ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 280, No. 1, July, pp. 112-121,199O

Purification, Characterization, and Complete Amino Acid Sequence of a Thioredoxin from a Green Alga, Chlamydomonas reinhardtii Paulette Decottignies,“,’ Jean-Marie Schmitter,? Jean-Pierre Sophie Dutka,? Alain Picaud,$ and Pierre Gadal* *Laboratoire TLaboratoire fonctionnelle

Jacquot,*

de Physiologie V&g&ale Mo&laire, U.A. 1128 CNRS, University Paris-&d, 91405 Orsay Cedex, France; de Biochimie, Ecole Polytechnique, 91128 Palaiseau Cedex, France; and $Laboratoire de Biochimie des membranes v&&ales, 91198 Gif-sur-Yvette Cedex, France

Received December 6,1989, and in revised form February

27,199O

Two thioredoxins (named Chl and Ch2 in reference to their elution pattern on an anion-exchange column) have been purified to homogeneity from the green alga, Chlamydomonas reinhardtii. In this paper, we describe the properties and the sequence of the most abundant form, Ch2. Its activity in various enzymatic assays has been compared with those of Escherichia coli and spinach thioredoxins. C. reinhardtii thioredoxin Ch2 can serve as a substrate for E. coli thioredoxin reductase, with a lower efficiency when compared to the homologous system. In the presence of dithiothreitol (DTT), the protein is able to catalyze the reduction of porcine insulin. Thioredoxin Ch2 is as efficient as its spinach counterpart in the DTT or light activation of corn NADP-malate dehydrogenase, but it only activates spinach fructose- 1,6-bisphosphatase at very high concentrations. The complete primary structure of the C. reinhardtii thioredoxin Ch2 was determined by automated Edman degradation of the intact protein and of peptides derived from trypsin, chymotrypsin, clostripain, and SV8 protease digestions. It consists of a polypeptide of 106 amino acids (MW 11,808) and contains the well-conserved active site sequence Trp-Cys-GlyPro-Cys. The sequence of the algal thioredoxin Ch2 has been compared to that of thioredoxins from other sources and has the greatest similarity (67%) with the thioredoxin from Anabaena 7 I 19. o ISSO Academic PFW, IUC.

Thioredoxin is a small, ubiquitous, heat-stable protein which functions via a reversible disulfide-dithiol reac1 To whom correspondence

should be addressed.

tion (for a review, see Ref. (1)). It was first reported to be involved in the reduction of methionine sulfoxide (2) co& this protein has been and sulfate (3). In Escherichia shown to function as a hydrogen donor for the synthesis of deoxyribonucleotides by ribonucleotide reductase (4). E. coli thioredoxin also serves as a subunit of the bacteriophage T7 DNA polymerase (5) and it is required for the assembly of filamentous phages (6). More generally, this protein acts as a disulfide reductase; for example, it has been reported to catalyze the reduction of insulin disulfides (7). Thioredoxins have been found in a large number of organisms and the primary structures of some of them have been determined (8-22). All the thioredoxins sequenced so far contain an active site composed of -TrpCys-Gly-Pro-Cys-, except for the second protein isolated from Corynebacterium nephridii, where the glycine residue is replaced by an alanine (19). In higher plants, several different thioredoxins have been found. Chloroplastic thioredoxins are involved in the light regulation of fructose-1,6-bisphosphatase, a carbon metabolism enzyme, and NADP-malate dehydrogenase (23). In this system, usually called the ferredoxin-thioredoxin system, thioredoxin is reduced by a ferredoxin-thioredoxin reductase and, in turn, can activate the target enzyme (24). Two types of thioredoxins have been described, according to the enzyme that is preferentially activated (25). The m-type thioredoxin activates NADP-malate dehydrogenase while the activation of fructose-1,6-bisphosphatase specifically requires the f-type protein. Both types have been purified from spinach leaves and sequenced (11,20). It has been shown that their primary structures display few sequence identities. Recently, two cytosolic thioredoxins

112 All

0003-9861/90 $3.00 Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.

THIOREDOXIN

FROM

(called h) have been purified from spinach leaves and characterized (26) but their amino acid sequence and function remain unknown. Little is known about the presence of thioredoxin in green algae and no sequence is available. Several thioredoxins (cytoplasmic and chloroplastic) were purified from Scenedesmus obliquus (27, 28), but their primary structure is still unknown. It has been shown that the ferredoxin-thioredoxin system is present in the green alga, Chlamydomonas reinhardtii (29), including the ferredoxin, the ferredoxin-thioredoxin reductase and three thioredoxins (two m-types and one f-type, based on their reaction specificity). We recently characterized and sequenced the ferredoxin from this organism (30). In this paper, we report the purification, the characterization, and the primary structure of an m-type thioredoxin from C. reinhardtii. MATERIALS

AND

METHODS

Materials E. coli NADPH thioredoxin reductase and thioredoxin were supplied by IMCO (Stockholm, Sweden). Iodo (l-“Clacetic acid (2 mCi/ mmol) was purchased from CEA (Saclay, France). TPCK*-treated trypsin, TLCK-treated chymotrypsin and clostripain were obtained from Sigma; Staphylococcus aureus V8 protease was from Miles Laboratory. CNBr, trifluoroacetic acid (TFA) and HPLC grade solvents were purchased from Merck. Reagents used for amino acid sequencing were supplied by Applied Biosystems Inc. (Foster City, CA). Corn NADP-malate dehydrogenase was purified as previously described in (31). Ferredoxin (32), ferredoxin-thioredoxin reductase (33), and thioredoxin n (34) from spinach leaves were purified as previously described. Spinach leaf FBPase and thioredoxin f were generous gifts from Professor P. Schiirmann (Neuchatel, Switzerland).

Chlamydononas

reinhardtii

113

fuged at 13,OOog for 30 min. Hence, the supernatant could be directly fractionated by addition of solid ammonium sulfate, and the fraction precipitating between 46 and 80% saturation was collected by centrifugation (l3,609g, 15 min). The precipitate was dissolved in a minimal volume of Tris-HCl buffer pH 7.9 containing 0.1 mM EDTA (TE buffer) and 150 mM NaCl. The fraction was then centrifuged for 16 h at 150,OOOg.The clear supernatant was concentrated using an Amicon cell equipped with a YMlO membrane, under a N2 pressure of 3 atm. The concentrated fraction (20 ml) was applied to a Sephadex G-50 column (5 X 80 cm) equilibrated in TE containing 150 mM NaCl. Fractions (5 ml) were eluted by gravity flow. The fractions showing activity in the DTT-dependent NADP-MDH assay were pooled, concentrated to 200 ml using a YMlO membrane, and dialyzed twice against 5 liters in the of a 10 mM sodium acetate buffer, pH 4.5. Proteins precipitating sample were pelleted by centrifugation for 30 min at 10,OOOg.Most of the thioredoxin activity (determined by the DTT-dependent NADPMDH test) was found in the pellet, which was dissolved in 200 ml TE and applied to a DEAE-Sephacel column (2.5 X 20 cm) equilibrated in TE. The column was washed with excess TE and eluted with a 2 X 250-ml gradient of NaCl (O-300 mM) in TE. The fractions showing thioredoxin activity were pooled, concentrated using a YMlO membrane to 20 ml, and reduced with 1 mM DTT for 5 min at room temperature. The reduced sample was applied to a Sephadex G-50 column (5 X 80 cm) equilibrated in TE supplemented with 200 mM NaCl. The active fractions were pooled, concentrated using a YMlO membrane, and stored at -80°C. Further purification by HPLC was achieved on a Mono Q HR 5/5 anion exchange column (Pharmacia). The column was equilibrated with TE and eluted with a 20-min gradient of O-200 mM NaCl in TE at a flow rate of 1 ml mini. Thioredoxin activity was eluted as two peaks (Chl and ChZ), which were concentrated and dialyzed against TE using a microconcentrator Cl0 (Amicon).

Electrophoresis Electrophoresis of native or SDS-denatured thioredoxins was performed on 15% polyacrylamide slab gels at pH 8.6 using the methods of Davis (36) and Laemmli (37), respectively.

Fluorescence Measurements Culture Conditions C. reinhardtii wild type cells (Dangeard, strain 137c, coming directly from Levine’s lab) were grown photoheterotrophically in the presence of acetate (35). After 4-5 days culture under constant illumination in lo-liter carboys, the algae were harvested and concentrated using an artificial kidney dialyzer and centrifugation (4OOOg, 10 min). The cells were resuspended in a minimal volume of 20 mM potassium phosphate buffer, pH 7.0, containing 0.1 mM EDTA and 5% glycerol and stored at -90°C. One-hundred twenty liters was reduced to a volume of ca. 500 ml.

Extraction

and Purification

All steps were carried out at 3°C unless otherwise indicated. The cells were thawed, causing their disruption, and immediately centri-

’ Abbreviations used: DTNB, 5,5’-dithiobis (2.nitrobenzoic acid); DTT, dithiothreitol; FBP, fructose 1,6-bisphosphate; FBPase, fructose-1,6-bisphosphatase; HPLC, high performance liquid chromatography; NADP-MDH, NADP-dependent malate dehydrogenase; PTH, phenylthiohydantoin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TFA, trifluoroacetic acid, TLCK, N-tosylL-lysine chloromethyl ketone; TPCK, tosyl-L-phenylalanine chloromethyl ketone.

Fluorescence emission spectra were determined using a Perkin-Elmer MPF 44B spectrofluorometer. Samples were excited at 280 nm, and the emission was recorded between 300 and 450 nm, at 25°C. Thioredoxin (0.2 mg/ml) was dissolved in 30 mM Tris-HCI buffer, pH 7.9, for 10 and reduced by incubating the protein in 4 mM dithiothreitol min at 25°C.

Activity

Measurements

Reduction by thioredorin reductuse (38). Oxidized thioredoxin was tested as a substrate of E. coli NADPH thioredoxin reductase in the presence of NADPH and DTNB. In a final volume of 1 ml, the reaction medium contained 100 mM phosphate buffer pH 7.1, 10 mM EDTA, 150 PM NADPH, 200 pM DTNB (dissolved in 95% ethanol), and variable amounts of thioredoxin. The reaction was started by adding 10 nM thioredoxin reductase and the reduction of DTNB determined by monitoring the absorbance change at 412 nm. The activity was expressed as micromoles of thioredoxin reduced per minute, using 13,600 M-i cm-’ as the molar absorption of DTNB (2 SH are formed per mole of thioredoxin reduced). DTT-dependent activation of NADP-malate dehydrogenase (39). The activation medium (60 ~1) contained 100 mM Tris-HCl buffer, pH 7.9, 5 mM DTT, 0.6 pM NADP-MDH, and thioredoxin as indicated. After variable incubation times at 20-C, a 10.~1 aliquot was used to determine the activity of NADP-malate dehydrogenase at 30°C in a medium containing 100 mM Tris-HCl buffer, pH 7.9, 0.2 mM NADP,

114

DECOTTIGNIES 1

‘- 5

ET AL. moved by five filtrations using a Centricon 10 microconcentrator (Amicon), and in the same step Tris-HCl buffer was replaced by 0.1 M NH* HCOB.

Enzymatic

-2

,’

I’

Digestions

[r4C]Carboxymethylthioredoxin Ch2 (140 pg, 12 nmol) was digested in 0.1 M NHIHCOB, pH 8.5, at 37°C with TPCK-treated trypsin (3 rg) for 2 h, or with TLCK-treated chymotrypsin (3 pg) for 5 h. Cleavage at Glu residues was achieved with SV8 protease (5 lg) for 7 h at 37°C in 0.15 M NHIHCO,, pH 8.5. Clostripain cleavage was performed by adding 3 pg of clostripain previously reduced with 2 mM DTT to 120 gg of [r4C]carboxymethylthioredoxin. The digest was carried out at 25°C in the presence of 1 mM CaCl*.

Cyanogen Bromide Cleavage

10

20

min

FIG. 1. High performance liquid chromatography of C. reinhardtii thioredoxins on a Mono Q HR 5/5 anion-exchange column. After purification steps performed as described under Material and Methods, 300 pg of protein was applied to the column and eluted at a flow rate of 1 ml min-’ , with a linear gradient from 100% A, 0% B buffer to 60% A, 40% B buffer, in 20 min, where A is 30 mM Tris-HCl, pH 7.9, 0.1 mM EDTA (TE) and B is 500 mM NaCl in TE.

Cleavage at methionyl residues was achieved by reacting [“‘Clearboxymethylthioredoxin (20 nmol) in 70% formic acid with a 250-fold molar excess of CNBr for 40 h at room temperature in the dark. The mixture was evaporat.ed and submitted to sequence analysis without further purification.

Peptide Separation The digests were dried reverse-phase HPLC on Monitor D), using a 0.46 a 0.21 X IO-cm Aquapore Absorbance was recorded

and 0.7 mM oxaloacetic acid. The amount of NADPH oxidized was measured by monitoring the decrease in absorbance at 340 nm. One unit of enzymatic activity represents 1 kmol NADPH oxidized per minute. The Light activation of corn NADP-malate dehydrogenase (31,33). activation medium (40 ~1) contained 100 mM Tris-HCl buffer, pH 7.9, washed thylakoids extracted from pea leaves (equivalent to 15 Gg chlorophyll), 10 +M spinach ferredoxin, 2 pM spinach ferredoxin-thioredoxin reductase, 0.3 yM NADP-MDH and thioredoxin as indicated. The activation was carried out at 25°C under N, and constant illumination (300 W/mr). After 9 min incubation, a 30-/Al aliquot was used to determine the activity of NADP-MDH as described after DTT activation. Activation of spinach FBPase (40). The activation medium (40 ~1) contained 150 mM Tris-HCl buffer, pH 7.9,8 mM DTT, 0.6 fiM FBPase (0.3 units) and thioredoxin as indicated. After a 30-min preincubation at 2O”C, a 35~1 aliquot was used to determine the activity of FBPase in a reaction medium containing 100 mM Tris-HCl buffer, pH 7.9, 0.1 mM EDTA, 1.5 mM MgS04, 2.5 mM FBP, 0.6 units of glucose 6 phosphate dehydrogenase, 2 units of glucose isomerase and 0.2 mM NADP. The activity was followed spectrophotometrically by monitoring the increase in absorbance at 340 nm. Reduction of ilzsulin (7). The reaction medium contained 100 mM phosphate buffer, pH 7.1, 0.13 mM porcine insulin, 2 mM EDTA, and variable amounts of thioredoxin in a final volume of 1 ml. The reaction, carried out at 3O”C, was started by adding 0.5 mM DTT, and the reduction was monitored by an absorbance change at 650 nm. A cuvette without thioredoxin served as a control.

1

+ Preparation

of [14C]Carboxymethylthioredoxin

Thioredoxin Ch2 (2 mg/ml) was reduced with 4 mM dithiothreitol for 1 h at 37°C under N, and carboxymethylated with 10 mM iodo[lr4C 1acetic acid for 15 min in the dark (41). Excess reagents were re-

under vacuum and the peptides purified by an LDC system (miniMetric pumps, spectra X 25-cm Ultrasphere ODS (Altex) column or Butyl column (Brownlee) for longerpeptides. at 215 nm. Peptides were eluted with a linear

2

3

-

FIG. 2. Native polyacrylamide gel electrophoresis of thioredoxins from E. coli, C. reinhardtii and spinach, at pH 8.6. Lane 1, E. coli thioredoxin, 2 Gg; lane 2, isomers of spinach leaf thioredoxins, 2 pg; lane 3, thioredoxin Ch2,l Gg.

THIOREDOXIN

FROM

Chlamydomonas

115

reinhardtii

RESULTS

Purification

and Properties

Purification.

C. reinhardtii thioredoxin was purified sulfate precipitation, anion-exchange chromatography, and two gel filtrations. The test routinely used during the purification process was the DTTdependent activation of NADP-MDH. After these steps, two major bands were observed in SDS-polyacrylamide gel electrophoresis, corresponding to two proteins Chl and Ch2 with apparent molecular masses of 10 and 10.5 kDa, respectively. These two proteins were further resolved by HPLC on an anion-exchange column (Fig. 1). Both exhibited thioredoxin activity. Chl, eluting first, represented about 30-35% of the total amount of thioredoxin and Ch2 represented 70-65%, based on their absorbance at 280 nm. At pH 7.9, the retention times of Chl and Ch2 correspond to 20 and 40 mM NaCl, respectively (Fig. l), while the E. coli counterpart eluted at 160 mM NaCl and the two isomer thioredoxins from spinach leaves, previously described by Schiirmann (40) at 120130 mM NaCl (data not shown). Minor contaminants were removed during the HPLC and after this step, Ch2 thioredoxin was homogeneous as determined by polyacrylamide gel electrophoresis in native (Fig. 2) and denaturating conditions (data not shown). About 3-4 mg of Ch2 thioredoxin and 1.5-2 mg of Chl were obtained in a pure form from 1 kg of wet algal cells. In this paper, we describe the properties and the sequence of the major thioredoxin, Ch2. Molecular mass. The molecular mass of Ch2 estimated by SDS-PAGE with marker proteins including spinach thioredoxin was 10,500 (data not shown). This value is slightly different from the one calculated from the amino acid sequence (11,808). It can be noted that thioredoxin Ch2 always migrated further than its spinach counterpart, although they have approximately the same molecular mass (11). Fluorescence emission spectra. Fluorescence emission spectra of oxidized and reduced thioredoxin Ch2 (Fig. 3A) and E. coli thioredoxin (Fig. 3B) were recorded after excitation at 280 nm. The spectrum of the oxidized algal thioredoxin Ch2 shows a unique maximum at 340 by

Wavelength

(nm)

FIG. 3. Fluorescence emission spectra of thioredoxins from A, C. reinhardtii (Ch2) and B, E. coli, in oxidized and reduced forms, at pH 7.9. The proteins were excited at 280 nm and the emission recorded between 300 and 450 nm. Reduction was achieved by incubation with 4mM DTT for 10 min.

gradient from 0 to 60% acetonitrile in 0.1% TFA. After collection of the major peaks, the solvent was evaporated and, when needed, the fraction was rechromat,ographed on the same system with a linear gradient from 0 to 90% methanol in 10 mM ammonium formate pH 7.5.

ammonium

Sequence Determination Automated Edman degradation was performed using an Applied Biosystems Model 470 A sequencer. Thirty-microliter aliquots (100 to 1000 pmol) of peptide and [r4C]carboxymethylthioredoxin samples were applied on the glass fiber disc of the cartridge, after precycling with polybrene (2.5 mg). Phenylthiohydantoins (PTHs) were analyzed in the on-line mode. The part returned to the fraction collector of the sequencer (60% of available PTH at each cycle) was used for liquid scintillation counting, in order to confirm the identification of PTH-carboxymethylcysteines (42). Sequence identities among the thioredoxins were calculated and alignments performed using the CLUSTAL program (43).

TABLE

I

Constants of E. coli NADPH-thioredoxin Reductase Using Thioredoxins from Various Sources as Substrates

Kinetic

Thioredoxin

E. colt' C. reinhardtii (Ch2) Spinach (m-type)

%, (/.tM) 2.6 6 28

kat (5’) 37 4 4

116

DECOTTIGNIES

ET AL.

A

A

/I’ 8,. .

r----7-

.’ i’

/’

.,’/I .’

,/ 6’

30

40

50

2,

FIG. 4. Insulin reduction. Rate of thioredoxin-catalyzed insulin reduction at different concentrations of thioredoxin from (A) E. coli, 0, without thioredoxin; 0, 2.5 pM thioredoxin; n , 5 pM thioredoxin; (ES) spinach (m-type), 0, without thioredoxin; 0, 2 @M thioredoxin; n , 5 PM thioredoxin, 0,7 FM thioredoxin; and (C) C. reinhardtii (Ch2), Cl, without thioredoxin; l ,2.5 gM thioredoxin; n , 5 +M thioredoxin.

nm, while the oxidized E. coli thioredoxin shows a second maximum at 310 nm, attributed to tyrosine fluorescence. When thioredoxin Ch2 was reduced, a small fluerescence increase of about 28% was observed. In the same conditions, a 2.5-fold increase was obtained for reduced E. coli thioredoxin (Fig. 3B), as previously described by Holmgren (44).

Enzyme Activity C. reinReaction with E. coli thioredoxin reductase. ha&ii thioredoxin Ch2 was tested as a substrate for E. coli NADPH thioredoxin reductase with DTNB used as an electron acceptor. The algal protein was reduced by the bacterial reductase as shown in Table I. However, the apparent K,,, value was 2-fold higher and the velocity

THIOREDOXlN

.E’

0.1

FROM

-

z

a

;

tpp”,,l 2

4

6

Thioredoxin

6

concent

10

12

JIM

ration

FIG. 5. DTT-dependent activation of corn NADP-MDH. Effect of the concentration of different thioredoxins on the rates of appearance of activated NADP-MDH. The amount of NADP-MDH in the 10.~1 aliquot used for the assay was 0.5 pg. Rates of appearance of activated enzyme (expressed in units of enzymatic activity per minute) were determined as follows: NADP-MDH activity (expressed in units) was measured at various incubation times for several thioredoxin concentrations then the slope was plotted versus thioredoxin concentration. Thioredoxins were from 0, C. rein&&ii (Ch2); V, E. coli; l , spinach (m-type).

decreased by lo-fold when compared with the homologous bacterial system. Spinach thioredoxin m exhibited the same behavior with a 4-fold higher K, value and the same Kcatas its algal counterpart. Reduction of insulin. Holmgren has shown that E. coli thioredoxin is able to catalyze the reduction of insulin disulfides by DTT (7). The rate of insulin reduction is routinely measured spectrophotometrically at 650 nm as an increase in turbidity due to the precipitation of the free insulin chain B. We have compared the rates of porcine insulin reduction by DTT at pH 7.0, in the presence of different thioredoxins (Ch2, E. coli and spinach mtype) at various concentrations. The three proteins efficiently catalyze the reaction as shown in Fig. 4. The highest rate was obtained with the protein from E. coli: 0.1 AAGsOmin-’ at 5 PM (corresponding to a specific activity of 1.7 AAssO min-‘mg--I). This value is in good agreement with those obtained by Holmgren in the reduction of bovine insulin (7). Both Ch2 and spinach mtype thioredoxins catalyzed the reaction at slightly lower rates of 0.06 AAs5,, mini* and 0.05 AAssO min-’ at 5 PM, respectively. Activation of chloroplastic enzymes. The effect of C. reinhardtii thioredoxin Ch2 on the activation of two chloroplastic enzymes, NADP-malate dehydrogenase and fructose-1,6-bisphosphatase, has been investigated. The activation rates of corn NADP-MDH in the pres-

Chlamydomonas

117

reinhardtii

ence of three different thioredoxins at various concentrations were compared (Fig. 5). Thioredoxin Ch2 was efficient in the DTT-dependent activation of purified NADP-MDH in the concentration range tested (0.35 to 14 ~.LM). At concentrations higher than 2 FM, the proteins from E. coli and spinach activated the enzyme at a slightly higher rate than the algal thioredoxin. In addition, thioredoxin Ch2 was as efficient as its spinach mtype counterpart in the light activation of NADP-MDH, measured in the in vitro-reconstituted system consisting of thylakoids, ferredoxin, ferredoxin-thioredoxin reductase, thioredoxin, and NADP-MDH (Fig. 6). In all cases, NADP-MDH activity decreased when high concentrations of thioredoxin were used, especially with the E. coli protein which was efficient only at low concentrations (