Aspartate-beta-Semialdehyde Dehydrogenase from Escherichia coli. Purification and General Properties

Eur. J. Biochem. 104, 53-58 (1980)

Aspartate-P-Semialdehyde Dehydrogenase from Escherichia coli Purification and General Properties Jean-Frdnqois BIELLMANN, Pierre EID, Christian HIRTH, and Hans JORNVALL Laboratoire associi. au CNRS, Institut de Chimie, Universite Louis Pasteur, Strasbourg, and Karolinska Institutet, Kemiska Institutionen I, Stockholm (Received March 30, 1979)

Aspartate-P-semialdehyde dehydrogenase, from an Escherichia coli mutant derepressed for the biosynthesis of L-lysine, has been purified to homogeneity. Its isoelectric point is pH 4.3. This enzyme has a molecular weight of 77000 and is composed of two identical or highly similar subunits of molecular weight 38000 f 2000. Their N-terminal amino-acid sequence is Met-Lys-Asx-Val-Gly-. Three cysteine residues per subunit were detected: two are reactive in the native enzyme and one is partially protected by the substrate. Formation of an acyl-enzyme intermediate was also detected. Correlation of the 'H nuclea'r magnetic resonance spectrum of [4-'H]NADPH produced from [4-2H]NADP+indicated that aspartate P-semialdehyde dehydrogenase transfers the p r o 3 hydrogen from NADPH (class B dehydrogenase). A short comparison with the corresponding yeast enzyme is given. Aspartate-P-semialdehyde dehydrogenase is one of the three enzymes of the common pathway from L-aspartic acid to L-lysine, L-methionine, L-threonine and L-isoleucine. It catalyzes the reduction of p-asparty1 phosphate to aspartate 8-semialdehyde by NADPH according to the scheme:

L-HOOC- CH(NH2)- CH2- COOPOg

+ NADPH

11 L-HOOC- CH(NH2)CHz - CHO + PO:

+ NADP'.

The general outline of this reaction is similar to that of glyceraldehyde-3-phosphate dehydrogenase. The latter enzyme, extracted from different sources, has been extensively studied, including its cooperativity of coenzyme binding and its half-site reactivity with some reagents [I j. It has been proposed that it carries Ahbveviations. NMR, nuclear magnetic resonance; dansyl, 5-dimethylaminonaphthalene-I -sulfonyl ; QAE, quaternary diethyl(2-hydroxypropy1)aminoethyl. Definition. 1 enzyme unit, U, is defined as the quantity of enzyme able to reduce 1 pmol NADP+/min at 25 C in 0.02 M triethanolamine-HC1 buffer, pH 8. Enzymes. L-Aspartate-fi-semialdehyde :NADP' oxidoreductase (phosphorylating) (EC 1.2.1.11) ; ~-glyceraldehyde-3-phosphate: NAD' oxidoreductase (phosphorylating) (EC 1.2.1.12); glucose6-phosphate dehydrogenase (EC 1.1.1.49); isocitrate dehydrogenase (NADP') (EC 1.1.1.42); aldehyde dehydrogenase (EC 1.2.1.3); hydroxymethylglutaryl-CoA reductase (NADPH) (EC 1.1.1.34).

out the catalytic reaction through a thioester and a thiohemiketal. In order to investigate similarities and differences between phosphorylating aldehyde dehydrogenases, structural and mechanistic studies of the aspartate8-semialdehyde dehydrogenase were undertaken. This enzyme had been purified from yeast [2,3] and from Escherichia coli [4j. In the present study an improvement of the purification of the E. coli enzyme is given, and the amino acid composition, chemical and physicochemical properties are described. MATERIALS AND METHODS

Chemicals and Buffers Pig heart isocitrate dehydrogenase was from Sigma, yeast glucose-6-phosphate dehydrogenase, glucose 6-phosphate, sodium isocitrate and NADP' from Boehringer Mannheim, DEAE-cellulose 23 from Whatman and DEAE-Sephadex A- 50, quaternary diethyl (2-hydroxypropyl)aminoethyl (QAE)-Sephadex and Sepharose 4B from Pharmacia Fine Chemicals. Hydroxyapatite Bio-Gel HTP, P-2 and P-10 was from BioRad Laboratories and Ultrogel ACA-44 from LKB. Other chemicals were of the best available grade. Sepharose C3 was prepared according to Shaltiel [5j.

54

Buffer 1 contained imidazole (12.25 g 1-I), barbital (free acid, 3.78 g 1 '), EDTA disodium salt (0.372 g 1-') and dithioerythritol (1.5 g 1-'). The pH was adjusted to 8.0. Buffer 2 contained 0.01 M potassium phosphate pH 7.2, 0.1 mM EDTA and 1 mM dithioerythritol. Buffer 3 contained 0.01 M potassium phosphate pH 7.2,0.5 M potassium chloride, 0.1 mM EDTA and 1 mM dithioerythritol. Enzyme Purification Escherichia coli K12 strain OE.73 derepressed for aspartate-b-semialdehyde dehydrogenase biosynthesis was a gift from Prof. J. C. Patte. It was cultured in Gif-sur-Yvette by Dr Hours. This strain gives at least a fivefold increase of the quantity of aspartate /3-semialdehyde dehydrogenase compared to the E. coli K12 strain. Cell-free extracts in 30 mM potassium phosphate buffer, pH 7.2, were prepared by treatment in a Gifford-Wood homogenizer (Hudson, NY, U.S.A.) in the presence of glass beads (0.1 mm diameter) followed by centrifugation at 15000 x g (1 h). The DEAE-cellulose (200 g) was equilibrated in 30 mM phosphate buffer, pH 6.8. The filtered DEAEcellulose was added to the protein solution. After filtration, the gel was suspended in the same buffer and filtered. The enzymic activity was eluted with 350 mM phosphate buffer, pH 6.8. The enzyme was dialyzed against buffer 1, ten times diluted and containing 0.15 M potassium chloride, with several changes. The DEAE-Sephadex column (3 x 30 cm) was equilibrated with the same buffer. A linear gradient (2 x 1 1) of potassium chloride (0.15-0.5 M) was applied. The enzyme was eluted at 0.35 M. The protein was dialyzed against buffer 2 (without EDTA) and the hydroxyapatite column (3 x 20 cm) was equilibrated and eluted with the same buffer (without EDTA). The enzyme was eluted after the void volume. After ammonium sulfate precipitation (65 % saturation), the concentrated enzyme solution, 20 mg ml-' in buffer 2, was layered on an Ultrogel ACA-44 column (4 x 80 cm). The column was equilibrated and eluted with buffer 2. The enzyme (2mg ml-' in buffer 2) was adsorbed on a Sepharose C3 column (3 x 35 cm) equilibrated with buffer 3 containing 0.5 M ammonium sulfate. A linear gradient from 0.5 M ammonium sulfate in buffer 3 (0.5 1) to buffer 3 (0.5 1) was applied [6]. The enzyme was eluted at 0.2 M ammonium sulfate. The enzyme (10 mg ml-' in buffer 2, 50 % glycerol) was stored at - 20 "C in sealed tubes under argon.

Aspartate P-Semialdehyde Dehydrogenase

0.02 M triethanolamine-HC1 buffer pH 8.0. The substrate aspartate b-semialdehyde was prepared from allylglycine [7]. The protein concentration was monitored at 280 nm or estimated by the biuret method [8]. Determination of the Absorption Coejficient

The absorption coefficient A l',",P at 280 nm was determined using a solution of known absorbance and measuring the protein concentration by quantitative analysis of the amino acid composition of the solution and by weighing dried aliquots. Ultracentr ijiugation

Equilibrium experiments were carried out (near 23 "C) with a Beckman model E ultracentrifuge equipped with interference optics in a 1.2-cm double-sector cell during 31 h at a rotor speed set at 20410 rev./min. Photographic plates were taken with a Nikon profile projector model 6ST2. Molecular Weight of the Enzyme Subunil

The subunit molecular weight was determined by sodium dodecyl sulfate gel electrophoresis according to Weber and Osborn [9]. Isoelectric Focusing

Studies were performed either on acrylamide sheets using an LKB ampholine polyacrylamide gel plate (pH 3-9.5) or in a column (1.5 x 25 cm) containing a linear gradient of glycerol (lO-50%) and LKB ampholine (4 %, pH 3 - 6). Amino Acid Composition

For carboxymethylation, the protein was dissolved (10 mgml-') in 6 M guanidine-HC1,O.l M Tris, 5 mM EDTA, pH 8.1, reduced with dithiothreitol (0.4 pmol/ mg protein) for 2 h at 37 "C, and then reacted with neutralized iodo [2-'4C]acetic acid (1.25 pmol/mg protein) for 2 h at room temperature. After dialysis against distilled water, samples of the carboxymethylated protein were hydrolyzed with 6 M HCl/ 0.1 % mercaptoethanol in evacuated tubes at 110 "C for 20 h, 48 h and 72 h (triplicate samples for each time). Amino acids were determined on a Beckman 121 M amino acid analyzer. Tryptophan was determined after hydrolysis with methanesulfonic acid.

Enzymic Activity

Sequence Determination

Enzymic activity was determined by reduction of NADP' with aspartate P-semialdehyde at 25 "C in

The N-terminus was determined with the dansyl method after dissolving the protein in 8 M urea [lo].

55

J.-F. Biellmann, P. Eid, C. Hirth, and H. Jiirnvall

The subsequent residues were similarly determined on different protein samples that were first submitted to Edman degradations. Dansyl amino acids were detected by thin-layer chromatography on polyamide sheets I l l ] in four solvent systems as previously described [121.

(30 x 2 cm) with a 50 - 500 mM elution gradient of potassium chloride. Salts were removed from the NADP' solution by passage through a Biogel P-2 column(100 x 3 cm). The deuterium content at position C-4 of the isolated NADPt was higher than 95% according to the NMR spectroscopy.

- SH

Enzymic Reduction o ~ ( ~ - ~ H ] N A D P +

Titration

Measurements were carried out at room temperature with 0.5 mM 5,5'-dithiobis(2-nitrobenzoic acid) using 0.1 M sodium phosphate buffer or TrisHCl, pH 8. The enzyme concentration was 5 pM (subunit concentration). Denaturation agents were 1 % sodium dodecyl sulfate or 8 M urea [13]. Esterase Assay The hydrolysis of p-nitrophenyl acetate was followed as is described in [14]. Esterase assays were performed at 30 "C in a total volume of 1 ml, containing 10 mM sodium barbital buffer pH 8.0, 1 mM EDTA, and 0.2 mMp-nitrophenyl acetate. Before use, the enzyme was dialyzed extensively against 10 mM sodium barbital buffer pH 8.0, 1 mM EDTA. Formation of Acyl-enzyme Phosphate ions were removed from the enzyme solution by extensive dialysis against 10 mM imidazole-HCI buffer, pH 7.2, or using a Sephadex G-25 column (20 x 1.5 cm) equilibrated with the same buffer. The absorbance at 340nm of a solution (0.2 ml) of aspartate-/?-semialdehyde dehydrogenase (2- 3 mgiml), 0.5 mM NADP+ and 0.5 mM aspartate P-semialdehyde in the same buffer, was followed. After 30 min the acyl-enzyme was isolated at 0 "C by chromatography on a Biogel P 10 column (11 x 1.4 cm) using the same buffer. The protein was detected by absorption at 280 nm. To the solution containing the acyl-enzyme (concentration determined by the absorption at 280nm as for the apoenzyme) a 20 mM solution of NADPH (5 pl) was added and the decrease of the absorption at 340 nm was measured. Preparation of [4-2H ] N A D P ' NADP+ (0.76 g) was dissolved in a 1 M potassjum cyanide solution in 2H20 (12 ml), to which a 5 M potassium hydroxide solution (0.4 ml) was added [15]. After 4 h at 20 "C, the pH of this solution was brought to 5.0 by the addition of Dowex 50 (washed with 'H20). After filtration the solution was freeze-dried. The residue was dissolved in 'H20 and the exchange procedure was performed a second time. The next step was chromatography on DEAE-Sephadex A-25

With Aspar ta te-/?- Sem ialdehyde Dehydrogenase .To a 10 mM sodium bicarbonate and 40mM sodium arsenate solution (20 ml, pH 8.0) of [4-2H]NADP+ (50 mg), aspartate-/?-semialdehyde dehydrogenase (2 mg) was added, followed by successive aliquots of aspartate P-semialdehyde until no increase of absorbance at 340 nm was detected. With Glucosed-Phosphate Dehydrogenase. The reduction of [4-'H]NADP+ (50 mg) by glucose 6phosphate (50 mg) in the presence of glucose-6-phosphate dehydrogenase (1 mg) was performed in a 50 mM potassium phosphate buffer pH 7.6, 60 mM magnesium chloride (50 ml). The reaction went rapidly to completion, after addition of the enzyme. With Isocitrate Dehydrogenase. The reduction of [4-'H]NADP+ (50 mg) by sodium isocitrate (20 mg) in the presence of isocitrate dehydrogenase (1 mg) was performed in a 0.1 M imidazole-HCI buffer pH 8.0, 30 mM magnesium chloride (50 ml). The reaction was stopped when about 35 mg NADPH were obtained. Purification of Enzymatically formed [4-2H]NADPH [4-2H]NADPH was isolated by chromatography on QAE-Sephadex (20x2 cm; 10 mM sodium bicarbonate, 50 mM sodium chloride) using a 50 mM 0.5 M elution gradient of sodium chloride. [4-2H]NADPH was eluted at 0.35 M, and desalted on a Biogel P2 column (100 x 3 cm) equilibrated with 10 mM ammonium bicarbonate solution. The solvent was removed by lyophilization. Determination of the Stereochemistry of the Hydride Transfer Samples were dissolved in 2H20containing sodium trimethylsilyl ('H4)propionate as the internal reference. p2H was brought to 7.6 (not corrected). Spectra were taken at 25°C with a Cameca 250-MHz nuclear magnetic resonance spectrometer. RESULTS AND DISCUSSION The extraction method of the enzyme, described by Hegeman et al. [4] was modified in order to permit work on a larger scale. A purification of the enzyme is summarized in Table 1.

56

Aspartate 8-Semialdehyde Dehydrogenase

Table 1. Pur fieation scheme of aspartate-P-semiuldehydedehydrogenase from E. cob Steps starting from 700 g cells (net weight)

1. 2. 3. 4. 5. 6.

Crude extract DEAE-Cellulose DEAE-Sephadex A-50 Hydroxyapatite Ultrogel ACA-44 Sepharose C3

Volume

Total activity

Protein concentration

Specific activity

ml

U

mg m1-I

U mg-'

700 1000 190 220 60 1so

67 700 43 SO0 33 700 21 600 14500 12400

-

-

8.1 7.6 1.5 2.3 0.73

5.4 23 65 100 10s

The highest specific activity previously described for the enzyme was 120 U mg-' determined at pH 9.0 [4]. The values obtained in the present activity test are not directly comparable to those observed previously since the instability of the substrate at high p H led us to work at pH 8.0. The previous activity test [4] gave for pure preparation an activity of 150 U mg-l. The enzyme was stable for several days in buffer 2, at 4°C. It was stored in buffer 2 in the presence of 50 "/, glycerol at - 20 "C. Under these conditions, no loss of activity was detected within two months. A final specific activity of 100- 110 U mg-l was routinely obtained. The enzyme proved to be homogeneous by electrophoresis, ultracentrifugation, electrofocusing and N-terminal sequence determination : dansylation in 8 M urea revealed that the N-terminus is methionine. Similar determinations after successive Edman degradations showed the N-terminal sequence of the protein to be : Met-Lys-Asx-Val-Gly-. The molecular weight of the native enzyme determined by ultracentrifugation is 77500 f 1500 [16]. The subunit molecular weight on sodium dodecylsulphate gel electrophoresis is 38000 2000. This is in agreement with the results obtained earlier from an impure preparation by centrifugation of the active enzyme [17]. Therefore, the protein is dimeric and composed of two probably identical subunits. In contrast to aspartate-P-semialdehyde dehydrogenase from Escherichia coli, the same enzyme from yeast is tetrameric and composed of four identical subunits of molecular weight 41 000 f 4000. Glyceraldehyde3-phosphate dehydrogenase has been isolated as a tetramer from numerous sources, including both yeast and E. coli [19]. The absorption spectrum ( A : :$?&:, = 1 ; AZ8,,/ A260 = 1.8) agrees with the absence of any bound nucleotide [18]. Charcoal treatment, used for the removal of bound nucleotides from glyceraldehyde3-phosphate dehydrogenase [ l ] did not change the absorption spectrum. The isoelectric point is low (pH 4.3) in agreement with the behavior of the enzyme during isolation.

+

Yield

z 100 70 52 35 21 18

Table 2. Results of amino acid analysis of carboxymethylated aspartate-8-semiuldehydedehydrogenase from E. coli Values for serine, threonine, valine and isoleucine estimated from the hydrolysis curves as given in the text Amino acid

Amount

Nearest integer per subunit of 38000M,

mo1/100 mol

Carboxymethylcysteine Aspartic acid Threonine Serine Glutamic acid Proline Glycine Alanine Valine Methionine Isoleucine Leucine Tyrosine Phenylalanine Tryptophan Lysine Histidine Argininc Total

0.9 9.6 6.6 5.7 9.1 4.4 10.1 8.4 7.5 2.6 5.1 11.6 1.4 3.3 1.3 4.4 1.5 6.5

3 34 23 20 32 15 36 30 26 9 18 41 5 12 5 15

5 23

100.0

352

The results of the amino acid analysis are given in Table 2. Values for serine and threonine have been extrapoled to zero time and those for valine and isoleucine are taken at 72 h. The amino acid composition indicates the presence of three cysteine residues per subunit agreeing with the 5,5'-dithiobis(2-nitrobenzoicacid) titration under denaturing conditions. The native enzyme has two titratable - SH per subunit, which are not protected by the coenzymes NADP' or NADPH (Table 3). The substrate aspartate P-semialdehyde partially protects one - SH per subunit. In the presence of NAD' and aspartate p-semialdehyde and in the absence of any acyl acceptor, a single turnover of the enzyme is observed, leading to the formation of a stable acyl-enzyme. The amount of NADPH produced corresponds to 1 mol/subunit.

57

J.-F. Biellmann, P. Eid, C. Hirth, and H. Jornvall Table 3. Determination of thiol groups Total -SH groups were estimated in the presence of 1 % sodium dodecyl sulfate or 8 M urea Enzyme

Reactive thiol groups/subunit

Total - SH groups/ subunit

Native With 0.8 mM NADP’ With 1 mM NADPH With 1 mM aspartate p-semialdehyde

2 2 1.7

3 3.2 2.7

1.2

2.2”

Table 4. N M R data of NADPH Chemical shifts taken from sodium trimethylsilyl (2H4)propionate internal standard (pH 7.6). Concentration was 20 mg NADPH/ ml. The spectra were taken at 25°C with a Cameca 250-MHz NMR spectrometer. The chemical shifts of protons at C-8 and C-2 of the adenine ring and at C-2 of the dihydronicotinamide ring are indicated in order to show the reproductibility Sample

Adenine protons Dihydronicotinat amide protons at c-8

a With sodium dodecyl sulfate a precipitate appears. T, therefore, measurement was only made with 8 M urea

8.48

[4-’H]NADPH from [4-’H]NADP+ With glucose-5-phosphate dehydrogenase (B) 8.48 With isocitrate-dehydrogenase (A) 8.48 With aspartate-p-semialdehyde dehydrogenase 8.48

0

10 Time (rnin)

20

Fig. 1. Formation of the acyl enzyme. Amount (mol) of NADPH formed at 20°C per subunit as a function of time in a mixture of aspartate-p-semialdehyde dehydrogenase (80 pM subunit concentration), 0.5 mM NADP’ and 0.5 mM aspartate p-semialdehyde in a phosphdte-free medium (10 mM imidazole-HC1 pH 7.2)

A slow spontaneous hydrolysis of the acyl-enzyme could explain the slow increase of the absorbance due to the formation of NADPH with time (Fig.1). However, the acyl-enzyme could be isolated by rapid chromatography and reduced by NADPH. This experiment indicates the formation of an acyl-enzyme which has previously been found also for glyceraldehyde-3-phosphate dehydrogenase [20]. The esterolytic activity towards p-nitrophenyl acetate is low. For glyceraldehyde-3-phosphate dehydrogenase from yeast and rabbit muscle [ l ] and for aspartateP-semialdehyde dehydrogenase from yeast [21], a high esterolytic activity is observed. But this activity does not seem to be a general property in aldehyde dehydrogenases not even among the glyceraldehyde3-phosphate dehydrogenases where the apoenzymes from sturgeon and from Bacillus stearothermophilus show only a slight activity (M. A. Abdallah and L. Wallkn, unpublished results). N M R spectroscopy was used to determine the stereochemistry of the hydrogen transfer with NADP’

c-2

c-4

8.23

6.93

2.83 and 2.71

8.24

6.93

2.72

8.24

6.93

2.80

8.23

6.94

2.71

PPm NADPH commercial

0

c-2

[23]. The two protons at C-4 of NADH or NADPH are diastereotopic and thus distinguishable by N M R spectroscopy. Indeed for NADH and NADPH, the signals attributed to these protons have different chemical shifts; for NADH the signal at high field was shown to originate from the p r o 3 proton [23]. Prepared [4-’H]NADP+ was reduced to [4-*H]NADPH in the presence of aspartate-P-semialdehyde dehydrogenase and two enzymes of known stereochemical course : glucose-6-phosphate dehydrogenase from yeast which transfers the p r o 3 hydrogen (class B dehydrogenase) [24] and isocitrate dehydrogenase from pig heart which transfers the pro-R hydrogen (class A dehydrogenase) [25,26]. The spectroscopical results (Table 4) show that aspartate P-semialdehyde dehydrogenase transfers the pro-S hydrogen (class B deh ydrogenase). The stereochemistry of the aldehyde dehydrogenases have been determined only for a few systems disregarding the dehydrogenases acting on sugar aldehydes. The aldehyde dehydrogenase from beef liver [27] and hydroxymethylglutaryl-CoA reductase (NADPH) from yeast and rat liver [28,29] belong to the A class dehydrogenase. Glyceraldehyde-3-phosphate dehydrogenase belongs to the B class dehydrogenase [30]. The stereochemical courses are therefore different for various aldehyde dehydrogenases. While clear differences appear between E. coli aspartate-P-semialdehyde dehydrogenase and glyceraldehyde-3-phosphate dehydrogenases from different sources in the quaternary structure, the formation of an acyl-enzyme suggests a catalytic similarity for these two enzymes.

58

J.-F. Biellmann, P. Eid, C. Hirth, and H. Jornvall : Aspartate p-Semialdehyde Dehydrogenase

We would like to thank Mr G. De Marcillac for performing the ultracentrifugation and F. Hemmert and R. Graff for NMR spectroscopy. This work was supported in part by grant 73-7-1312 from Dtitgation Gtntrale a la Recherche Scientijique et Technique as well as by funds from the Fondation pour la Recherche Midicale FranGaise and the Swedish Medical Research Council (project 13X-3532).

REFERENCES 1 . Harris, J. & Waters, H. J. (1976) in The Enzymes (Boyer, P. D., ed.) 3rd edn, vol. 13, pp. 1-49, Academic Press, New Nork. 2. Surdin, Y. (1967) Eur. J. Biochem. 2, 341-348. 3. Holland, M. J. & Westhead, E. W. (1973) Biochemistry, 12, 2264- 2270. 4. Hegeman, G. D., Cohen, G. N. & Morgan, R. (1970) Methods Enzymol. 17A, 708-713. 5. Shaltiel, S. (1974) Methods Enzymol. 34B, 126-139. 6. Raibaud, O., Hogberg-Raibaud, A. & Goldberg, M. E. (1975) FEBS Lett. 50, 130-134. 7. Black, S. &Wright, N. G. (1955) J . Biol. Chem. 213, 39-50. 8 . Gornall, A. G., Bardawill, C. J. & David, M. M. (1949) J . Biol. Chem. 177,751 - 766. 9. Weber, K. & Osborn, M. (1969) J . Biol. Chem. 244,4406-4412. 10. Bruton, C. J. & Hartley, B. S. (1968) Biochem. J . 108,281-288. 11. Woods, K. R. & Wang, K.-T. (1967) Biochim. Biophys. Acta, 133, 369 - 370. 12. Jornvall, H. (1970) Eur. J . Biochem. 14, 521-534.

13. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70-77. 14. Kezdy, F. J. & Bencler, M. L. (1962) Biochemistry, 1, 10971106. 15. San Pietro, A. (1955) J . Biol. Chem. 217, 579-587. 26. Yphantis, D . A. (1966) Biochemistry, 3, 297-317. 17. Cohen, R. & Mire, M. (1971) Eur. J . Biochem. 23,276-281. 18. Kirschner, K. & Voigt, B. (1968) Hoppe-Seyler’.y Z. Physiol. Chem. 349,632- 644. 19. D’Alessio, G. & Jasse, J. (1971) J . Biol. Chem. 246,4326-4333. 20. Seydoux, F., Bernhard, S., Pfenninger, O., Payne, M. & Malhotra, 0. P. (1973) Biochemistry, 12, 4290-4300. 21. Holland, H. J. & Westhead, E. W. (1973) Biochemistry, 12, 2276 - 228 1. 22. Reference deleted. 23. Arnold, L. J., Kwan-sa Yon, Allison, W. S. & Kaplan, N. 0. (1976) Biochemistry, 15,4844-4849. 24. Stern, B. K. & Vennesland, B. (1960) J . Biol. Chem. 235, 205 - 208. 25. Nakamoto, T. & Vennesland, B. (1960) J . Biol. Chem. 235, 202 - 204. 26. Englard, S. & Colowick, S. P. (1957) J. Bid. Chem. 226, 1047- 1058. 27. Levy, H. R. & Vennesland, B. (1957) J . Biol. Chem. 228, 85 - 96. 28. Dugan, R. E. & Porter, J. W. (1971) J . Biol. Chem. 246, 5361 5364. 29. Beedle, A. S., Munday, K. A. & Wilton, D. C. (1972) Eur. J . Biochem. 28, 151 - 155. 30. Loewus, F. A., Levy, H. R. & Vennesland, B. (1956) J . Biol. Chem. 223.589 - 597.

J. F. Biellmann, C. Hirth* , lnstitut de Chimie, Universite Louis Pasteur, Rue Blaise-Pascal 1, BP 296/R8, F-67009 Strasbourg-Cedex, France P. Eid, Unite d’Ecologie Virale, Institut Pasteur, Rue du Docteur-Roux 25, F-75724 Paris-Cedex 15, France H. Jornvall, Karolinska Institutet, Kemiska Instutionen I, S-10401 Stockholm 60, Sweden

* To whom correspondence should be addressed.