presence of 20 lM

DCMU (added at 0 h light); lane 4, cells taken after 3 h in the light in the p resence of 1 lM DBMIB (added at 0 h light). (C) Induction of Ch-Prx1
gene expression in the dark by bubbling with oxygen. Cultures were grown in 12-h light/dark cycles. Lane 1, cells taken after 11 h in the d ark; lane 2,
cells taken a fter 11 h in the dark after 90 min bubbling with 100% O
2
.
Ó FEBS 2002 Peroxiredoxin from Chlamydomonas (Eur. J. Biochem. 269) 279
in E. c ol i [23]. B ecause T rx-de®cient mutants o f Chlamydo-
monas are not available, we combined the mixed-disul®de
method with af®nity chromatography, u sing an af®nity
column m ade of Trx h (C39S Trx h) mutated at the less
reactive Cys of i t s active site. T he major protein retained on
the af®nity column was a thioredoxin-dependent peroxidase
(Prx) that b elongs to t he same family as a peroxidase
isolated in vivo in yeast using mutated Trx of Arabidopsis [8].
The predominant formation of mixed disul®des between
Prx and C39S Trx h in the presence of a number of well-
known T rx-dependent enzymes appears surprising but may
be explained by t he natural abundance of t he peroxidase
that might compete with other targets, but also by structural
features of the r egulatory sites of the various target proteins.
Extensively studied enzymes, such as NADP-MDH [12] and
fructose bisphosphatase [31], can be slowly activated by a
single cysteine mutant thioredoxin while almost no complex
between the mutant thioredoxin and wild-type MDH is
formed [12]. Oxidized Prx, o n the other hand, is linked by
intermolecular disul®de bonds. Upon cleavage by Trx the
subunits separate leaving no proximal Cys that could attack
the mixed disul®de formed between Trx and its target. Thus,
the mixed disul®de approach seems to favour the isolation
of targets bearing an intermolecular disul®de bridge.

Structural and functional characteristics of Ch-PRX1
The amino-acid sequence of Ch-Prx1 shares highest identity
with the BAS1 protein of Brassica, spinach, barley and
A. thaliana,PR1ofPhaseolus and MHF
9
of A. thaliana.
These proteins b elong to the 2Cys-Prx subfamily. A ll plan t
2Cys-Prx proteins, except BAS1 of barley, the complete
cDNA of which has not been isolated, contain putative
chloroplast-targeting sequences. C h-Prx1 is likely to be a
chloroplastic protein because it is synthesized as a precursor
protein containing a short transit peptide that is predicted to
be cleaved at a conserved s ite. The homologous BAS1
protein of Ara bidopsis wasshowntobeimportedinto
isolated plastids after post-translational modi®cation [6].
Like other 2Cys-Prx enzymes previously described i n
yeast, mammals and plants, Ch-Prx1 d isplayed antioxidant
and peroxidase activities. The enzyme could reduce hydro-
gen peroxide as well as alkyl hydrogen peroxide and exerted
a s trong protective effect against DNA degradation by f ree
radicals of oxygen. More extensive biochemical character-
izations, including K
m
determinations, are needed to better
de®ne the substrate speci®city of Ch-Prx1.
Peroxiredoxin and thioredoxin speci®city
There is i ncreasing evidence for a role o f Trx in coping with
oxidative stress. A m utant strain o f yeast Sacc haromyces
cerevisiae in which both Trx genes were disrupted has been
found to be particularly sensitive to hydrogen peroxide [8]

and to heavy-metals [32]. Heterologous complementation of
this yeast mutant with Arabidopsis type h Trx3 or type m
Trx1, 2, or 4, or type x Trx conferred hydrogen peroxide
tolerance [8,33]. The fact that the thioredoxin-dependent
Ch-Prx1 of Ch lamydomonas is able to detoxify hydrop er-
oxides provides additional s upport f or a f unction of Trx in
response to oxidative s tress.
All types of plant T rx, whether cytosolic or chloroplas tic,
were ab le to serve a s hydrogen donors for Ch-Prx1 in vitro.
Interestingly, the spinach f-type chloroplastic isoform was
more ef®cient with Ch-Prx1 than the m-type chloroplastic
Trx suggesting t hat T rx f is t he preferred electron donor to
Prx in vivo. This result differs from the results of yeast
complementation experiments in which several m-type
Arabidopsis Trx proteins have been shown to confer
hydrogen peroxide tolerance, while complementation with
f-type Trx d id not [33]. H owever, it cannot be e xcluded that
the yeast cytosolic NTR is unable to r educe Trx f.
The lack o f s peci®city of T rx toward Trx-dependent
proteins can explain why a presumably chloroplastic P rx
could be isolated with c ytosolic Trx h as bait even though
Southern blot analysis and genome sequencing indicate that
a cytosolic Prx exists in Chlamydomonas. It is also possible
that the putative cytosolic Prx was present i n our protein
extracts at much lower concentration than the chloroplast
Prx. It can b e noted that a s imilar loss of speci®city w as
recently reported b y Motohashi et al. [ 34] who trapped
targets o f t hioredoxin f on an af®nity column comprised of
thioredoxin m.
Regulation of peroxiredoxin gene expression

and defence against oxidative stress
Light is an important environmental factor inducing,
directly and indirectly, the production of ROS. ROS is
known to impair photosynthesis by damaging chloroplast
structures such as the D1 protein, LHCII, the chloroplast
ATPase an d r ibulose 1 ,5-bisphosphate carboxylase/oxygen-
ase (RubisCO) [35]. In Arabidopsis, peroxiredoxins have
been found to protect chloroplast structures from damage
by ROS [35]. Regulation of Ch-Prx1 gene expression may
be controlled either d irectly by ROS, e.g. by H
2
O
2
,whichis
known for its role in signal transduction [36], or indirectly by
sensors of redox conditions i n the chloroplast, e.g. ascorbate
[37]. We f ound that transcript levels of the Ch-Prx1 gene
markedly increased in illuminated cells (Fig. 8A,B) but a lso
upon bubb ling cultures w ith 100% oxygen in t he dark
(Fig. 8 C), conditions in which p roduction of ROS is likely
to be high. Blocking noncyclic photosynthetic e lectron ¯ow
with DCMU or DBMI B
10
inhibited the accu mulation of
Ch-Prx1 transcripts in the light (Fig. 8B), suggesting an
in¯uence o f the photosynthetic electron ¯ow on Ch-Prx1
gene expression. The redox state of plastoquinone, known
to regulate the expression of s ome genes [38,39], does not
seem to be the r esponsible for this r egulation, because both
DCMU and DBMIB exert an inhibitory effect. The

regulation of Ch lamydomonas Trx m gene expression was
also shown t o b e d ependent on the photosynthetic electron
¯ow [28] but independent of the redox state of the
plastoquinone pool. The expressions of other chloroplastic
genes, such as those encoding phosphoribulokinas e and
ferredoxin-NADP-reductase, follow the same pattern [40].
These results are in favor of a coordinated regulation o f
both Trx and Prx in the c hloroplast a nd fully support the
hypothesis that Ch-Prx1 is a c hloroplastic protein.
Because Ch-Prx1 gene expression correlates with pro-
duction of ROS, and because C h-Prx1 is able to remove
peroxides, it is likely that Ch-Prx1 is involved in detoxi®-
cation of ROS in the Chlamydomonas chloroplast. It should
be of interest to over- and under-express Ch-Prx1 in
Chlamydomonas to better understand its importance in
protection against oxidative stress.
280 A. Goyer et al. (Eur. J. Biochem. 269) Ó FEBS 2002
ACKNOWLEDGEMENTS
We would l ike to thank Prof. Peter Schu
È
rmann for a generous gift of
thioredoxins m a nd f f rom s pinach and Dr E. Issakidis-Bo urguet for
stimulating d iscussions.
REFERENCES
1. Tartaglia, L.A., Storz, G., Brodsky, M.H., Lai, A. & Ames, B.N.
(1990) Alkyl hydroperoxide reductase from Salmonella typhimu-
rium. J. Bio l. Chem. 265, 10535±10540.
2. Chae, H.Z., Kim, I H., K im, K. & Rhee, S.G. (1993) Cloning,
sequencing, and mu tation of thiol-speci®c antioxidant ge ne of
Saccharomyces c erevisiae. J. Bio l. Chem. 26 8 , 16815±16821.

3. Chae, H.Z., Robison, K., Poole, L.B., Church, G., Storz, G. &
Rhee, S.G. (1994) Cloning and sequencing of thiol-speci®c
antioxidant from m ammalian brain: alkyl reductases and thiol-
speci®c antioxidant de®ne a large family o f antioxidant enzymes.
Proc. N atl Acad. Sci. USA 91, 7 022±7026.
4. Kim, K., Kim, I.H., Lee, K.Y., Rhee, S.G. & Stadtman, E.R.
(1988) The isolation and puri®cation of a speci®c ÔprotectorÕ
protein which inhibits enzyme inactivation b y thiol/Fe (III) /O
2
mixed function oxidation system. J. Biol. Chem. 263, 4 704±
4711.
5. Baier, M. & D ietz, K J. (1996) Pr imary structure and e xpression
of plant h omologues of animal and fu ngal thio redoxin -dependent
peroxide reductases and bacterial alkyl hydroperoxide reductases.
Plant Mol . Biol. 31 , 553±564.
6. Baier, M. & Dietz, K J. (1997) The plant 2-Cys peroxiredoxin
BAS1 is a nuclear-encoded chloroplast protein: its expressional
regulation, phylogenetic origin, and implications for its speci®c
physiological function in plants. Plant J. 12 , 179±190.
7. Cheong, N.E., Choi, Y.O., Lee, K.O., Kim, W.Y., Jung, B.G.,
Chi, Y.H ., Jeong, J.S., Kim, K., Cho, M.J. & L ee, S.Y. (1999)
Molecular cloning, expression, and functional characterization of
a 2Cys-peroxiredoxin in Chinese cabbage. Plant Mol. Biol. 40,
825±834.
8. Verdoucq, L., Vignols, F., Jacquot, J.P., Chartier, Y. & Meyer, Y.
(1999) In vivo characterization of a thioredoxin h target protein
de®nes a new peroxiredoxin family. J. Biol. Chem. 274, 19714±
19722.
9. Jeong, J.S., Kwon, S.J., Kang, S.W., Rhee, S.G. & Kim, K. (1999)
Puri®cation and characte rization of a second type t hioredoxin

peroxidase (type II TPx) from Saccharomyces cerevisiae. Bio-
chemistry 38, 7 76±783.
10. Chae, H.Z., Chung, S.J. & Rhee, S.G. (1994b) Thioredoxin-
dependent peroxide reductase from yeast. J. Biol. Che m. 269 ,
27670±27678.
11. Sueoka, N. (1960) Mitotic replication of deoxyribonucleic acid
in Chlam ydomonas reinhardii. Pr oc. N atl Acad. Sci. USA 46, 83±
91.
12. Goyer, A ., Decottignies , P., Lemaire, S ., Ruell and, E ., Is sakidis-
Bourguet, E., Jacquot, J.P. & Miginiac-Maslow, M. (1999) The
internal Cys-207 of sorghum leaf NADP-malate d ehydrogenase
can form mixed disul®des with thioredoxin. FEBS Lett. 444, 165±
169.
13. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) In Molecular
Cloning: a Laboratory Manual. Cold S pring H arbo r Labo ratory
Press, Cold Spring Harbor, NY.
14. Merchant, S. & Bogorad, L . ( 1986) Re gulation by copper of the
expression of plastocyanin and cytochrome c 552 in Chlamydo-
monas r einhardtii. Mol. Cell. Biol. 6, 462±469.
15. Salvador, M .L., Kle in, U. & B ogorad, L. (1993) Light-regulated
and endogenous ¯uctuations of c hlo roplast transcript levels in
Chlamydomonas. Regulation by transcription and RNA degra-
dation. Plant J. 3, 213±219.
16. Rochaix, J.D., May®eld, S., Goldschmidt-Clermont , M. &
Erickson, J. (1988) In Plant Molecular Bi ology: a Practical
Approach. ( Shaw, C.H., ed.) pp. 253±275. I RL Press, Oxford.
17. Church, G. & Gilbert, W. (1984) Genomic sequencing. Pr oc. Natl
Acad.Sci.USA81, 1991±1995.
18. Stacy, J.E., Ims, R.A., S tenseth, N.C. & Jakobsen, K.S. (1991)
Fingerprinting of diverse species with DNA probes generated

from immobilized s ingle-stranded DNA templates. Nu cleic Acids
Res. 19, 400 4.
19. Lim, Y.S., Cha, M.K., Kim, T.B., P ark, J.W., Kim, K. & Kim,
I.H. (1993) R emoval of hy drogen peroxide a nd h ydroxyl r adical
by thiol-speci®c antioxidant protein as a possible r ole in vivo.
Biochem. Biophys. Res. Com m. 192, 273±280.
20.Lim,Y.S.,Cha,M.K.,Yun,C.H.,Kim,H.K.,Kim,K.&
Kim, I.H. (1994) Puri®cation and characterization of thiol-speci®c
antioxidant pro tein from human red blood cell: a new type
of antioxidant protein. Biochem. Biophys. Res. Comm. 199, 199±
206.
21. Jacquot, J P., Rivera-Madrid, R., Marinho, P., Le Mare
Â
chal, P.,
Miginiac-Maslow, M. & Meyer, Y. (1994) Arabidopsis thaliana
NADPH thioredoxin reductase: cDNA characterization and
expression of the r e combinant protein in Escherichia c oli. J. Mol.
Biol. 235, 1357 ±1363.
22. Stein, M ., Jacquot, J P., Jeannette, E., Decottignies, P ., Hodges,
M.,Lancelin,J M.,Mittard,V.,Schmitter,J M.&Miginiac-
Maslow, M. (1995) Chlamydomonas reinhardtii thioredoxins:
structure o f the genes coding for the chloroplastic m and cyt osolic
h isoforms: expression in Escherichia coli of the r ecomb inant
proteins, puri®cation and biochemical properties. Plant Mol. Biol.
28, 487± 503.
23. Kishigami, S ., Kanaya, E., Kikuchi, M. & Ito, K. (1995) DsbA±
DsbB interaction thro ugh their a ctive site c ysteines. J. Biol. Chem.
270, 170 72±17074.
24. Hodges, M., M iginiac-Maslow, M., Decottignies, P., J acquot ,
J P., Stei n, M., L epiniec, L., C retin, C. & Gadal, P. ( 1994) Puri-

®cation and characterization of pea thioredoxin f e xpressed in
Escherichia c oli. Plant Mol. Biol. 26 , 225±234.
25. Geck, M. & H artmann, F.C. (2000) Kinetic and mutational
analyses of th e regulation o f p hosphoribulo kinase by thioredoxins.
J. Biol. Chem. 275, 18034±18039.
26. Rivera-Madrid, R., M estres, D., Marinho, P., Jacquot, J P.,
Decottignies, P., Miginiac-Maslow, M. & Me yer, Y. (199 5) Evi-
dence for ®ve divergent thioredoxin h s equences in Ar abidops is
thaliana. Proc. Natl. Acad. S ci. (USA) 92 , 5620±5624.
27. Emanuelsson, O., Nielsen, H. & von Heijne, G. (1999) ChloroP, a
neural network-based m ethod f or pred icting ch loroplast transit
peptides and their c leavage sites. Pro tein Sci. 8, 9 78±984.
28. Lemaire, S., Stein, M., Issakidis-Bourguet, E., Keryer, E.,
Benoit, V ., Pineau, B .G., E
Á
rard-Hirne, C ., Miginiac-Maslow, M.
& Jacquot, J .P. (1999) The c omplex r egulation o f f erredoxin/
thioredoxin-related genes by light and the circadian clock. Planta
209, 221 ±229.
29. Brandes, H.K., Larimer, F.W. & Hartman, F.C. (1996) The
molecular pathway fo r the regulation of ph osphoribulokin ase by
thioredoxin f. J.Biol. Chem. 271 , 3333±3335.
30. Wang, P F., Veine, D. M., Ahn, S.H. & Williams, C.H. Jr (1996)
A stable m ixed disul® de bet ween t hioredo xin re ductase a nd its
substrate, thiore doxin: preparation and characterization. Bio -
chemistry 35, 4812±4819.
31. Brandes, H.K., Larimer, F.W., Geck, M.K., Stringer, C.D.,
SchcË rmann, P. & Hartman, F.C. (1993) Direct identi®cation of the
primary nucleophile of thioredoxin f . J. Biol. Chem. 268, 18411±
18414.

32. Vido,K.,Spector,D.,Lagniel,G.,Lopez,S.,Toledano,M.&
Labarre, J . (2001) A proteome analysis of the c admium response
in Saccharomyces ce revisiae. J. Bio l. Chem. 276, 8469 ±8474.
Ó FEBS 2002 Peroxiredoxin from Chlamydomonas (Eur. J. Biochem. 269) 281
33. Issakidis-Bourguet, E., Mo uaheb, N., M eyer, Y. & Migin iac-
Maslow, M. (2001) Heterologous complementation of yeast
reveals a new putative function for chloroplast m-type thioredoxin.
Plant J. 25 , 127±135.
34. Motohashi, K., Kondoh, A., Stumpp, M.T. & Hisabori, T. (2001)
Comprehensive survey of proteins targeted by chloroplast
thioredoxin. Proc. N atl Acad. Sci. USA 98, 11224±11229.
35. Baier, M. & Dietz, K J. (1999) Protective function of chlor o plast
2-cysteine peroxiredoxin in photosynthesis. Evidence from trans-
genic Arabidopsis. Plant Physiol. 119, 1407±1414.
36. Delaunay, A., Isnard, A.D. & Toledano, M.B. (2000) H
2
O
2
sensing through oxidation of t he Yap1 transc ription factor.
EMBO J. 19 , 5157±5166.
37. Baier, M., Noctor, G., Foyer, C.H. & Die tz, K J. (2000) Antisense
supression of 2-cysteine pe roxiredoxin in Ar abidopsis speci®cally
enhances the activities and expression of enzymes associated with
ascorbate m etabolism but no t glutathione m etabolism. Plant
Physiol. 124 , 823±832.
38. Karpinski, S., Escobar, C., Karpinska, B., Creissen, G. & Mulli-
neaux, P.M. (1997) Photosynthetic electron transpo rt regulates the
expression of cytosolic ascorbate peroxidase ge nes in Arabidopsis
during excess l ight stress. Plant C el l. 9, 627±640.
39. Escoubas, J.M., Lomas, M., Laroche, J. & Falkowski, P.G. (1995)

Light intensity regulation of cab g ene transcription is s ignaled by
the redox state o f t he plastoqu inone p ool. Proc. N atl Acad. Sci.
USA 92, 10237±10241.
40. Lemaire, S. (1999) Etude structurale et fonctionnelle des
syste
Á
mes f erre
Â
doxine et thiore
Â
doxine de
Â
pendants chez Chlamy-
domonas Reinhardtii. PhD thesis. Universite
Á
Paris XI, Orsay,
France.
282 A. Goyer et al. (Eur. J. Biochem. 269) Ó FEBS 2002