Plant Cell Reports

Plant Cell Reports (1996) 15:786-790

9 Springer-Verlag 1996

Purification and characterization of an alcohol dehydrogenase from Lithospermum erythrorhizon cell cultures Shu-Ming Li, Zhao-Xin Wang, and Lutz Heide Pharmazeutisches Institut, Eberhard-Karls-Universit~itTiibingen, Auf der Morgenstelle 8, D-72076 Tiibingen, Germany Received 21 October 1995/Revised version received 3 January 1996 - Communicated by W. Barz

Abstract. An NAD-dependent alcohol dehydrogenase has been purified to apparent homogeneity from cell suspension cultures of Lithospermum erythrorhizon Sieb. et Zucc. (Boraginaceae), using protamine sulphate and ammonium sulphate precipitation and chromatography on DEAE-Sephacel, Superdex 200, hydroxyapatite and HiTrap blue. The enzyme is a homodimer with a M r of ca. 77,000. Each subunit with a Mr of 40,000 contains two zinc atoms. Its isoelectric point was found at pH 5.0. The best alcohol substrate of the enzyme is ethanol. The pH optimum for ethanol oxidation is at pH 8.7 and for acetaldehyde reduction at pH 4.6. The Michaelis constants for ethanol and NAD are 2.49 and 0.05 (pH 8.7), and for acetaldehyde and NADH 2.2 and 0.078 mM (pH 4.6), respectively. Partial amino acid sequences of the purified enzyme showed high homology to alcohol dehydrogenases from other plants. Abbreviations: ADH = alcohol dehydrogenase; DTT = dithiothreitol; PMSF = phenylmethylsulfonyl fluoride; PVPP = polyvinylpolypyrrolidone; IAA = indole-3-acetic acid; TFA = trifluoroacetic acid

Introduction Alcohol dehydrogenase (ADH; alcohol:NAD§ oxidoreductase; EC 1.1.1.1.), has been extensively studied in many organisms, e.g. in yeast or in horse liver. This enzyme also occurs in intact plants and plant cell cultures. The preferred substrate of the plant NAD-dependent ADH (EC.1.1.1.1) is ethanol, whereas the NADPdependent ADH (E.C. 1.1.1.2.) utilizes aromatic substrates (Davies et al. 1973; Yamashita et al. 1978; Igaue and Yagi 1982; Bicsak et al. 1982). Several NAD-dependent ADHs have been purified and characterized from intact plants; e.g. from strawberry (Yamashita et al. 1978), tomato (Nicolas and Crouzet 1980; Bicsak et a1.1982) and citrus fruits (Bruemmer and Roe 1971), and from tea (Hatanaka et al. 1974) and maize

Correspondence to." L. Heide

seeds (Felder et al. 1973). All these enzymes apparently belong to the well-established family of medium-chain zinc-containing ADHs (J6mvall et al. 1993). A completely different type of ADHs has been reported from insects (Danielsson et al. 1994). The main physiological function of the NAD-dependent ADH in plants is acetaldehyde reduction, allowing anaerobic glycolysis and therefore ATP production during periodes of oxygen deprivation. Therefore the enzyme is important for the response to anaerobic stress in intact plants (Freeling and Bennett 1985). Besides under oxygen deprivation, submersion under water (Irigoyen et al. 1992), and other stress factors, e.g. low temperature (Jarillo et al. 1993), the enzyme is only expressed at certain developmental stages of plant tissues, e.g. in ripening fruits. In contrast, plant cell cultures continuously show high ADH activity (Igaue and Yagi 1982, Ashihara et al. 1988, Wasternack et al. 1985). Sucrose, the usual carbon source in plant cell culture media, is rapidly hydrolyzed to glucose and fructose by the cells. The produced hexoses undergo glycolysis, and a large part of the pyruvate produced is converted to ethanol (Ashihara et al. 1988). In rice cell cultures, about 40% of the sucrose consumed from the medium is converted into ethanol, assuming that 4 mol of ethanol are produced from 1 mol sucrose (Igaue and Yagi 1982). Therefore, ADH is of major importance for the primary metabolism in plant cell cultures. Interestingly, it has been reported that ADH activity in cell cultures may not be regulated by the oxygen level (Igaue et al. 1982). Despite the metabolic importance of ADH in plant cell cultures, only a single report is available on the purification and characterization of this enzyme from such a source. Igaue and Yagi (1982) have purified ADH 36-fold from rice cell cultures. They reported the enzyme to be homogeneous after this procedure. Its Km value for ethanol was determined as 64.5 mM, which is

787

considerably higher than the reported values of ADHs from intact plant sources, e.g. from tomato fruits (2.67 mM, Bicsak et al. 1982). Therefore, there are some indications that the ADH expressed in plant cell cultures may differ in its regulation or even in some biochemical properties from the ADH expressed in intact plants. In order to investigate this question, we have purified ADH from Lithospermum erythrorhizon cell cultures to homogeneity, examined its biochemical properties and partial amino acid sequences, and compared it to the known plant ADHs. The cell cultures of Lithospermum erythrorhizon Sieb. et Zucc. are used in large-scale fermentations for industrial production of shikonin, a naphthoquinone pigment of pharmaceutical importance (Tabata and Fujita 1985), and therefore represented a suitable, well-established cell culture system for our

study. Materials and Methods Cell cultures. Cell cultures of Lithospermum erythrorhizon Sieb. et Zuec. were derived from germinating seeds (Tabata et al. 1974). Cell suspension cultures were initiated and maintained as described elsewhere (Fukui et al. 1984). Strain YM is a derivative of strain M18. For enzyme purification, YM cells were cultured in liquid LinsmaierSkoog medium supplemented with 10-6 M IAA and 104 M kinetin in the dark at 25~ Enzyme assay. ADH activity was measured at 37~ in a spectrophotometer UVicon 820 (Kontron Instruments, Munich, Germany). The total volume of each assay was 1 ml. The enzyme activity was expressed in nkat, calculated with the NADH molecular absorption coefficient of 6.2x 106. Ethanol oxidation (standard condition). The assay mixture contained 85/amol glycine-NaOH (pH 8.7), 170/amol ethanol, 0.1/amol B-NAD, and enzyme fraction. The initial rate of NADH production was measured at 340 nm after addition of ethanol to the reaction mixture. For examination of NADP dependence of the reaction, NAD was replaced by NADP. Acetaldehyde reduction. The assay mixture contained 85/amol KPi (pH 4.6), 100/amol acetaldehyde, 0.05/amol NADH and enzyme solution. The initial rate of NADH oxidation was measured at 340 nm after addition of acetaldehyde to the reaction mixture. Protein extraction and enzyme purification. All procedures were carried out on ice or between 4 and 8~ Chromatography of proteins was performed on an FPLC system of Pharmacia (Uppsala, Sweden), and protein content was either monitored by UV at 280 nm or using the method of Bradford (1976). After each chromatographic step, enzyme activity was assayed under standard conditions at 37~ All buffers were filtrated through 0.22/am Millipore filters. DTT and PMSF were added to the buffers immediately before use. Enzyme extraction, protamine sulphate and ammonium sulphate fractionation. Cells (1.4 kg) were harvested 5 days after inoculation, filtered, washed and mixed with 1.4 1 0.1 M KPi buffer, pH 6.5 containing 20 mM DTI', 40/aM PMSF and PVPP (140 g) and then ground to a fine slurry. After homogenization by a Branson sonifier (Danbury, Connecticut, USA), the homogenate was centrifuged at 17,000 x g for 20 min. A solution of 2.0% (g/ml) protamine sulphate in 0.5 M KPi (pH 6.5) was added to the enzyme extract (1 ml solution/200 mg protein). After stirring for 10 min and centrifugation at 17,000 x g for 15 min, the supematant was fractionated with solid (NH4)2SO4. The pH value of the enzyme solution was kept at 6.5 by addition of 5 M ammonia solution during the precipitating procedures. The precipitate between 33 and 55% salt saturation was collected by centrifugation at 17,000 x g for 20 min. DEAE-Sephacel chromatography. The protein pellet of the ammonium sulphate precipitation was dissolved in buffer A [50 mM Tris-HC1, pH

7.6, containing 1 mM DTT, 20/aM PMSF and 15% (v/v) glycerol] and desalted by gel filtration on Sephadex G-25 columns preequilibrated with buffer A. The desalted enzyme fraction (120 ml) was chromatographed on a DEAE-Sephacel column (2.5x25 cm, 120 ml of gel) previously equilibrated with buffer A. Proteins were eluted with a linear gradient from 0-0.25 M NaC1 in buffer A at a flow rate of 1.5 ml min -I. Fractions of 9 ml were collected and assayed for ADH activity. Superdex 200 chromatography. The active fractions of DEAE-Sephacel column chromatography (81 ml) were pooled and concentrated with an Amicon cell using a 30 kD membrane (Filtron) to a volume of 12 ml and loaded onto a Superdex 200 HR 26/60 column 0aharmacia, Uppsala, Sweden) previously equilibrated with buffer C (150 mM NaC1 in buffer A). Proteins were eluted with the same buffer at a flow rate of 1.5 ml rain l. Fractions of 6 ml were collected and assayed for enzyme activity. Hydroxyapatite chromatography. The buffer of the active fractions from Superdex 200 (18 ml) was changed to buffer D (10 mM KPi, pH 7.6 containing 1 mM DTT, 20/aM PMSF and 15% glycerol) by gel filtration on Sephadex G-25 columns, and the obtained solution was applied to a hydroxyapatite column (2.5x15 cm, 100 ml of BioGel HT, BioRad, USA) equilibrated with buffer D. The protein was eluted with a linear gradient of 10 - 150 mM KPi in the otherwise same buffer at a flow rate of 2.0 ml min l . Fractions of 5 ml were collected and assayed for ADH activity. HiTrap blue chromatography. The active fractions from the hydroxyapatite column were applied directly to a HiTrap blue column (5 ml bed volume, Pharmacia, Uppsala, Sweden) equilibrated with buffer A. After removal of unbound proteins by washing with 15 ml of buffer A, ADH was eluted with a linear gradient of 0-5 mM NAD in buffer A, and residual proteins with 2M NaCI in buffer A. Fractions of 3 ml were collected and assayed for ADH activity. Enzyme characterization. All the experiments described below were carried out using homogeneous enzyme after purification on HiTrap blue column. Affinity for substrate and coenzymes. The apparent K m values of the enzyme to substrates or coenzymes were determined by the LineweaverBurk method (Lineweaver and Burk 1934) with the computer programme Enzyme Kinetics (Trinty Software, Campton, Nit).

Chromatofocusing and determination of the isoelectric point. Chromatofocusing was performed on a Mono P column (HR 5/5) from Pharmacia (Uppsala, Sweden). Bis-Tris-HC1, 25 mM, pH 7.0 containing 15% glycerol, 1 mM DTT and 20/aM PMSF was used as start buffer, and the protein was eluted with a diluted solution (1 to 10) ofpolybuffer 74 (pH 4.0) containing 15% glycerol, 1 mM DTT and 20 /aM PMSF. Fractions of 0.5 ml were collected and assayed for enzyme activity, pH and protein concentration. Gel electrophoresis. Purity of proteins and the M r of the subunits were analysed on SDS-PAGE using a discontinuous buffer system containing 0.1% SDS (Laemmli 1970). Gels (1.5 mm, 12%) were usually run at 3040 mA constant current per gel for 4.5 h at 6 to 10~ and stained with silver nitrate (Sambrook et al. 1989). Protein concentration. The protein concentration was determined according to the method by Bradford (1976) using bovine serum albumin as a standard. Molecular weight determination. The Mr of native ADH was estimated by gel filtration on a Superdex 200 column (60x2.6 cm) using 50 mM Tris-HCl, pH 7.6 containing 0.15 M NaC1, 15% glycerol, 1 mM DTT and 20/aM PMSF as eluting buffer. The following proteins were used for column calibration: bovine serum albumin (Mr 67,000); hen ovalbumin (43,000); chymotrypsinogen A (25,000); ribonuclease A (13,700). Zinc analysis. Quantitative determination of the zinc content of purified ADH was performed by atomic absorption spectrophotometry on a Beckman 1272 M absorption spectrophotometer (Beckman, California, USA). The zinc containing sample was atomized in graphite tube furnaces of a Perkin-Ehner HGA-500 and absorption was detected at 213.86 nm. The buffer was used as a control and its value was subtracted from the sample values. The enzyme in 20 mM Bis-Tris-HC1, pH 6.0 was analyzed by the above method

788

Results and discussion Cell culture. ADH activity was examined in cell suspension cultures of Lithospermum erythrorhizon, strain YM. During 14 days culture period, ADH activity was monitored. Specific enzyme activity showed a maximum after seven days, whereas activity per g fresh weight decreased sharply after day 4 (Fig. 1). Considering these two factors, cells cultured for five days after inoculation were chosen as starting material for the enzyme purification. Extraction and separation. Crude extracts obtained by extraction of ground cells were first fractionated using protamine sulphate and ammonium sulphate precipitation. Protein precipitated at 33-55% ammonium sulphate saturation was redissolved, desalted and further purified successively by ion exchange chromatography on DEAE-Sephacel, gel filtration on Superdex 200, hydroxyapatite chromatography and semi-affinity chromatography on blue sepharose (HiTrap blue) (Fig. 2). All of the steps mentioned showed very good reproducibility. Although several proteins bound to HiTrap blue, ADH could be selectively eluted by a gradient of its coenzyme f3-NAD (0-5 mM) from the column. The total purification procedure resulted in an increase in specific ADH activity by a factor of 110, compared to the crude extract (Tab. 1). The purified enzyme showed a single band at about 40,000 on SDSPAGE (Fig. 3) after staining with silver nitrate (Sambrook et al. 1989). Stabili(y of the purified enzyme.The purified enzyme was stable at -70~ No loss of enzyme activity was observed after freezing in liquid N2 and storage at -70~ for 4 months in 50 mM Tris-HC1, pH 7.6, containing 15% glycerol and 10 mM DTT. ~-,.3o S

~nkat/mg protein ~ nkat/gcells +growth

-6

o.35-

-/~

UV-280 -a- nkat/ml

a )

~

o.=~-

~

~o ~

~.

40 0.15-

E

~.. ~

~

~

_

Fraction

o.14-

--

o.,~:

no.

UV-280

b)

~

_6o

~

~oo~

~o~ _ /

~'oo6

I

_-30

J

o.o.-

Fraction

O.OTq

o

~ ~ ~

--

UV-280 -*-

no.

nkat/ml

-35

c) -2m .~

o.o5 0.04

9

o 03

-20

~ ~

-~5 N

~

-~o~

~

no.

no

~

loo

d)

/-/g/ml ' ~

nkat/ml

~

:x4o

2 M NaCI : 1 o o

6o

~

0

~.~

l

5 mM NAD

2~

~

-70

nkat/ml

5

N

N

20 =zs~

0.05-

m ~__,15.

.>

-; ~

0 M NaCl

o. 1 -

Fraction

N 25

-eO

O.S M NaCI

e~

4o

qt

; Fraction

no.

days after inoculation Fig.l: Time course of ADH activity in cultured cells of Lithospermum erythrorhizon: Verticalbars showdivergenceof two independentresults.

Fig. 2:Purification of ADH from Lithospermum erythrorhizonby column chromatography: a) DEAE-Sephacel;b) Superdex200 HR 26/60, c) Hydroxyapatite;d) HiTrapblue. For detailssee materialand methods.

789

Table 1: Purification of ADH from L, erythrorhizon Protein T o t a l Specific Purifiation Yield Purification (mg) activity activity (fold) (%) steps (nkat) (nkat/mg) 8.24 1 100 Crude extract 1925 15862 1755 15062 8.58 1.04 95.0 Protamine sulphate 1310 14170 10.8 1.31 89.3 (NH4)2SO4 (33-55%) 880 13589 15.4 1.87 85.7 Sephadex G-25 97.2 2664 27.4 3.33 16.8 DEAESephacel 23.6 2218 94.0 11.4 14.0 Superdex 200 6.16 1727 280 34.0 10.9 Hydroxyapatite 0.44 397 903 110 2.50 HiTrap blue

activity was shown towards methanol. Likewise, the enzyme showed almost no activity towards aromatic, alicyclic and secondary aliphatic alcohols (Tab. 2). Determination o f Km values for the substrates of the reversible reaction CH3CH2OH + N A D + +-" CH3CHO + N A D H + H + gave 2.49 and 2.2 mM for ethanol and acetaldehyde, respectively, and 0.05 or 0.078 for NAD + and NADH, respectively. The Vm~, value for ethanol oxidation was approximately 1230, for acetaldehyde reduction approximately 2220 nkat/mg, respectively. The turnover number for ethanol and N A D was calculated to be 2940 min -1. Table 2: Substrate specificity of ADH from L. erythrorhizon alcohol relative Km Vmaxnkat/mg activity mM protein