plant seeds contain several thioredoxins of regular size

Fur J Biochem 131, 139-344 (1983) FbBS I983

Plant Seeds Contain Several Thioredoxins of Regular Size

Andrea BERSTERMANN, Klaus VOGT, and Hartmut FOLLMANN

Fachbereich Chemie (Biochemie) der Philipps-Universitit, Marburg

(Received August 27/ December 6, 1982) - EJB 5944

Thioredoxin systems composed of several thioredoxin isoproteins and a NADPH : thioredoxin reductase are contained in the albumin-globulin fraction of wheat and soy-bean seed proteins. Two wheat thioredoxins I and I1 were separated on CM-cellulose whereas soy-bean extracts could be resolved into three thioredoxins I, 11, and I11 on DEAE-cellulose. These proteins were purified to apparent homogeneity and were shown by sodium dodecylsulfate/polyacrylamide gel electrophoresis to possess the molecular weight M, z 12000 typical of the single bacterial and animal thioredoxin. In contrast, gel filtration runs may yield erroneous estimates of thioredoxin molecular weights. The seed thioredoxins can serve as ribonucleotide reductase (Escherichia coli) substrates. They stimulate spinach NADP: malate dehydrogenase but are inactive towards chloroplast fructose-bisphosphatase. These results demonstrate that the number of thioredoxins in nongreen plant tissues approaches that of leaves; additional explanations must therefore be sought for the multiple thioredoxin profiles of plants besides diversification for light-dependent and light-independent functions.

Plants are known to contain several protein fractions with thioredoxin activity [l - 51. It is an open question whether this is so because in addition to established thioredoxin functions like hydrogen transfer for deoxyribonucleotide synthesis, sulfate, or protein disulfide reduction there are extra, specific thioredoxins that participate in light-affected chloroplast reactions, or whether one observes the not uncommon case of isoproteins having common origin and basically similar, although modulated reactivities. Previously reported properties of plant thioredoxins appeared to support the first view, and the proteins (especially those isolated from leaves) have been grouped into ones activating fructose-bisphosphatase (‘f’ type), others that preferentially stimulate NADP: malate dehydrogenase (‘m’ type), and ‘c‘ thioredoxins of presumed cytoplasmic origin [I]. Such typification bears some arbitrariness as other enzymes like ribonucleotide reductases [6, 71, glutamine synthetase [28] or NADP: glyceraldehyde-3-phosphate dehydrogenase [30] are also activated but are not usually measured for practical reasons. In fact, the more or less unlimited exchangeability in v i m of many bacterial. animal, and plant thioredoxins in completely unrelated enzyme systems [6, 71 suggests that all these small -SH proteins are closely related and not highly specialized. The long-known existence of two thioredoxin isoproteins in yeast [S] and the preliminary characterization of more than one thioredoxin in photosynthetic bacteria and cyanobacteria [9, 101 also indicate that thioredoxin multiplicity is not restricted to eukaryotic plant cells, nor to photosynthetic organisms at all.

Abbreviutions. PAdoPS, 3’-phosphoadenosine 5’-phosphosulfate: SDS, sodium dodecyl sulfate; 2’,5’-ADP-Sepharose, 2’,5’-ADP linked to Sepharose via a 6-aminohexyl group.

Enzymes. Malate dehydrogenase (NADP’ ) (EC 1.1.1.82); thioredoxin reductase (NADPH) (EC 1.6.4.5); thioredoxin reductase (ferredoxin) (EC 1.8.7.-); ribonucleoside-diphosphate reductase (EC I . 17.4.1); PAdoPS: thioredoxin sulfotransferase (EC 2.8.2. -); fructose-I .6- bisphosphatase (EC 3.1.3.1 I ) .

More information about the precise thioredoxin pattern of plant tissues, as well as pure plant thioredoxin samplcs for protein characterization are thus desirable. Unfortunately it is not easy to unambiguously establish the ‘true’ number of thioredoxins in a plant extract because these redox proteins, lacking enzyme activity themselves, require enzyme-catalyzed indicator reactions of varying sensitivity and reliability. A clear-cut definition of which enzymatic and molecular properties characterize a thioredoxin is missing. Moreover there is evidence that with these proteins gel filtration experiments do not always yield valid estimates of their molecular size. Three thioredoxins present in spinach leaves have now been characterized in detail [4] and were shown to possess the typical molecular weight of 11 000- 12000 known from the bacterial and animal proteins. In contrast other studies of plant thioredoxins describe a number of protein fractions that were assigned molecular weights in the range from 8000 up to 30000 and are very difficult to compare among each other [I - 3, 5 , 111.

To alleviate some of these difficulties we have focussed our attention on the thioredoxin content of nongreen plant cells and on extensive purification of individual proteins. The presence of a thioredoxin system in wheat-flour protein fractions was reported earlier [12]. In this communication we describe the separation and properties of two and three new thioredoxins found in dry seeds of a monocotyledonous and a dicotyledonous plant, respectively, wheat (Tririrunz ursrivum) and soy bean (Glycine max). The existence of a NADPH- dependent thioredoxin reductase, not found in green tissues, is demonstrated in both plant seeds.

MATERIALS AND METHODS

Wheat seeds and various wheat-processing products were kindly provided by v. Lochow-Petkus Saatzucht, Northeim, and by Bundesforschungsanstalt fiir Getreideverarbeitung.

340

Detmold, respectively. Soy-bean seeds (‘Gieso’ variety) came from local sources. Thioredoxin and ribonucleotide reductase from Escherirhia coli, fructose-bisphosphatase and NADP: malate dehydrogenase from spinach were purified and assayed by published procedures [I 3 - 161. The thioredoxin/ thioredoxin reductase assay using 5,5’-dithiobis(2-nitro- benzoate) was also carried out as described [8, 121.

Thiorerloxitz Preparation

All preparative work and column chromatography was done at 0- 5 “C. Activation of spinach NADP:malate dehydrogenase served as routine assay of thioredoxins in column effluents. Typical thioredoxin preparations were started by extracting 100-g batches of seed flour for 3 h with 500 ml (wheat) or 2000 ml (soy bean) of 50 mM Tris/HCl buffer pH 7.5, containing 1 mM EDTA, followed by centrifugation at 30000 x g ; these extracts contained 1.5 - 2.5 mg protein/ml. The brownish supernatants were made 30 y” saturated in ammonium sulfate and the precipitates discarded. A 30- 85 :< ammonium sulfate fraction was then prepared, the precipitates redissolved in and dialyzed vs. 10 mM sodium phosphate pH 6.5 (wheat proteins) or 50 mM sodium acetate pH 5.5 (soy- bean proteins). Thioredoxins and thioredoxin reductase were then separated and purified according to the protocal outlined in Fig. 1.

Column chromatography systems were : DEAE-cellulose (Whatman, DE 32; 2.7 x 6.0 cm), equilibrated in the above mentionedpH 6.5 orpH 5.5 buffersandelutedwitha0-0.3 M NaCl gradient in the same buffers; gel filtration on Sephadex G-75 (Pharmacia; 2.8 x 75 cm) i n 50 mM Tris/HCl buffer pH 7.0 containing 5 mM EDTA; CM-cellulose (Serva; 1.0 x 3.1 cm), equlibratcd in 10 mM sodium acetate buffer pH 4.6 and eluted with a 0 - 0.2 M NaCl gradient in the same buffer; 2’,5’-ADP- Sepharose (Pharmacia; 0.4 x 3.2 cm), equilibrated and washed with 10mM potassium phosphate buffer pH 7.6 containing 3 mM EDTA, followed by elution of bound enzyme with 1 mM NADPH in the same buffer.

Polyarrylamide Gel Electrolihori,sis

System I: thioredoxin samples were made in 0.125 M Tris/HCl buffer (pH 6.8) containing 2 % SDS, 1 :,; mercaptoethanol, and I0 sucrose. Electrophoresis was carried out on 2-mm-thick slab gels. Gel concentrations: stacking gel, 4 ”/, acrylamide in 0.125 M Tris/HCl buffer (pH 6.8), 0.1 SDS; separating gel, 15 % acrylamide in 0.375 M Tris/HCl buffer (pH 8 4 , 0.1 % SDS, 10% sucrose. Electrophoresis buffer: 0.025 M Tris/glycine (pH 8.3), 0.1 (>/: SDS. System 11: samples were made in 0.125 M Tris/HCl buffer (pH 6.8) containing 4 % SDS, 1 ”/: mercaptoethanol, and 20:: glycerol. Gel concentrations: stacking gel, 9.6% acrylamide in 0.125 M Tris/HCl buffer (pH 6.8), 0.1 ::: SDS; separating gel: 167{1 acrylamide in 0.75 M Tris/HCl buffer (pH 8.8), 0.274 SDS, 3.6 M urea, 13 ”/, glycerol. Electrophoresis buffer as above. Analytical gels were stained with Coomassie brilliant blue R 250.

For preparative gel electrophoresis in system I, gel slabs (3-mm thick, 10-cm wide) were loaded with up to 0.5 ml thioredoxin sample containing up to 5 mg protein; one sample of cytochrome r served as visible molecular weight marker. After electrophoresis the unstained gels were cut horizontally into 2 - 5-mm strips. The strips were cut in pieces and placed into glass tubes (6-mm inner diameter) which had been plugged at the bottom with a thin layer of stacking gel (system I) and were

attached to a small bag of dialysis tubing. Proteins were then electrophoretically (200 V; 5 mA/tube; in Tris/glycine/SDS buffer as above) eluted from the gel and were collected in the dialysis tubing. The recovered sample was diluted to 1 ml and the protein finally precipitated by addition of acetone to a final concentration of 75 %.

RESULTS

Starting Material

Considering the high storage protein content of seeds it appeared useful to establish the best source for characterization of thioredoxin system(s) in non-green plant paterial. We had previously shown that ‘protein disulfide reductase’ described in the albumin fraction of wheat flour can in fact be resolved into thioredoxin and NADPH: thioredoxin reductase [I 21. Ground whole wheat grains, the isolated embryos [17], bran, and various types of wheat flour were analyzed but thioredoxin activity was not found concentrated in any particular fraction, including the protein-rich embryo; it rather appears evenly distributed in the entire seed [18]. Extraction of wheat flour at low salt concentration is thus the most convenient isolation procedure. Likewise no qualitative difference was observed in the thioredoxin profile of dry soy beans and of 8-day-old, etiolated seedlings worked up with or without the cotyledons. However, thioredoxin yields declined significantly in soy beans two or more years old.

Purification

Resolution of the thioredoxin systems from wheat and soy- bean is summarized in Fig. 1. Most purification steps are standard procedures in the isolation of thioredoxins and need not be described in detail. Heat treatment of the extracts ( 5 min at 70 “C) is also possible but was usually omitted because it does not improve final yields and purity of the thioredoxins while it inactivates the thioredoxin reductases. One important difference in the two preparations is the initial DEAE-cellulose chromatography which separates all three thioredoxins I, 11,111 of soy bean but leaves the two wheat thioredoxins I, 11 un- resolved (Fig. 2 and 3); the latter species are only separable on CM-cellulose (Fig. 4). Soy-bean thioredoxins thus resemble the yeast isoproteins I and I1 in chromatographic behaviour [8] and the wheat thioredoxins chromatograph like isoproteins B I and BII found in Scerzedesmus ohliquus [7]. Blue Sepharose which has successfully been used in the purification of spinach leaf thioredoxins did not bind the seed proteins. Adsorption to hydroxyapatite was possible but not of major preparative advantage. Obviously there is no single purification scheme applicable to all thioredoxins from plant sources. As multiple species of equal molecular weight cannot be recognized by gel filtration or SDS-polyacrylamide gel electrophoresis it follows that one has to perform at least two more other chromatography steps to establish the total number of thioredoxins present in an extract.

It has been difficult in each individual case to purify a seed thioredoxin to homogeneity, in contrast to microbial proteins where purification is reasonably straigthforward. Some of the numerous other albumin and globulin proteins of the M , = 10000- 20000 size category present in seeds could not be removed by any conventional chromatography step, in particular a major soy bean protein of M, about 20000 probably corresponding to a component of glycinin [19]. While these contaminants have no thioredoxin activity and do not interfere

341

CHROMATOGRAPi ON ADP-SEPHAROSE CHROMATOGRAPH ON ADP-SEPHAROSE

A REDUCTASE ELUTED THIOREDOXIN REDUCTASE ELUTED

A THIOREDOXIN PASSES THROUGH WITH 1 MM NADPH PASSES THROUGH W I T H 1 MM NADPH , /

EXTRACT SEED FLOUR I N 50 MM TRIS-HCL, PH 7.5; CENTRIFUGE

GEL F I L T R A T I O N ON SEPHADEX G-75

THIOREDOXIN 1 THlOREDOXlN I 1 ELUTED W I T H

GEL F I L T R A T I O N ON ELUTED WITH

MR N 70,000 80 MM NACL 100 MM NACL SEPHADEX 5-75

MR 70,000

P R E C I P I T A T E PROTEINS FROM SUPERNATANT A T 30-35 % ( N H Q ) ~ S @ Q SATURATION

v t

REDISSOLVE, D IALYSE, AND CHROMATOGRAPH ON DEAE CELLULOSE

SOY BEAN THIOREDOXIV SYSTEM ( P H 5 . 5 ) WHEAT THIOREDOXIN SYSTEM (PH 6 .5 )

Fig. 1. Pur$ccrriorz scIienie f t r rife fhiorerlosbz . r j ~ ~ c w s f i o n z soy-hwn mid wheui seeds. The procedures in the two bottom lines may be used alternatively to produce highly enriched yet not homogenous samples for enzyme studies, or proteins of apparent homogeneity, respectively. See text for details

m

, ,/' , , ' 1 0 20 30 LO 50 60 70

Fraction number

20 : 1.0 4

0

I

05 0 C 0

0.2 g 0.1 : m

Fig. 2. Chromarography ojsoy-herin protpbi e.Ytrai't on DEA E-cellzdo.w. The column was eluted with a 50 mM sodium acetate buffer, pH 5.5, to which a 0-0.3 M NaCl gradient was added (dashed line). Thioredoxin activity (04) was measured with spinach NADP: malate dehydrogenase (MDH) and is expressed as the factor of rate increase vs. enzyme controls without thioredoxin. Drawn line: protein absorption at 280 nm

c 1c c / @

b - L , , 0 2 mnc c

5 2.0

0 -

o c .- - E ' E

0 1.0

0 5 U C 0

0 m f 0.2

9 0.1

* / /

, , / ' , * , , ~, , 10 20 30 LO 50 60 70

Fraction number

Fig. 3. Chrornutogruph-v of wheutflour jwotrin e-vtract on DEAE-cellulose. The column was eluted with a 10 mM sodium phosphate buffer, pH 6.5, under addition of a 0-0.3 M NaCl gradient (dashed line). Thioredoxin activitywasmeasured at 41 2 nmasdithionitrobenzoate reduction (A-A, upper left scale) with an independently prepared sample of thioredoxin reductase [12], and by the NADP:malate dehydrogenase (MDH) stimulation assay ( 0 4 , right scale). Drawn line: protein absorption at 280 nm

in enzyme activation studies, it was necessary to devise a preparative method of SDS gel electrophoresis in order to obtain analytically pure samples. Due to their very close similarity in size with a coloured marker protein, cytochrome c, it was possible to locate thioredoxins precisely enough on unstained 13- 16 polyacrylamide gels. They could be ex- tracted from the cut gels and recovered after acetone pre-

cipitation or after removal of SDS with an anion exchanger. Despite such treatment, the thioredoxin samples retained sufficient and qualitatively unaltered enzyme-stimulating activity to permit their safe identification.

The resulting purity of thioredoxins treated in this way is documented in Fig. 6. With the exception of one especially difficult preparation (wheat thioredoxin I) the proteins can

343

Fract ion number

Fig. 4. C ~ ~ r o r n u l o g r u / ? h ~ of ivhc.it tliiorerloui~i on CM-C~IIUI~J,W. The column was run in a 10 mM sodium acetate buffer, pH 4.6. and eluted with a 0- 0.2 M NaCl gradient (dashcd line). Thioredoxin activity was measured with thioredoxin reductase (A---A, upper left scale) and NADP: malatc dehydrogenase (MDH) (0+, right scale) as in Fig. 3. Drawn line: protein absorption at 280 nm

Fig. 6. Analysis of seed thioredosins on ~~DS-po~J'uc~r:c~umic/c. gds. Electrophoresis was carried out in system 1 (cf. Materials and Methods). Unless noted otherwise, samples purified by preparative electrophoresis have been reanalysed. Lane 1 :marker proteins; 2: wheat thiorcdoxin 11: 3: wheat thioredoxin I before preparative electrophoresis (this species is the most difficult one to purify); 4: E. coli thioredoxin, prepared after [13]: 5: soy-bean thioredoxin I before preparative gel electrophoresis: the protein of M,= 22000 has thioredoxin activity; 6: soy-bean thioredoxin 11; 7: soy-bean thioredoxin 111; 8 : marker proteins. The marker proteins are: aprotinin, M , = 6500: cytochrome c, M, = 12500: chymotrypsinogen. M, = 25000; egg-white albumin, M, = 45000

0 5 - - 2 :: I n

QJ 0 2 -

2 0 1 c

L

- 0

0 2 0 0 5 - NADPH NoCl Q , , , , , , , , , , , , , , , , , , , ,

0 5 10 15 20 ~~

Fract ion number Fig. 5. Sepururiori of ' stij~-hcurr ihiormh~i17 nntl rhiot.orItJ.ub rech.icto.re tiu

2',5'-ADP- ,%phurosr. The column was equilibrated with 10 niM PO- tassium phosphate buffer, pH 7.6, and was eluted at the indicated fractions with 1 mM NADPH and with 1 M NaC'1 solution, respectively. Thioredoxin was measured with spinach NADP: malate dehydrogenase (MDH) ( 0 - 4 , right scale) and the reductase was dctcrmined by dilhionitrobenzoate reduction after supplementation with an indepen- dentlyprepared thioredoxin sample (0 -a, upper left scale). Drawn line: absorption of protein and coenzyme at 280 nm

finally be obtained in one-milligram amounts, starting from several hundred grams of seeds. However i t is impossible to calculate precise figures of yield and purification factor because one deals with mixtures of differently active components and because the test enzymes vary considerably from step to step in their sensitivity to thioredoxin stimulation. A reasonable estimate is that the combined thioredoxins constitute up to 0.1 "" of low salt-soluble seed proteins, but are obtained in less than I0 ";, yield as the individual species.

The thioredoxin reductase activity present in wheat [I21 and soy bean extracts is not completely separated from thioredoxin in the initial DEAE-cellulose column chromatography. However it is easy to achieve separation by passage of enzyme- containing fractions over a column of 2',5'-ADP- Sepharose which retains the reductases and does not bind thioredoxins (201 (Fig. 5). This step together with subsequent gel filtration on

Sephadex (3-75 resulted in 250-fold purification of the seed thioredoxin reductases. The thioredoxin-free and strictly NADPH-dependent enzymes can then be used for assaying seed thioredoxins. As they appear very similar to the known proteins of Escherichiu coli, yeast, or those from mammalian tissues, the seed thioredoxin reductases have not been in- vestigated further at this stage.

Enzymutic orid Other Properties

Both seed specics contain the 'classical' thiorcdoxin system consisting of NADPH-dependent thioredoxin reductase and thioredoxin(s). In combination experiments it was shown that wheat thioredoxin reductase is less species-specific than the E. c d i or yeast enzymes in that is also reduces the soy-bean (but not E. coli) thioredoxins. Reduction of the plant proteins does not proceed in the presence of either microbial thioredoxin reductase. In contrast, wheat seed thioredoxin is as good a substrate as the spinach or E. coli proteins for ferredoxin- dependent thioredoxin reductase of spinach leaves ([21]; H. Follmann and B. Buchanan, unpublished). Two basically different electron donors for thioredoxin reduction in nongreen and in green plant tissues are obvious.

In order to define more clearly the isolated seed proteins as thioredoxins, and for selecting a generally applicable routine assay we have compared the stimulatory effect of the reduced proteins in other typical enzyme systems known to react with thioredoxin, viz. E. coli ribonucleoside-diphosphate reductase (substrate: cytidine diphosphate), spinach fructose-bisphosphatase, and NADP: malate dehydrogenase; some assays with Syi7r~choc:oc~crrs PAdoPS sulfotransferase [22] were kindly pcr- formed by Prof. A. Schmidt, Munchen. The results are summarized in Table 1. Although slight stimulation of leaf

343

Wheat, I 20 40 Wheat, I1 25 50 Soy bean, I 35 40 Soy bedn, I1 35 40 soy bedn, 111 5 85

Table 1. Sfimuhtion of' thiore~o.x~~i-cicrir'clrcrl cwzj'mt..s /q srcv l t/liorr%/o.xirf isoprofeim Relative figures are given because the enzymes differ gi-catly in specific activities, and mechanism of interaction with thioredoxin. The E. c,o/ i protein which is very active in all four systems is chosen as reference. Thioredoxin samples (1 - 100 pg) were individually adjusted to produce linear increase in en7yme activities. Typical specific activities (i. c., after subtraction of non-activated rates) for thioredoxins 1 or I1 were: 1.1 nmol CDP reduced min ~ mg-', catalyzed by 20 pg E. coli reductase at 30 C. pH 7.5; 3 p o l NADPH oxidized mill- ' mg-', catalyzed by 100 pg partially purified spinach malate dehydrogenase after 120 miii activation at 22 'C, pH 7.9

20 20

I < I

Thioredoxin Ribo- NADP: Fructose- PAdoPS source and nucleotide malate bisphos- sulfo- isoprotein reductase dehydro- phatase trans-

( E . di) genase (spinach) fcrase (spinach) (Syrlecllo-

m.(.u.s)

fructose-bisphosphatase was occasionally observed in initial purification steps, our purified seed thioredoxins do not stimulate that chloroplast enzyme to any measurable extent; this inactivity is not due to denaturation during SDS gel electrophoresis because the same samples had retained good activity towards NADP: malate dehydrogenase. They can all be utilized in the heterologous ribonucleotide reductase, the malate dehydrogenase, and sulfotransferase systems ill vitro, as established for bacterial and algal thioredoxins 171. It may therefore be stated that malate-dehydrogenase catalyzed NADPH oxidation, but not the fructose biphosphatase re- action is a reliable method for assaying plant thioredoxins in general.

All purified wheat and soy bean proteins show the high stability against thermal denaturation characteristic for thioredoxins. Heating to 70 'C for up to 10 min does not reduce the enzyme-stimulating activities, and inactivation is not even complete at 90 'C. Likewise full activity is retained in the pH range from 5 - 11. Resistance towards SDS or acetone treatment without inactivation, as mentioned above, is also com- patible with such properties.

Considering the carbohydrate-rich source of preparation and the irregularities in molecular weight determination dis- cussed below the seed thioredoxins were tested for possible sugar content. N o evidence for glycoprotein nature was observed, however, when the samples were chromatographed over concanavalin-A - Sepharose or when polyacrylamide gels were stained with Schiff's reagent 1231.

Drtcmiination of Molecular Size qf Thioredo.xiris

In view of conflicting reports about the molecular weight of plant thioredoxins we have taken great care to establish the size of the new seed proteins. Gel filtration alone is clearly unacceptable. It has been found in several experiments, not detailed here, in our and in other laboratories [4] that thioredoxins including authentic E. coli thioredoxin can elute from

calibrated gel chromatography column in unexpected fractions, usually corresponding to a molecular mass several thousand daltons higher, depending upon the type of gel matrix, ionic strength, and redox state. (It is tempting to speculate that such anomalous behaviour is due to the specific protein folding of thioredoxins with a protruding, active-site chain segment [24]). In contrast, thioredoxins fit the normal, linear molecular weight scale when analyzed on SDS or urea- containing polyacrylamidegels, calibrated with commonly used marker proteins plus E. coli thioredoxin of known molecular weight. Unfortunately cell extracts contain a rather wide variety of such small proteins, making it necessary to use already extensively purified thioredoxin samples for gel electrophoretic determination of molecular weight. When analyzed at that stage in two optimized gel systems all our seed thioredoxins migrated as 12000 - 12 5000-Da polypeptides (Fig. 6). Extraction from preparative gels in undenatured form further eliminated any doubt about the identity of these protein bands. The molecular weight of 15000 for wheat thioredoxin communicated earlier 1121 thus has to be corrected.

DISCUSSION

We have demonstrated the occurrence of several thioredoxins in dry plant seeds, i.e. in the absence of light- dependent processes and of functioning chloroplasts. lntracellular compartmentation in the cytoplasm and in nongreen proplastids cannot, of course, be excluded; however the very uniform distribution of thioredoxin throughout the entire wheat seed, or thc identical thioredoxin profile of soy-bean seeds and seedlings make it unlikely that individual seed thioredoxins assume highly specific functions in the early development of a plant. Ribonucleotide reduction for deoxyribonucleotide and DNA synthesis is an obvious thioredoxin- requiring process at that stage but is difficult to test in a homologous system in v i m because plant ribonucleotide reductases are not available in purified form [25, 261. Other enzymes of seed metabolism such as sucrose synthase or glutamine synthetase may also be linked to thioredoxin activation [27, 281.

On the other hand it is remarkable that the seed thioredoxins characterized here all lack the capacity to stimulate fructose-bisphosphatase, a typical light-dependent enzyme activity. One thus expects that additional thioredoxins be formed in greening plant cells or plastids. Evidence for that order of events has been found in developing barley [5] and in a chloroplast-free mutant of Sceizedesrnus ohliquus [29] (P. Langlotr, W. Wagner and H. Follmann, unpublished). Such differentiation must also include the change from a NADPH- dependent to a ferredoxin-dependent thioredoxin reductase system.

The unambiguous molecular weight assignments made here for five seed thioredoxins, similar figures recently reported for the three major spinach leafthioredoxins [4], and the properties of two cytoplasmic thioredoxins from green algae [7] clearly demonstrate that plant thioredoxins in general are of the same molecular sire as the well-charactcrized bacterial, yeast, and animal proteins. All thioredoxins thus belong to one family of proteins of M, x 12000, composed of approximately 100 amino acids. This does not rule out that certain proteins of thioredoxin activity, yet exceptionally large size do exist, e. g. chloroplast protein of Mr = 30000 in S. ohliquus which has also been purified to homogeneity 1291; such forms might be precursors or fusion products of regular-size thioredoxins. However, molecular weight data deviating from 12000 by a few

344

thousand, established in gel filtration runs with less purified thioredoxin fractions must be reassessed by SDS-polyacrylamide gel electrophoresis, for example in the case of bean leaf or root extracts [3] or of developing barley proteins [5]. At least at the present state of knowledge such figures are no reliable indication of specifically altered, or changing thioredoxin patterns.

All plant tissues so far analyzed, green or nongreen, possess two or more thioredoxins while (non-phototrophic) bacteria and animals have only one. The observed exchdngeability in vitro of the thioredoxins from one organism as well as among thioredoxins from different organisms argues against strongly varying polypeptide sequences and/or conformations in all these proteins. The availability of homogenous thioredoxins from various sources, including plants, should now permit analysis of possible sequence homologies and, on that basis, an evaluation of why the plant proteins are more numerous.

This work has been supported by grants from Deutsche For.ve~iunRsb.etneifischrrft (Fo 50/13) and by Fonds der Chemischepi Iudustrie. Wc wish to thank Mrs Gabriele Schimpff-Weiland for her expert technical assistance.

REFERENCES

1. Wolosiuk, R. A,, Crawford, N. A., Yee, B. C. & Buchanan, B. B. (1979)

2. Crawford, N. A,, Yee, B. C., Nishizawa, A. N. & Buchanan, B. B.

3. Jacquot, J.-P., Vidal, J., Gadal, P. & Schurmann, P. (1978) FEBS Lett.

4. Schurinann, P., Maeda, K. & Tsugita, A. (1981) Eur. J . Biochern. 116.

5. Crawford, N. A,, Yee, B. C. & Buchanan, B. B. (1981) Plant Sci. Left.

6. Holmgren, A,, Buchanan, B. B. & Wolosiuk, R. A. (1977) FEBS Lett.

J. Bi(J1. Chem. 254, 1627-1632.

(1979) FEES Le// . 104, 141 - 145.

96, 243 - 246.

37-45.

317 - 326.

82. 351 - 354.

7. Wagner, W., Follmann, H. &Schmidt, A. (1978) Z. Nuturfimch. 33c,

X . Porque, P. G., Baldesten, A. & Reichard, P. (1970) J. BkJ/. Chenl. 245,

9. Jacquol, J . P., Maudinas, B. & Gadal, P. (1979) Biuchem. Biophys. Res.

51 7- 520.

2363 - 2370.

Commun. 91, 1371 -1376. 10. Schmidt, A. (1980) Arch. Microhiol. 127, 259-265. 11. Soulie,J.-M.,Buc,J., Meunier, J.-C., Pradel,J.&Ricard,J.(1981) Eur.

12. Suske, G., Wagner, W. & Follmann, H. (1979) Z. Naturforsch. 34c,

13. Laurent, T. C., Moore, E. C. & Reichard, P. (1964) J . B id . Chem. 239,

14. Brown, N . C., Canellakis, Z. N., Lundin, B., Reichard, P. &Thelander,

15. Buchanan, B. B., Schiirmann, P. & Kalberer, P. P. (1 971) J . Bid. Chrm.

16. Wolosiuk, R . A., Buchanan, B. B. & Crawford, N. A. (1977) FEBS

17. Johnston, F. B. &Stern, H. (1957) Nuture (Lo7itl.i 179, 160-161. 18. Kellmann, L. (1979) Thesis, Marburg University. 19, Spielmann, A,, Schurmann, P. & Stutz, E. (1982) Plunr Sci. Lect. 24,

20. Piglet, V. P. & Conley, R. R. (1977) J. B i d . Chem. 252, 6367-6372. 21. de la Torre, A,, Lara, C., Wolosiuk, R. A. & Buchanan, B. B. (1979)

22. Schmidt, A. & Christen, U. (1978) Pluntu (Bed . ) 140, 239-244. 23. Neville, D. M. & Glossmann, H. (1971) J . Biol. Cheni. 24Y, 6339-

24. Holmgren, A, , Soderberg, B.-0.. Eklund, H. & Brinden, C.-I. (1975)

25. Schimpff, G., Muller, H. & Follmann, H. (1978) Biochim. Biophys.

26. Hovemann, B. & Follmann, H. (1979) Biochim. Biophys. Acta, 561,

27. Pontis, H. G., Rabio, J. R. & Salerno, G. (1981) Proc. Nut[ Acatl. Sci.

28. Schmidt, A. (1981) Z. Nn/urfi~rsth. 36c, 396-399. 29. Langlotz, P. (1982) Thesis, Marburg University. 30. Wolosiuk, R. A. &Buchanan, B. B. (1978) Plant Physiol. 61,669 - 671.

J . Biochem. 119, 497 - 502.

214-221.

3436 - 3444.

L. (1969) Eur. J . Biochern. 9, 561 -573.

246, 5952- 5959.

Lett. 82, 253-258.

137-145.

FEBS LPtt. 107, 141 - 145.

6346.

Proc. Nut/ Acad. Sci. USA, 72, 2305 - 2309.

act^, 520, 70-81.

42- 52.

USA, 78, 6667 - 6669.

A. Berstermann, K. Vogt and H. Folhdni i , Arbeitsgruppe Biochem~e, Fachbercich Chemie der Philipps-Universitiit Marburg, Lahnberge, D-3550 Marburg, Federal Republic of Germany

plant seeds contain several thioredoxins of regular size

Documents