Ferredoxin-thioredoxin reductase: A chromophore-free protein of chloroplasts

FEBS LETTERS

Volume 107. number 1

November 1979

FERREDOXIN-THIOREDOXIN REDUCTASE: A CHROMOPHORE-FREE PROTEIN OF CHLOROPLASTS Angel DE LA TORRE, Catalina LARA, Ricardo A. WOLOSIUK*

and Bob B. BUCHANAN

Department of Cell Physiology, Universityof California,Berkeley, CA 94720, USA Received 12 August 1979

1. Introduction The ferredoxin/thioredoxin system constitutes a mechanism whereby light regulates the activity of key enzymes of chloroplasts [ 1,2]. In this system, photochemically reduced ferredoxin serves as the reductant for thioredoxin - proteins which, when reduced, modulate the activity of enzymes of photosynthesis and related chloroplast processes. The ferredoxinlinked reduction of thioredoxin does not occur spontaneously but requires the enzyme ferredoxinthioredoxin reductase (eq. (1)): 2 Ferredoxintiuced

t

ferredoxin-thioredoxin reductase

Thioredoxinotiiz~ 2’Ferredoxm

+ 2 H’ 2. Materials and methods

oxidizedt (1)

Of the components of the ferredoxin/thioredoxin system, ferredoxin-thioredoxin reductase stands alone in that it has not been characterized. The reductase is of particular interest because its associated reaction (the reduction of a thioredoxin by an iron-sulfur protein) represents a new type of enzyme reaction that could occur not only in chloroplasts, as is * Present address: Fundacion Campomar, Buenos Aires, Argentina Address correspondence to: Professor Bob B. Buchanan, Department of Cell Physiology, 3 13 Hllgard Hall, University of California, Berkeley CA 94720, USA ElsevierlNorth-HollandBiomedical JVess

currently recognized, but in other ferredoxincontaining systems as well. We have therefore focused our attention on ferredoxin-thioredoxin reductase, and we now describe certain properties of a highly purified preparation obtained from spinach leaves. The enzyme (purified by measurement of its ability to promote the photochemical activation of chloroplast fructose 1,6-bisphosphatase (Fru-Passe)) seems to consist of a single subunit (M, 38 000) and to lack a chromophore group. The purification from spinach leaves of a 38 000 mol. wt protein showing ferredoxin-thioredoxin reductase activity has been reported [3].

Ferredoxin-thioredoxin reductase was purified from market spinach leaves by the procedure in [4] that was modified to include, among other changes, an additional adsorption and elution from DEAEcellulose. After the described acetone precipitation and dialysis (step III), the protein solution was applied to a DEAE-cellulose column (5 X 12 cm) that was equilibrated beforehand with a solution containing 50 mM Tris-HCI buffer (pH 7.9) 0.1% 2-mercaptoethanol and 0.1 M NaCI. Adsorbed protein was then eluted sequentially with the same buffer containing 0.1,0.3, and 0.5 M NaCl. The fraction eluted with 0.3 M NaCl was collected and used for the Sephadex G-100 step, as in [4]. An additional modification in the procedure [4] included a change in the salt gradient used for DEAEcellulose chromatography (step IV) and the use of a 141

FEBS LETTERS

Volume 107, number 1

second Sephadex G-100 column. The buffer used for DEAE-cellulose chromatography was 2 1 of a linear gradient between 0.12 and 0.30 M NaCl, 30 mM in Tris-HCl (pH 7.9) (hereafter called buffer A), containing 0.1% 2-mercaptoethanol. The new Sephadex G-100 chromatography step (introduced before the hydroxyapatite chromatography (step VII) was carried out with a column of 2.5 X 100 cm that was equilibrated and eluted with buffer A plus 0.1% 2-mercaptoethanol). Published procedures were used for the isolation from spinach of ferredoxin [S], thioredoxins f and m [6], chloroplasts FruPzase [4],thioredoxinscfandc, [7], chloroplast NADP-malate dehydrogenase [8], and twice-washed chloroplast membrane fragments [9]. Protein was estimated by a modified phenol procedure [lo]. Ferredoxin-thioredoxin reductase was routinely assayed by the ferredoxin/FruPsase calorimetric procedure [4]. Electrophoresis in native and in sodium dodecyl sulfate (SDS)-polyacrylamide gels was according to [ 1l] and [ 121, respectively. For metal analysis, samples were either dialyzed against buffer A containing 0.1% 2-mercaptoethanol and 3 mM EDTA or were passed through a 0.5 X 3-cm Cnelex 100 column (BioRad Labs, Richmond, CA)

November 1979

that was equilibrated with buffer A containing 0.1% 2-mercaptoethanol. Fe, Cu,Mn, and Zn were analyzed by atomic absorption spectrophotometry after 3 h digestion (110°C) in a nitric acid/perchloric acid mixture (85: 15). MOwas estimated calorimetrically according to [ 131. Chemicals were obtained from commercial sources and were of the highest quality available. Reference proteins for molecular weight estimation were obtained from Sigma Chemical Co. (St. Louis, MO).

3. Results and discussion Ferredoxin-thioredoxin reductase was followed throughout the purification procedure by measuring its capacity to promote the light-dependent activation of chloroplast Fru-Psase [ 1,4]. Light-dependent FruPzase activation by the purified reductase showed the expected requirements for soluble ferredoxin and thioredoxin f - the chloroplast thioredoxin specific for FruPsase and other chloroplast enzymes [4] (table 1). When tested in SDS-polyacrylamide gel electrophoresis, purified ferredoxin-thioredoxin reductase

Table 1 Requirements for lightdependent activation of chloroplast Fru-P,ase with purified components of ferredoxin/thioredoxin system Pi released (nmol/min) Light, Light, Light, Light, Light, Dark,

complete minus ferredoxin minus thioredoxinf minus ferredoxin-thioredoxin minus Fru-P,ase complete

reductase

153a 7 7 6 3 7

a Complete, light, under air showed 7 nmol Pi released per min The reaction was carried out at 20°C in Warburg vessels containing in the sidearm, 6 pmol fructose 1,6_bisphosphate and in the main compartment (in 1.5 ml final vol.) 28 pg FruP,ase, 120 pg ferredoxin, 40 pg thioredoxinf, 112 clg ferredoxinthioredoxin reductase, twice-washed chloroplast membrane fragments equivalent to 0.1 mg chlorophyll, and the following bmol): Tris-HCl buffer (pH 7.9), 100; MgSO,, 1; sodium ascorbate, 10; dichlorophenol indophenol, 0.1. Vessels were equilibrated with nitrogen for 6 min and then were incubated for 5 mm in saturating light (20 000 lux). The reaction was started by the addition of fructose 1,6_bisphosphate from the sidearm and continued for 30 min under ilhrmination. The reaction was stopped with 0.5 ml 10% trichloroacetic acid. After the precipitate was removed by centrifugation, Pi was estimated by a modified FiskeSubbaRow procedure [9]

142

Volume 107, number 1

I

FEBS LETTERS

>

Fig. 1. Densitometric trace of ferredoxin-thioredoxin reductase in SDS-polyacrylamide gel electrophoresis. Electrophoresis was performed with 20 pg protein in 0.5 X 7 cm gels of 10% acrylamide and 0.1% SDS. Gels were stained with Coomassie brilliant blue and, following destaining in 7.5% acetic acid + 5% methanol, were scanned at 560 nm in a Gilford M220 densitometer (full scale of 2.5).

November 1979

amide) and in Sephadex G-100 chromatography on calibrated columns (data not shown). The data collectively suggest that the ferredoxin-thioredoxin reductase from spinach chloroplasts contains a single 38 000 mol. wt subunit. This enzyme was also reported to have mol. wt 38 000 [3]. At 95% purity, chloroplast ferredoxin-thioredoxin reductase showed no evidence of a chromophore. The enzyme had a single absorption peak in the ultraviolet region (277 nm) and no significant absorption in the visible region (fig. 3). At 1 mglrnl, ferredoxin-thioredoxin reductase was colorless. When tested for metals, the purified enzyme contained negligibb amounts of Mn, Zn, Cu and MO, and a trace amount of Fe (5 0.1 g atom Fe/mol enzyme). Ferredoxinthioredoxin reductase thus appears to be devoid of these metals as well as of an enzyme-bound chromophore . The ability of purified ferredoxin-thioredoxin reductase to reduce the thioredoxinfneeded for activation of FruPzase raises the question of whether

.9-

11 277nm

.7

appeared to be -95% pure (fig. 1) and showed a minimum mol. wt 38 000 (fig. 2). A similar molecular weight was observed in native gel electrophoresis (determined with 7.5%, 9%, 1 l%, 13% and 15% acryl-

.6

g

.4 .3 x

.= .I

o-

I

200

I 300

I

400

I

500

I 600

I

700

Wavelength, nm

Fig. 2. Relative mobility of ferredoxin-thioredoxin reductase in SDS-polyacrylamide gel electrophoresis. Conditions as givenin frg. 1.

Fig. 3. Absorption spectrum of ferredoxin-thioredoxin reductase. The spectrum was obtained with 700 pg enzyme in 1 .O ml buffer A containing 0.1% 2-mercaptoethanol. Measurement was made in a Cary 14 M spectrophotometer with a cuvette of 1 cm lightpath.

143

Volume 107, number 1

FEBS LETTERS

November 1979

Table 2 Requirements for lightdependent activation of chloroplast NADP-malate dehydrogenase with purified components of the ferredoxinl thioredoxin system NADPH oxidized (nmol/min) Light, Light, Light, Light, Light, Dark,

complete minus ferredoxm minus thioredoxin m minus ferredoxin-thioredoxin reductase minus NADP-malate dehydrogenase complete

25 0 3 0 0 5

The reaction was carried out in Warburg-Krippahl vessels containing in the sidearm, 40 pg partially purified NADP-malate dehydrogenase and in the main compartment (in 0.4 ml final vol.) 72 fig ferredoxin, 90 fig thioredoxin m, 38 c(g ferredoxin-thioredoxin reductase, twice-washed chloroplast membrane fragments equivalent to 20 fig chlorophyll, and the following (Nmol): Tris-HCl buffer (pH 7.9), 20; sodium ascorbate, 2; dichlorophenol indophenol, 0.02; 2-mercaptoethanol, 6. Vessels were equilibrated for 5 mm with nitrogen, NADP-malate dehydrogenase was added from the sidearm, and the enzyme was activated by 10 min illumination (20 000 lux). Then, 0.2 ml of the mixture was removed with a microsyringe and injected into a cuvette (1 ml capacity, 1 cm lightpath) containing, in 0.8 ml, the following (pmol): Tris-HCl buffer (pH 7.9), 100; NADPH, 0.25 ; 2-mercaptoethanol, 11. Oxalacetic acid (2.5 rmol) was added immediately and the LL43e,,was followed for 5 min in a Cary 14 M recording spectrophotometer

the enzyme also catalyzes the reduction of thioredoxin m - the type of thioredoxin that activates NADP-malate dehydrogenase in chloroplasts [4]. The data in table 2 provide an affirmative answer to this question: ferredoxin-thioredoxin reductase catalyzed a light-dependent activation of chloroplast NADPmalate dehydrogenase that was dependent on thioredoxinm as well as on soluble ferredoxin. Ferredoxinthioredoxin reductase seems to have the capability of reducing each of the two types of chloroplast thioredoxins - - thioredoxmfand thioredoxin m [4]. A similar conclusion can be reached from [ 141. It should be mentioned that, in addition to thioredoxin m, purified ferredoxin-thioredoxin reductase catalyzed the reduction of cytoplasmic thioredoxin cm [6] (based on capacity for light-dependent NADF malate dehydrogenase activation, as given in table 2) and of cytoplasmic thioredoxin cf [6] (based on capacity for light-dependent FruPzase activation, as given in table 1). However, in contrast to chemical reduction with dithiothreitol [6], the reduction of the cytoplasmic thioredoxins by ferredoxin-thioredoxin reductase was sluggish: in the photochemical 144

systems, the activity of thioredoxin cm was 24% that of thioredoxin m, and the activity of thioredoxin cf was 17% that of thioredoxinf(data not shown). Thioredoxin from Escherichiu coli [l], unlike that from rabbit liver [ 151, was shown earlier to undergo light-dependent reduction by ferredoxin-thioredoxin reductase.

4. Concluding remarks The present findings provide evidence that ferredoxin-thioredoxin reductase, the enzyme shown to catalyze the ferredoxin-linked reduction of thioredoxin in chloroplasts, is devoid of a metal or other type of protein chromophore. The results are therefore consistent with the idea that the transfer of electrons from ferredoxin to thioredoxin occurs via a proteinaceous group on this enzyme, e.g., through the reversible reduction/oxidation of a cyst(e)ine residue. It is significant that a single ferredoxin-thioredoxin reductase enzyme catalyzes the ferredoxin-linked reduction of each of the two types of chloroplast

Volume 107, number 1 thioredoxins, thioredoxins f and m . The mechanism for the ferredox~-liked

FEBSLETTERS

November 1979

References reduction

thus differs from the NIP-led mechanism that was described in the classical work [ 161. Those studies demonstrated that the NADPHdependent reduction of thioredoxin is catalyzed by an enzyme that contains FAD as the prosthetic group ASP-thioredox~ reductase). The NAP-liked reduction of thioredoxin may apply to aerobic bacteria, yeast, and animal tissues. Recently, NADP-thioredoxin reductases were also isolated from wheat seeds [ 171 and a photosynthetic bacterium [ 181. There is, however, no evidence for such an enzyme in c~oroplasts or in their parent leaves. It would appear that the reduction of thioredoxin in chloroplasts is achieved independently of NADP via ferredoxin and the ferredoxin-thioredoxin reductase enzyme described above. The mechanism by which cytoplasmic thioredoxins are reduced in leaf tissue remains an open question. of thioredoxin

Acknowledgements

This work was aided by grant ‘78-l5287 from the National Science Foundation, C. L. gratefully acknowledges support of a fellowship from The Spanish Ministry of Education and Science.

111Wolosiuk, R. A. and Buchanan, B. B. (1977) Nature 266,565-567.

PI Buchanan, B. B., Wolosiuk, R. A. and Schtirmann, P. (1979) Trends Biochem. Sci. 4,93-96. [31 Schtlrmann, P. (1979) Experientia in press. [41 Wolosiuk, R. A., Crawford, N. A., Yee, B. C. and Bu~a~n,B.B. (1979) J. Biol. Chem. 254,1627-1632. [51 Buchanan, B. B. and Arnon, D. I. (19’71) Methods Enzymol. 23,413-440. [61 Crawford, N. A., Yee, B. C., Nishizawa, A. N. and Buchanan, B. B. (1979) FEBS Lett. 104,141-146. [71 Buchanan, B. B., Schiirmann, P. and Wolosiuk, R. A. (1976) Biochem. Biophys. Res. Commun. 69,970-978. PI Wolosiuk, R. A., Buchanan, B. B. and Crawford, N. A. (1977) FEBS Lett. 81,253-258. PI Buchanan,B. B., Kalberer,P. P. and Arnon, D. I. (1967) Biochem. Biophys. Res. Commun. 29,74-79. [lo] Lovenberg, W., Buchanan, B. B. and Rabinowitz, J. C. (1963) J. Biol. Chem. 238,3899-3913. [ 111 Hedrick, J. L. and Smith, A. J. (1968) Arch. Biochem. Biophys. 126, X5-164. 112) Weber, K. and &born, M. (1961) J. Biol. Chem. 244, 4406-4412. [ 13) Cirdenas, J. and Mortenson, L. E. (1974) Anal. Biothem. 60,372-381. [ 141 SchUrmann,P. and Jacquot, J. P. (1979) Biochim. Biophys. Acta in press. [ 15 ] Holmgren, A., Buchanan, B. B. and Woloshrk, R. A. (1972) FEBS Lett. 82,351-354. [ 161 Reichard,P. (1968) Eur. J. Biochem. 3,259-266. [17] Suske, G., Wagner, W. and Follmann, H. (1979) Z. f. Naturforsch. 34c, 214-221. [18] Clement, J. D. (1979) FEBS Lett. 101, X16-120.

145