Eur. J. Biochem. 97, 503-509 (1979)

Enzymic Synthesis of Lignin Precursors Further Studies on Cinnamyl-Alcohol Dehydrogenase from Soybean-Cell-Suspension Cultures Dorothea WYRAMBIK and Hans GRISEBACH Lehrstuhl fur Biochemie der Pflanzen, Institut fur Biologic I1 der Universitat Freiburg (Received January 12, 1979)

Isoenzyme 2 of cinnamyl-alcohol dehydrogenase from soybean suspension cultures was purified about 3800-fold to apparent homogeneity by an improved purification procedure involving biospecific elution of the enzyme from a NADP'-agarose column. On sodium dodecylsulfate gels the dehydrogenase showed only one protein band with M , 40000 k 500. The enzyme is strongly inhibited by thiol reagents. Various metal chelators as well as the nonchelating 7,8-benzoquinoline also inhibited enzyme activity. Inhibition by 10 mM 1,lo-phenanthroline could be partially reversed by addition of Zn2+.1,jO-Phenanthroline and 7,8-benzoquinoline are non-competitive inhibitors with resepct to NADP'. The presence of zinc in the dehydrogenase was proved by atomic absorption spectroscopy and by specific incorporation of 65Zn into the enzyme. In steady-state kinetics inhibition patterns were obtained which are consistent with an ordered bi-bi mechanism in which NADP(H) is the first substrate to bind and the last product released. The cinnamyl-alcohol dehydrogenase belongs to the A-specific dehydrogenases and removes the pro-R hydrogen from coniferyl alcohol. The enzyme shows many similarities with alcohol dehydrogenases from horse and rat liver and from yeast.

Cinnamyl-alcohol dehydrogenase catalyses the last step in a series of reactions leading from L-phenylalanine to substituted cinnamyl alcohols, the precursors for polymerization to lignin [1,2]. The enzyme was first isolated from Forsythia suspensa [3] and cell cultures of soybean (Gljxine max L.) [4] and was found to be widely distributed in the plant kingdom [3,5j. Multiple forms have been detected by starch-gel electrophoresis in a few plants [ 5 ] . All cinnamylalcohol dehydrogenases are specific for NADP(H) as coenzyme. Two isoenzymes with quite different substrate specificities were isolated from soybean cell cultures. While isoenzyme 1 can only reduce coniferaldehyde to coniferyl alcohol (or oxidize coniferyl alcohol), isoenzyme 2 is specific for the reduction of a number of cinnamaldehydes including 4-coumaraldehyde, coniferyl aldehyde, and sinapaldehyde 141. In this publication we wish to report an improved purification procedure for isoenzyme 2 of soybean cell cultures which leads to an apparently homogenous enzyme. Further properties of this enzyme were investigated Enzymes. NAD' kinase (EC 2.7.1.23) ;isocitrate dehydrogenase (NADP*) (EC 1 . I .1.42); glucose-6-phosphate dehydrogenase (EC 1.1.1.49); alcohol dehydrogenase (EC 1.1.1.1).

to determine how closely related this dehydrogenase is to other alcohol dehydrogenases. Some of the results have been published in proceedings of a symposium [6]. MATERIALS AND METHODS Materials

NADP'-agarose (NADP' attached through ribose hydroxyls) was obtained from Sigma (Miinchen). Chelex-100, 200 - 400 mesh, was purchased from Bio-Rad labs. (Miinchen). [4-3H]NAD+ and 65ZnC12("Zn in 0.1 M HCI; 53.3 CilmoI) were purchased from Radiochemical Centre (Amersham). 7,8-Benzoquinoline was obtained from Fluka (Buchs). A11 biochemicals were purchased from Boehringer (Mannheim) or Serva (Heidelberg). NAD' kinase was a gift from Boehringer AG. Methods

The synthesis of substrates, the cultivation of cell suspension cultures of Glycine max L. var. Mandarin and the enzyme assays were carried out as described previously [4].

Cinnamyl-Alcohol Dehydrogenase

504

Enzyme Purification All operations were carried out at 4 T.The following buffers used contained 42 mM 2-mercaptoethanol and 10% (by vol.) ethyleneglycol: A, 0.1 M TrisHCI, pH 7.5; B, 20mM potassium phosphate, pH6.5; C, 10 mM potassium phosphate, pH 6.3. Extraction of the cells (2 kg wet weight); treatment with Dowex 1-X2 (step l), (NH4)2S04fractionation (step 2) and the first DEAE-cellulose chromatography (step 3) were carried out as described previously [4] with the following modifications. After (NH4)2S04 fractionation the precipitate was dissolved in buffer A and the solution dialysed against this buffer. The DEAEcellulose column (4.5 x 35 cm) was equilibrated with buffer A and proteins were eluted with a linear gradient of 1100 g each of 0.1 M and 0.35 M buffer A. Fractions containing isoenzyme 2 were pooled, protein precipitated by 80 "/, (NH4)2S04, and the protein collected by centrifugation. The precipitate was redissolved in 25 ml buffer B and dialysed against the same buffer. Step 4 . The dialysate was adsorbed on a column (3.6 x 14 cm) of hydroxyapatite (Biogel HTP, BioRad, Richmond, U.S.A.) and equilibrated with buffer B, and the column washed with 300 ml of this buffer. Elution was then performed with a linear gradient of 300 g each of 20 mM and 360 mM potassium phosphate, pH 6.5. The enzyme fraction was dialysed against buffer C. Step 5. The dialysate was applied to a column (1.5 x 11 cm) of NADP'-agarose equilibrated with buffer C, and the column was washed with seven times its volume of buffer C. Dehydrogenase was eluted with a linear NADP4 gradient from 0 to 2 mM in buffer C at a flow rate of 30 mlih. Step 6 . The pooled enzyme fractions from the affinity column (20 ml) were applied to a column of DEAE-cellulose (1.4 x 13 cm) equilibrated with buffer A, and the column was washed with three times its volume of this buffer. Enzyme was eluted with a linear gradient of 75 g each of 0.1 M and 0.4 M buffer A at a flow rate of 30 ml/h. The enzyme fractions (16 ml) were concentrated either by ultrafiltration (UM-10 membrane from Amicon) or by dialysis against 20 polyethyleneglycol 20000 (Serva, Heidelberg) in 0.01 M sodium phosphate, pH 7.1.

Sodium Dodecylsuljute Electrophoresis The method of Weber and Osborn [7] with the modification of Maize1 [12] was used. Calibrating proteins were chymotrypsinogen (MI 25 700), ovalbumin (hi(, 45 000), bovine serum albumin (hi(, 67000), and catalase (MI 60000).

Po1yacrylamicl.e Gel Electrophoresis The system 6 of Maurer [I61 was used without the upper gel. The gel was prerun for 3 h with 10 mM thioglycolic acid. The electrode buffers contained 42 mM mercaptoethanol. 100- 200 pl protein solution ( 5 - 10 pg protein) was applied to the gel. Metal Analysis A Perkin Elmer 400 atomic absorption spectrometer with graphite furnace (HGA 76) and programmer (HGA 76B) was used. Only polyethylene vessels and quartz-distilled water were used. Protein samples were dialysed before analysis against 0.01 M sodium phosphate, pH 7.1, containing 10% (by vol.) ethyleneglycol and 0.042 M mercaptoethanol. The dialysis buffer had earlier been freed from metal ions by filtration over a Chelex-100 column. The dialysis bags were pretreated according to the method of McPhie [13]. Controls were run with the phosphate buffer. Incovporution 0f65Zninto Dehydrogenase In growth medium I for the soybean cell cultures 1171zinc sulfate was replaced by zinc chloride (22 pCi, 0.41 pmol, for 40 ml medium). Five flasks containing 40 ml suspension culture each were grown for 7 days as described previously [171. Isoenzyme 2 was isolated and purified including (NH4)2S04 fractionation as described above. After step 2 the enzyme fraction was dialysed against 10 mM potassium phosphate, pH 6.5, and the dialysate (6 ml) was applied to an NADP+-agarose column. The enzyme was eluted as described above. The enzymatically active fractions were concentrated by ultrafiltration to 0.5 ml and the concentrate dialysed against 50 mM Tris-H3P04, pH 6.2. The dialysate was subjected to analytical polyacrylamide gel electrophoresis in system 6 of Maurer as described above. The gel was cut with an automatic gel cutter (Gilson Medical Electronics) into 2-mmthick discs. Each disc was eluted with 300 pl buffer A and analyzed for dehydrogenase activity and radioactivity. Synthesis of ( 4 s )-[4-3H]NADPH A modified procedure of Walton [14] was used. 2 pmol [4-3H]NAD+ (2.2 x lo8 dis./min), 5.2 pmol ATP, 20 pmol MgC12, 4 Fmol D,L-isocitrate disodium salt, 17 nkat NAD+ kinase and 17 nkat isocitrate dehydrogenase were incubated for 60 min at 30 'C in a total volume of 3 ml buffer A. The reaction was terminated by heating for 1.5 min in a boiling water bath. After removal of the protein by centrifugation the labelled NADPH was isolated by DEAE-cellulose chromatography [15].

D. Wyrambik and H. Grisebach

Synthesis of ( 4 R )-[4-3H]h'ADPH The above incubation was used with the following changes: 30 pmol MgCI2, 5 pmol glucose 6-phosphate disodium salt and 23 nkat glucose-6-phosphate dehydrogenase. Incubation qf Cinnamyl-Alcohol Dehydrogenase with Stereospecifkdly Labelled (4-3H]NADPH Two parallel incubations containing 0.48 pmol coniferaldehyde, 300 pmol (4s) or (4R)-[4-3H]NADPH, 1.44 nkat dehydrogenase and 0.2 M potassium phosphate, pH 6.5, in a total volume of 20 ml each were kept for 20 min at 30 "C. [2-3H]Coniferyl alcohol was isolated by paper Chromatography with butan-1-01 saturated with 2 % NH3. The yield of labelled coniferyl alcohol was 280 nmol.

505 Table 1. Pui?f?cation procedure .fiw isoenzyrne 2 of ~iri~~urii~~l-crlc.0h01 L1ehydr.ogenu.w 1 enzyme unit is defined as the amount of activity which cotiverts 1 nmol conifcryl alcohol/(l nkat) Purification step

Protein

Specific activity

Purification

ing

nkat/mg protein

-fold

1. Dowex supernatant

9200"

2. Ammonium sulfate (0.4-0.7 sdtd) 3. DEAE-cellulose 4. Hydroxyapatite 5. NADP+-agarose 6. DEAE-cellulose

4520 735 75 1 0.35

0.25 0.48 2.2 10.2 545 940

1

Recovery

"(,

100

1.9 8.8 41 2180 3760

95 71 34 24 14

From 2 kg wet weight soybean cells.

Oxidation oj [l-3H]Conijevyl Alcohol with Liver Alcohol Dehydrogenase The incubation mixture contained in a total voIume of 1 mI: 280 nmol [l-3H]coniferyl alcohol (3 x 10' counts/min), 1 pmol NAD', 0.27 unit alcohol dehydrogenase and 0.2 M Tris-HCI (pH 8.8). After equilibrium was reached the incubation mixture was adjusted to pH 6.5 by addition of 1 M HCI. The mixture was then extracted with ether. Coniferaldehyde was isolated from the ether phase by paper chromatography as mentioned above. The aqueous phase was adjusted with 0.5 M KOH to pH 9.0 and NADH isolated by chromatography on DEAE-cellulose [15]. 0 1 2 3 4 5 6 7 8 9 1 0 0 5 K ) 1 5 2 0 Fractim number

Detevmination of Radioactivity For 3H determination a solution of 15 g PPO in 1600 ml toluene and 800 ml Triton X-100 was used. "Zn was counted in a solution of 5 g PPO and 100 g naphthaline in 1000 ml dioxane.

Protein Determination Protein was determined with the biuret method [18] or with the method of Schaffner and Weissmann 1191. Bovine serum albumin was used as standard.

RESULTS

Fig. 1 . Affinity chrornatogrqdzy o f isorrzzynw 2 of c~itinuni~.l-rilc~ohol ciehydr-ogenaseon NADP+-ugurose.(a. -*) Dehydrogenase activity; (----) NADP' gradient; (--) protein

the isoenzymes on DEAE-cellulose 141isoenzyme 2 was chromatographed on hydroxyapatite and the dehydrogenase fraction then applied to an iVADPfagarose column. The enzyme could be eluted specifically with a linear NADP' gradient (Fig.1). This affinity step gave a 53.6-fold purification with a yield of 71 %. Essential for this purification step was to use not more than 3.6 mg protein/ml gel bed. A second DEAE-cellulose step lead to an apparently homogenous enzyme (see below).

Improved Purification qf Cinnamyl-Alcohol Dehydrogenase

Molecular Weight and Number of Subunits

The improved purification procedure for isoenzyme 2 is summarized in Table 1. After separation of

The purified dehydrogenase showed only one protein band in dodecylsulfate gel electrophoresis (Fig. 2).

Cinnamyl-Alcohol Dehydrogenase

506

-a2

-0.1

o

0.1 0.2 0.3 l/(NADP'] (pM')

Fig. 3. Double-reciprocal ploi of inhibition of cinnamyl-alcohol dehydrogenase by I,lO-phenaiithroline. 1,lo-Phenanthroline concentrations: (A--A) 1 mM, ( 0 - 4 )0.1 mM, (-0) without inhibitor. The reaction was started by addition of enzyme Fig. 2. Dodecylsulfate gel of cinnamyl-ulcohol deliydrogenasr ufter purification step 6 (Toble I ) . Electrophoresis was from cathode to anode. Front is marked with a wire

Table 2. Inhibition of' cinnamyl-alcohol dehydrogenase by tliiol reagents and metul chelators All compounds except p-chloromercuribenzoate were assayed without preincubation. p-Chloromercuribenzoate was preincubated for 2 min Compound

Concentration for 50% inhibition of dehydrogenase

Table 3. Zinc content of cinnamgl-alcohol dehydrogenase after the la.ct two purificution steps (Table 1 ) Purification step

NADP+-agaroseRib Second DEAEcellulose step a

mM A'-Ethylrnaleinimide p-Chloromercuribenzoate 1,lo-Phenanthroline 4-Methylpyrazol r,r-Dipyridyl 8-H ydroxyquinoline 7,8-Benzoquinoline

10 0.005 0.38 4 2 0.15 0.015

Determination of the molecular weight for the subunit according to the method of Weber and Osborn [7] gave a value of 40000 500. Since the M , for the dehydrogenase, estimated by its elution volume from Sephadex G-100, was about 69000 [4] the enzyme is composed of two apparently identical subunits. Inhibition of Dehydrogenase by Thiol Reagents and Metal Chelators The dehydrogenase is strongly inhibited by thiol reagents such as N-ethylmaleinimide and p-chloromercuribenzoate (Table 2). Inhibition was also observed with metal chelators and with the nonchelating

Protein

Total protein

Zinc

Specific zinc content

mg/ml

mg

pg/ml

mol/mol enzyme

0.6

1.05

0.215

0.45"

0.125

0.35

0.108

l.lh

Mean value from three determinations Mean value from four determinations.

7,8-benzoquinoline. The concentrations needed for 50% inhibition are shown in Table 2. Inhibition by 10 mM 1,lO-phenanthroline could be reproducibly reversed to a small extent by addition of 1 mM Zn". The following values are representative for such an experiment. Uninhibited reaction dA4oo/2 min = 0.09, with 10 mM 1,lO-phenanthroline dA400/2min = 0.01, after addition of 1 mM Zn2+ dA4oo/2 min = 0.03. A higher concentration of Zn2' interfered with the enzyme assay. The inhibition by 1,lO-phenanthroline and by the non-chelating 7,s-benzoquinoline [8,21] was determined quantitatively. As shown in Fig.3 a doublereciprocal plot of enzyme activity versus [NADP+] at fixed o-phenanthroline concentrations gave a series of straight lines intersecting below the abscissa. 1,lO-Phenanthroline is therefore a non-competitive inhibitor with respect to NADP'. 7,s-Benzoquinoline gave a similar inhibition pattern. Both inhibitors gave a hyperbolic inhibition in the Dixon plot.

507

D. Wyrambik and H. Grisebach

8-

-

7-

I

L

E 6-

I

in

B

5+00

/I/ \

/

m 1 c

- 5-

..->

A

1

5 4aJ

E, 3 W c

2t-

0

5

10

5

10 15 20 Fraction number

25

Fig. 4. Elution projl'le of cinnamyl-alcohol dehydrogenusefrom NADP' -ugurose. The enzyme was isolated from soybean cells grown in presence ) protein of 65ZnC12.(0- - 0 ) Dehydrogenase activity; (0-0) radioactivity; (----) NADP' gradient; (--

I

1.0

0.5

f 1

0.4

0.3 ,p W

0.2

0 Distance from cathole (crn)

Fig. 5 . Polyucrylumide gel ofcinnumjl-ulcohol dehydrogenusefructionsfrom NADP+-ugurose (Fig.4 ) . Upper part : ( ap-.) dehydrogenase; (0-0) 65Zn;BPB, bromophenylblue marker. Lower part: protein bands on gel after staining with Coomassie blue. Direction of electrophoresis was from cathode to anode

Presence of Zinc in Dehydrogenase The partial reactivation of enzyme activity by addition of Zn2 to the 1,lO-phenanthroline-inhibited enzyme pointed to the participation of Zn2+ in the catalytic reaction. The Zn content of the enzyme was determined by atomic absorption spectroscopy. Table 3 shows the zinc content of the dehydrogenase after steps 5 and 6 (Table 1) of purification. The increase in the zinc content after step 6 is correlated with the increase in specific activity. The specific zinc content found for the purest enzyme preparations was about 1 mol zinc/mol enzyme. +

Incorporation of 65Zninto Dehydrogenase Soybean cell cultures were grown in a medium in which ZnS04 was replaced by 65ZnC12(22 pCi in 40 ml

medium). Isoenzyme 2 of cinnamyl-alcohol dehydrogenase was purified as described under Materials and Methods from 46 g of wet cells. The elution profile of enzyme activity and radioactivity from the NADP+agarose column is shown in Fig.4. The peak of enzyme activity coincided with the peak of radioactivity. Fractions containing enzyme activity were concentrated by ultrafiltration and analyzed by polyacrylamide gel electrophoresis (Fig. 5). Dehydrogenase activity corresponded with the peak of radioactivity of one of the three radioactive fractions. The strong protein band towards the anode coincides with enzyme activity, whereas the second strong protein band towards the cathode contains no enzyme activity and no radioactivity. This result excludes unspecific adsorption of zinc to protein.

508

Cinnarnyl-Alcohol Dehydrogenase

Table 4. Inhibition patterns of cimumyl-ulcohol d~4ydrogenuse Inhibitor

Variable substrate

Inhibition type

K,

Is;

FM NADP' NADPH cinnamyl alcohol NADP' cinnamyl alcohol

NADPH NADP' NADPI-1 Cinnamaldeh yde Cinnamaldeh yde

competitive competitive mixed non-competitive mixed non-competit ive mixed non-competitive

150

--

100 -

. m

1

h

I

0

I

I

16 15 36 8.6

2

~

-

90 15.6 12.3

hydrogenase respectively. Parallel incubations of (4R)[4-3H]NADPH and (4S)-[4-3H]SADPH with coniferaldehyde and the dehydrogenase gave labeled (3.78 x 10' counts/min) and unlabeled (2.2 x lo2 counts/min) coniferyl alcohol respectively. The results prove the pro-R(A) specificity of the dehydrogenase with respect to NADPH. The isolated [l-3H]coniferylalcohol obtained from the above incubation was reoxidized with liver alcohol dehydrogenase, which is specific for the pro-1 R position of primary alcohols [ll]. The isolated coniferaldehyde had 347 counts/min and the NADH 29005

Kinetics of Product Inhibition Steady-state kinetics of the enzyme reaction were investigated. Lineweaver-Burk plots with substrates and coenzymes for oxidation and reduction had been shown previously to give an intersecting pattern of straight lines, as was to be expected for a sequential mechanism 141. Inhibition patterns were now investigated with the combinations listed in Table 4. The inhibition type, Ki and Ki values were obtained from Lineweaver-Burk and Cornish Bowden [9] plots. As an example the Lineweaver-Burk plot proving the competitive inhibition of NADPf versus NADPH is shown in Fig. 6. The inhibition pattern obtained is consistent with the ordered bi-bi mechanism in which NADP(H) is the first substrate to bind and the last product released [lo]. Stereospecifkity of Hydride Transfir [4-3H]NADP+ was obtained from [4-3H]NAD+ with NAD kinase. (4R)-[4-3H]XADPH and (4s)[4-3H]NADPH were then obtained by reduction with glucose-6-phosphate dehydrogenase and isocitrate de+

DISCUSSION A number of experiments were performed to investigate whether cinnamyl-alcohol dehydrogenase is a Zn-containing enzyme as has previously been found for alcohol dehydrogenase from various sources [24- 261. The presence of zinc in the dehydrogenase was proved by atomic absorption spectroscopy and by the specific incorporation of "Zn into the enzyme. The mean value of 1.1 mol zinc/mol enzyme must be considered with caution because of the error in protein determination and on account of the possibility that zinc was lost in the purification procedure. The partially reversible inhibition of the dehydrogenase by 1,lo-phenanthroline could involve zinc chelation, as has been reported for yeast alcohol dehydrogenase [20]. However, the fact that the nonchelating 7,8-benzoquinoline is also an inhibitor suggests that binding of phenanthroline and benzoquinoline could occur through hydrophobic interaction with the enzyme. In the case of yeast alcohol dehydrogenase a competitive inhibition by 1,lo-phenantholine [29], 7,8-benzoquinoline and other nitrogen bases

D. Wyrambik and H . Grisebach

509

with respect to NAD' was found (211. Anderson et al. [21] concluded that binding occurs at the pyridinium ring region of the NAD' binding site. In contrast we have found a non-competitive inhibition with respect to NADP', and binding of these bases to the substrate site could be considered [22]. The cinnamyl-alcohol dehydrogenases from both soybean cultures and forsythia [3] have now been shown to be A-specific with respect to NADPH, whereas the cinnamoyl-CoA oxidoreductase is B-specific [23]. The different stereospecificity of the two dehydrogenases catalizing consecutive reactions is an exception to the generalization expressed by Davis et al. [15,28], that NAD(P)-dependent dehydrogenases have the same stereospecificity when they catalyze consecutive steps in a metabolic sequence. The cinnamyl-alcohol dehydrogenase from soybean cell cultures bears a close resemblance to alcohol dehydrogenase from yeast [24], horse liver [25], and rat liver 1261 with respect to its subunit molecular weight, zinc content, steady-state kinetics, and stereospecificity with regard to coenzyme and substrate. More detailed studies on this relationship on the basis of protein composition and structure are unfortunately hampered by the low yield of the soybean enzyme. Although the improved purification procedure leading to an apparently homogenous enzyme included an affinity chromatography step on NADP+-agarose, which gave a 53-fold purification with a yield of 71 the fact that an approximately 4000-fold purification was necessary to obtain 350 pg pure dehydrogenase from 2 kg wet soybean cells made it difficult to obtain large amounts of enzyme for chemical characterization.

x,

The work was supported by Deutsche For.scl7ungsgernc~insc.hafr (SFB 46) and by Fm7d.s der Chr~mischenIndustrie. We thank Prof. U. Weser, Tubingen, for help in the zinc determinations.

REFERENCES 1. Grisebach. H. (1977) A~~itui~~~issensthcrfien, 64, 619- 625. 2. Gross, G . G . (1977) Recmt Adv. Phytochem. I / , 141 - 184.

3. Mansell, K.L., Gross, G. G., Stockigt, J., Franke, H. Sr Zcnk, M. H. (1 974) PI1ytoc~l7arnrstrl1., 13, 2427 - 2435. 4. Wyrambik, D . Sr Grisebach, H. (1975) Eur. J . Bioch~m.59, 9-15. 5. Mansell, R. L., Babbel, G . R . Sr Zenk, M. H . (1976) PIiJ.rochemistry, IS, 1849- 1853. 6. Griscbach, H., Wengenmayer, H. & Wyrambik, D. (1977) in Pj.ridine hlucleotide-Drpendent Delijdrogonasrs (Sund, H .. ed.) pp. 458-471, W. de Gruyter, Berlin, New York. 7. Weber, K. Sr Osborn, M. (1 969) J . Biol. Chc~rn.244,4406 - 441 2 8. Anderson. B. M. Sr Reynolds, M . L. (1966) Arch. Bio~i7o77. Biophys. 114, 299 - 308. 9. Cornish-Bowden, A. (1974) Biochenz. J . 137, 143- 144. P. D.,cd ) v o l . 11. 10. Cleland, L. W.(1970)in TlleGi~~nies(Boyer, 3rd edn, pp. 1-65, Academic Press, New York, London. 11. Bentley, R. (1970) Mokmdar. As5ymefr.y in Biologj.. vol. I I . pp. 18, Academic Press, New York, London. 12. Maizel, J. V., J r (1970) Mcthod.s Viral. 5 , 179-248. 13. Mc Phie, P. (1971) Method.7 Enzymol. 22, 23-32. 14. Walton, D. J. (1973) Bioclz~mi.s/ry,12, 3472-3478. 15. Davies, D. D., Teixeira, A. Sr Kenworthy, P.(1972) Biocheni. J . 127, 335 - 343. 16. Maurer, H . R. (1968) Di.sk-Elektro~~lzorc.sc., p. 42. W. de Gruyter, Berlin. 17. Ebel, J . , Schaller-Hekeler, B., Knobloch, K. H., Wellmann. E.. Grisebach, H. Sr Hahlbrock, K. (1974) Biochinz. Biop1ij.s Acta, 362, 417 - 424. 18. Layne, E. (1 957) Methods Enzj~mol.3, 447 - 454. 19. Schaffner, W. Sr Weissmann, C. (1973) Anul. Biochcv7. 56. 502-514. 20. Hoch, F. L., Williams, R. J. P. Sr Vallee, B. L. (1958) J . Biol. (%em. 232,453-464. 21. Anderson, B. M., Reynolds, M. L. Sr Anderson. C. D. (1966) Biochim. Biophys. Acta. 113, 235 -243. 22. Grisebach, H., Wengenmayer, H. Sr Wyrambik, D. (1977) in Pyidinc~ h~ucleotirle-Dependnt Dehyr1rogenusc.s (Sund, H.. cd.) p. 469, W. de Gruyter, Berlin, New York. 23. Gross, G . G. & Kreiten, W. (1975) FEBS LPI!.54, 259-262. 24. Branden, C. I., Jornvall, H., Eklund, H. Sr Furngren, B. (1975) in The Enzynres (Boyer, P. D., ed.) vol. XI, 3rd edn, pp. 171 173, Academic Press, New York, San Francisco, London. 25. Drum, D. E., Harrison, J. H., Li, T. K., Bethune, J. L. Sr V a lee, B. L. (1967) Proc. Ncitl Acad. Sci. U.S.A. 57, 1434- 1440 26. Arslanian, M. J . Sr Pascoe, E. (1971) Biochenz. J . 125, 10391047. 27. Klischics, M., Stockigt, J . Sr Zenk, M . H. (1978) P/iy/oc.hfw/ . ~ / i . y17, , 1523- 1525. 28. D o Nascimento, K . H. Sr Davies, D. D. (1975) Biochem. .I. 149, 553-557. 29. Hoch, F. L., Williams, R. J. P. SC ValIee, 5. L. (1958) J . Biol Cl7ern. 232,453 -464.

D. Wyrambik and H. Grisebach, Lehrstuhl fur Biochemie der Pflanzen am Biologischen Institut I1 der Albert-Ludwigs-Universitiit Freiburg, SchHnzlestraBe 1, D-7800 Freiburg i. Br., Federal Republic of Germany