Purification and characterization of glycerol-3-phosphate dehydrogenase (NAD+) in the salt-tolerant yeast Debaryomyces hansenii

180

Biochimica et Biophysica A cta, 1034 (1990) 180-185

Elsevier BBAGEN 23300

Purification and characterization of glycerol-3-phosphate dehydrogenase (NAD +) in the salt-tolerant yeast Debaryomyces hansenii Anders Nilsson and Lennart Adler Department of Marine Microbiology, University of Grteborg, Grteborg (Sweden)

(Received28 December1989)

Key words: Glycerol-3-phosphatedehydrogenase(NAD+); Salt tolerance; Glutamate; Osmoregulation;(D. hansenii)

The NAD-dependent glycerol-3-phosphate dehydrogenase (EC 1.1.1.8) of the salt-tolerant yeast Debaryomyces hansenii was purified by poly(ethylene glycol) precipitation and a combination of chromatographic procedures. The enzyme existed in two forms with different ionic characters and specific activity. On SDS-polyacrylamide gel electrophoresis, both forms yielded one predominant band with an apparent molecular weight of 42 000. The specific activity of the enzyme was dependent on the concentration of the enzyme and on the ionic strength of the dissolving medium. All ions tested stimulated the enzyme activity in the ionic strength range 0-100 mM, with glutamate yielding the highest activity. Above these concentrations, the dehydrogenase showed high tolerance for glutamate in concentrations up to 0.9 M, whereas malate, sulfate and chloride were inhibitory. Enzyme activity showed little sensitivity to the type of cation present and was only slightly affected by 5 M glycerol. The true K m values for the substrates were 6 . 6 / t M for NADH, 130 pM for dihydroxyacetone phosphate, 0.3 mM for NAD and 1.2 mM for glycerol 3-phosphate, and the enzyme showed specificity for these four substrates only. It is proposed that the enzyme functions in cellular osmoregulation by providing glycerol 3-phosphate for the biosynthesis of glycerol, the main compatible solute in D. hansenii, and that the enzyme is well adapted to function in yeast cells exposed to osmotic stress.

Introduction The yeast Debaryomyces hansenii, which has been isolated from saline environments such as sea-water [1] and concentrated brines [2], is one of the few eukaryotic microorganisms that tolerates wide ranges of salt concentrations [3,4]. When exposed to osmotic stress, the yeast produces and accumulates glycerol as the major intracellular compatible solute [5]. The glycerol production increases with increased salinity of the medium [5] and at high salinity the intracellular concentration may reach several molar. The glycerol is assumed to be synthesized via reduction of dihydroxyacetone phosphate to glycerol 3-phosphate by a cytoplasmic NADdependent glycerol-3-phosphate dehydrogenase (EC 1.1.1.8) followed by dephosphorylation of glycerol 3phosphate to glycerol [6].

Correspondence: L. Adler, Department of Marine Microbiology, Botanical Institute, S-41319 Grteborg, Sweden.

The glycerol-3-phosphate dehydrogenase (NAD +) has been purified from many organisms [7-11], and the enzyme appears to play several distinct metabolic roles, such as in the production of glycerol and lipids and in the control of the N A D H / N A D balance via the glycerol 3-phosphate shuttle [12]. This article reports the purification and characterization of the NAD-dependent glycerol-3-phosphate dehydrogenase of the yeast D. hanseniL with the aim of obtaining an increased understanding of the role of this enzyme in the osmotically regulated glycerol production. The influence on enzyme activity of major intracellular solutes in D. hansenii, such as glycerol, K ÷ [13] and glutamate [14] and of other ions was studied and the Michaelis constants were determined.

Materials and Methods Organism and growth conditions. D. hansenii (Zopf) (van Rij strain 26) [1] was cultured and harvested as described previously [15], except that the medium was not supplemented with NaCl. The cells were grown for

0304-4165/90/$03.50 © 1990 ElsevierSciencePublishers B.V. (BiomedicalDivision)

181 20 h in a 15 1 medium which was harvested at an absorbance of 10 (610 nm) giving a total yield of 75 g wet weight of cells. Preparation of cell homogenate. The cells were suspended in 2 vol. ( v / w ) of 30 mM Tris-H2SO4, 2.5 mM Na2EDTA, 5 mM dithiothreitol, 0.05% NaN 3 (pH 8 at 4 ° C) buffer. This suspension was treated with 130 g of 0.5 mm glass beads for 10 rain in a bead mill (Edmund Btihler) at 4 ° C. After addition of phenylmethylsulfonyl fluoride to a final concentration of 0.1 raM, the homogenate was clarified by centrifugation at 79000 × g (rmax) for 30 min at 4°C. Enzyme assays. Enzyme activity was recorded in a Shimadzu UV-240 spectrophotometer at 2 5 ° C in 1 ml cuvettes. The production of oxidized or reduced NAD(P) was followed at 340 nm. The assay buffer was 20 mM imidazole and substrates were 1 mM dihydroxyacetone phosphate and 0.16 mM N A D H (standard conditions) or 10 mM DL-glycerol 3-phosphate and 1 mM NAD. The assay buffer was supplemented with 200 mM potassium glutamate in all experiments except where the influence of various salts and glycerol was studied. Except for experiments where the influence of pH was examined, the final pH (25 ° C) of the reaction mixtures was adjusted to 8.0 (with H2SO4). Measurements were started by the addition to the cuvette of 5-25/~1 enzyme extract, which yielded linear reactions for 1 min under standard conditions at an absorption change of less than 0.1 min -1. One unit of enzyme activity is defined as the amount that converts 1 /xmol of substrate per rain under the prevailing conditions. Kinetic studies. For kinetic measurements in the direction dihydroxyacetone phosphate to glycerol 3-phosphate (forward reaction), dihydroxyacetone phosphate and N A D H were varied within the concentration ranges 0.05-1.5 mM and 2-120 /LM, respectively. The reverse reaction was studied with 1-24 mM DL-glycerol 3-phosphate and 0.15-3 mM NAD. Substrate specificity. To determine the substrate specificity, a standard concentration of the nonvariable substrate was added. Dihydroxyacetone phosphate was replaced by 7-30 mM DL-glyceraldehyde phosphate or 2-30 mM dihydroxyacetone, and N A D H by 40-260 /~M NADPH. Inhibition studies. Product inhibition was studied in the forward reaction using the standard concentration of the nonvariable substrate and both products were tested for inhibition of each substrate. N A D H was varied within the concentration range 0.01-0.16 mM and the inhibitor concentrations within 0 - 8 mM (NAD) or 0-120 mM (DL-glycerol 3-phosphate). The dihydroxyacetone phosphate concentration was varied in the range 0.05-1.5 raM, and the inhibitor concentrations within 0-16 mM (NAD) or 0-80 mM (DL-glycerol 3-phosphate).

Protein determination. Protein concentrations were determined with a Coomassie dye binding technique according to Sedmak and Grossberg [16], using bovine serum albumin as the standard. Purification of the enzyme. All steps were carried out at 4 ° C and the purification procedure was completed within 3 days. Poly(ethylene glycol) precipitation. Poly(ethylene glycol) 4000 was added to the centrifuged homogenate using a 50% ( w / v ) stock solution in buffer A (10 mM Tris-H2SO4, 1 mM Na2-EDTA, 2 mM dithiothreitol, 0.02% NAN3; pH 7.5 at 4 ° C) to a final concentration of 20% (w/v). The mixture was stirred for 30 rain and centrifuged for 15 rain at 48000 x g (rmax). Ion-exchange chromatography 1. The supernatant from above was diluted 1.5-times with buffer A to decrease the viscosity and then applied to a 2 x 5 cm column of DEAE-Sepharose (Pharmacia Fine Chemicals, Sweden) equilibrated with buffer B (buffer A supplemented with 7.5% w / v poly(ethylene glycol) 4000). After application, the column was washed with 3 vol. of equilibration buffer and the enzyme was eluted using a gradient of 2 x 150 ml of 0-0.5 M DL-glycerol 3-phosphate in buffer B (flow rate 10 c m / h ) . The peak activity was eluted at a salt concentration of 0.3 M. Ultrafiltration. Tubes containing glycerol-3-phosphate dehydrogenase activity were pooled and concentrated by ultrafiltration to a final volume of 50 ml (Amicon YM 10 membrane). Gel filtration. Due to the limited capacity of the gel-filtration column, the dehydrogenase pool was processed through the following steps in four portions. Each portion (12.5 ml) was separately applied to a 130 x 2.5 cm column of Ultrogel AcA-34 (LKB, Sweden) equilibrated with buffer C (buffer A supplemented with 10% ( w / v ) poly(ethylene glycol) 4000 and 30 mM (NH4)2SO4) and eluted with buffer C at a flow rate of 5.4 c m / h . The peak activity emerged after more than one column volume, indicating interactions with the gel matrix. Dye affinity chromatography. Active fractions from above were pooled and MgC12 was added to 10 mM concentration. This pool was immediately applied to a 1 x 3.2 cm column of Procion red HE-3B Sephacryl S-200, prepared as described by Boehme et al. [17] and equilibrated in buffer C supplemented with 10 mM MgC12. The addition of MgC12 reduced the time required for binding and increased the capacity of the dye column (cf. Ref. 18). After washing with 3 column volumes of buffer C, a gradient of 2 x 45 ml 0 - 3 mM N A D H in the same buffer was used for elution (flow rate 8.0 cm/h). The peak activity eluted at a N A D H concentration of 1 mM. Ion-exchange chromatography 2. The pooled dye affinity fractions were diluted 1.5-times to decrease the viscosity, filtered through a 0.2/~m filter (Flow Labora-

182 tories) and applied to a 5 x 0.5 cm M o n o - Q column (Pharmacia) equilibrated with buffer A (flow rate 50 c m / h ; Pharmacia FPLC-system). After washing with 5 ml equilibration buffer containing 30 m M DE-glycerol 3-phosphate, the enzyme was eluted with a 25 ml gradient of 3 0 - 2 1 0 m M DL-glycero1 3-phosphate in the same buffer (flow rate 150 c m / h ) . The ultraviolet absorbance of the effluent was recorded continuously at 280 nm, and 1 ml fractions were collected. The activity was eluted in two peaks and poly(ethylene glycol) 4000 was added to the pooled samples to a final concentration of 15% ( w / v ) poly(ethylene glycol) and the purified enzyme was stored at 4 ° C. Polyacrylamide gel electrophoresis. SDS-polyacrylamide gel electrophoresis was performed in a 10-20% gradient gel as previously described [15], using molecular weight markers from Pharmacia (Kit No. 17-044601). The gel was stained with Coomassie blue. Results

Purification (?CAD +)

of

glycerol-3-phosphate

dehydrogenase

A s u m m a r y of the purification procedure is presented in Table I. During the purification of the enzyme it was observed that the specific activity was dependent on the concentration of the enzyme in the storage buffer. The gel-filtration step caused a large loss in purification and yield (Table I), which was partly reversed by increasing the enzyme concentration in the eluate (Fig. 1); a 10-fold increased concentration b y ultrafiltration caused a 3-fold increase in the specific activity. This was not a general effect of increased protein concentration, since the addition of 10 m g / m l of bovine serum albumin to the eluate had no effect on the enzyme activity. However, in the enzyme assay used, the specific activity remained unaffected by the enzyme concentration over a 20-fold concentration range, provided the reaction was started by the addition of the enzyme extract. Initiation of the reaction by the addition of either substrate (dihydroxyacetone phosphate

E ~,

6.0'

v

~,

4.0'

'.g 2.0' d O C/)

0.0 0,0

0.2

0.4

0.6

0.8

protein cone. (mg/ml) Fig. 1. Effect on the specific activity of the D. hansenii glycerol-3phosphate dehydrogenase by concentrating the gel filtration pool by ultrafiltration. or N A D H ) , gave considerably lower activities. These observations indicate that the specific activity remained stable on dilution in the presence of the substrates. In developing the poly(ethylene glycol) precipitation step, it was observed that this substance dramatically increased the stability of the enzyme. At optimal concentration of poly(ethylene glycol) (15-17.5%, w / v ) , the loss of enzyme activity was insignificant ( < 5%) after 6 days at 0 ° C in a crude homogenate. Consequently, poly(ethylene glycol) was included in all buffers except in the last ( M o n o - Q ) step, where the high viscosity would have been an obstacle. It was also found that NaC1 and KCI caused inactivation of the enzyme. Therefore DL-glycero1 3-phosphate, in which solute the enzyme activity remained stable, was used for the elution of the D E A E - S e p h a r o s e and the M o n o - Q columns. The elution profile from the final ion exchange chrom a t o g r a p h y revealed two peaks of enzyme activity (Fig. 2), of which the second peak, which eluted at 130 m M DE-glycerol 3-phosphate, showed the highest specific activity (Table I). Samples from the two peaks of enzyme activity showed protein bands of identical mobility when subjected to SDS-polyacrylamide gel electrophoresis (Fig. 3). The p r e d o m i n a n t c o m p o n e n t of both pools had an apparent molecular weight of 42 000.

TABLE I

Summary of a purification of glycerol-3-phosphate dehydrogenasefrom D. hansenii

Ultracentrifugate PEG 4000 centrifugate DEAE-Sepharose Ultrogel AcA 34" Sephacryl red HE-3B MonoQ pool 1 pool 2

Protein (rag)

Volume (ml)

Total activity (units)

Specific activity (units/mg)

Purification (-fold)

Yield (%)

660 164 40.5 7.65 1.31 0.126 0.066

110 176 150 100 39.4 5.85 7.3

237 197 160 11.4 31.1 8.78 13.2

0.359 1.20 3.95 1.50 23.9 13.4 200

1 3.3 11 4.2 c 66 37.3 557

100 83 68 19 b,c 52 b 15 b 22 b

a The DEAE-pool was processed in four portions due to the limited capacity of the gel-filtration column. Results are presented from one run. b The yield for one of the four portions multiplied by 4. c The purification and yield for the gel-filtration step is underestimated due to reversible loss of activity on dilution (see Results).

183 E

0.50 A

= 0,050

"300

-

$

i

v

= ¢g

0.025'

0.25

-15o

.8

o

!

94 000

w

_e G, 0.000

•

,

10

•

i

0.00

-

20

i

67 000

o

43 000

30

fraction number

Fig. 2. The final purification step of the D. hansenii glycerol-3-phosphate dehydrogenaseby ion-exchangechromatographyon a Mono-Q column. - - , protein concentration; II, enzymeactivity; rn, glycerol 3-phosphate gradient.

Influence of pH, salts and glycerol on enzyme activity The purified enzyme showed maximum activity in the p H interval 7.9-8.2. The influence on enzyme activity of various K ÷- and Mg2÷-salts, is shown in Fig. 4. All salts stimulated the enzyme activity and the maximum activity occurred in the ionic strength range 0.10.3 M. Salts of chlorides and glutamate gave the highest activation (about 2-fold), but chloride salts became strongly inhibitory at higher concentrations, whereas potassium glutamate maintained a high activity also at high ionic strength. Salts of sulfate and malate gave slight activation at low ionic strength and moderate inhibition as the concentration was further increased, MgSO 4 being the most inhibitory. Replacing K + with Na ÷ or N H ~ yielded results similar to what was obtained with the K ÷ salts. Glycerol caused slight inhibition of the enzyme; 60-70% of the control activity remained at 5 M concentration, irrespective of whether the enzyme was assayed in standard buffer or stimulated by 0.2 M potassium glutamate. Enzyme that was partly inhibited by 0.4 M KC1 did not respond to glycerol additions.

30 ooo 20 100

A

B

Fig. 3. SDS-polyacrylamidegel electrophoresis of the Mono-Qpool 1 (A) and 2 (B) from fractions 13-16 and 17-21, respectively, from the final ion-exchange chromatography step (Fig. 2) of the D. hansenii glycerol-3-phosphatedehydrogenase.

Kinetic properties Initial velocity measurements with dihydroxyacetone phosphate and N A D H as variable substrates were indicative of Michaelis-Menten kinetics. Using Dixon as well as Comish-Bowden [19] plots, N A D was found to be a competitive inhibitor to N A D H , whereas DLglycerol 3-phosphate yielded a mixed type of inhibition, as was also the case for both inhibitors when dihydroxyacetone phosphate was the variable substrate. These results and the fact that converging lines were obtained in both Hanes and Lineweaver-Burk plots (not shown) suggested an enzyme mechanism of an ordered ternarycomplex type with N A D H as the first binding substrate [19], allowing for determination of the true K m values from secondary plots (Fig. 5; Table II). The inhibition

/

A .~

A 200

A .~

| 200

~

100'

.--

100 .-

B

1

__>

0

0,0

0,5

ionic strength ( M )

1,0

O" 0.0

0.5

Ionic strength ( M )

1.0

Fig. 4. Influenceof different salts on the activity of D. hansenii glycerol-3-phosphatedehydrogenase. In (A) the salts were: i potassium glutamate; D, KCI; A, MgC12;and in (B) the salts were: D, K2SO4; A, dipotassium malate; I, MgSO4. Data represent mean values of two experiments and the bars indicate S.D.

184 .o10

80

Q.

tion, linear converging lines were obtained in both Hanes and Lineweaver-Burk plots with DL-glycerol 3phosphate or N A D as variable substrates (not shown), indicating a ternary-complex enzyme mechanism. The true K m values were determined from secondary plots as above (Fig. 6; Table II).

B

c

E

>

.o

x

> 40

5

n m

Discussion

E

Z

!

0

D

<

!

!

i

0 50 50 100 NADH concentration ( gM )

100

Fig. 5. Secondary plots [19] for the determination of K m for dihydroxyacetone phosphate and NADH. The values on the ordinate are given as (A) ~M-mM/munits per ml and (B)/tM/munits per ml. constants ( K i ) for N A D and DL-glycerol 3-phosphate were calculated f r o m Dixon plots. U n d e r the conditions used, the enzyme showed no activity when dihydroxyacetone phosphate and N A D H were substituted by dihydroxyacetone, DL-glyceraldehyde phosphate or N A D P H . However, the enzyme displayed activity with DE-glycerol 3-phosphate and N A D as substrates (reverse reaction). As in the forward reacTABLE II Kinetic constants for the glycerol-3-phosphate dehydrogenase from D. hansenii

The K m values are based on triplicate values determined from secondary plots as shown in Figs. 5 and 6. The K i values are based on duplicates determined from Dixon plots. Values presented are means + S.D. Enzyme preparations from the final purification step (Mono-Q pool 2) of two different purifications were used. Km (DL-glycerol3-phosphate) Km (NAD) Km (dihydroxyacetone phosphate) K m (NADH) Ki (DL-glycero13-phosphate) Ki (NAD)

1.2 _+ 0.5 mM 0.30_+ 0.06 mM 130 _+30/~M 6.6 _+ 3/tM 9.0 _+ 1 mM 0.60_+ 0.2 mM

10

.Q 12 cg a. O. m

t,.i

x ¢1

E

>(

>E

5

2

m n n M

O <{ Z w

E

v

x

E ,<

0

i

i

2

4

0

i

i

2

4

NAD concentration ( mM )

Fig. 6. Secondary plots [19] for the determination of Km for glycerol 3-phosphate and NAD. The values on the ordinate are in (A) given as mM. mM/munits per ml and in (B) as mM/munits per ml.

In a preliminary study of the cytoplasmic N A D - d e pendent glycerol-3-phosphate dehydrogenase in cell homogenate of the yeast D. hansenii, it was observed that enzyme activity was unstable in dilute solutions. Hence, the stabilization achieved by the addition of poly(ethylene glycol) 4000 was crucial for the development of a purification procedure. The purification, based on precipitation with (poly(ethylene glycol) and a c o m b i n a t i o n of chromatographic procedures (Table I), yielded one p r e d o m i n a n t protein, which exhibited an apparent molecular weight of 42 000 on SDS-gel electrophoresis (Fig. 3), similar to values reported for the analogous enzyme from other sources [7,20,21]. In the final purification step, the enzyme activity was eluted in two peaks which differed in specific activity and ionic characteristics (Table I, Fig. 2), but showed identical behavior on SDS-gel electrophoresis (Fig. 3). This m a y indicate a limited proteolysis or deamidation of the enzyme to an extent that would be undetectable on the SDS gel or that the native enzyme exists in two forms having different catalytic properties. The specific activity of the enzyme appeared to be dependent on the concentration of the enzyme; the loss of activity in the gel-filtration step (Table I) was partly reversed by increasing the concentration of the eluate by ultrafiltration (Fig. 1), but was unaffected by increasing the protein concentration with bovine serum albumin. It has been reported that the dimeric form of glycerol-3phosphate dehydrogenase of rabbit skeletal muscle, reversibly dissociates into low-active m o n o m e r s u p o n dilution, yielding a changed specific activity with changed enzyme concentration [21]. We suggest that the yeast enzyme m a y function in a similar way. Such behavior would be in agreement also with the observed stabilization of the enzyme by the substrates (cf. Ref. 22) and with the activity-preserving effect of poly(ethylene glycol), a c o m p o u n d k n o w n to stabilize active, multimeric forms of enzyme complexes [22-24]. In the presence of the substrates (assay conditions), the enzyme was stimulated by salt (Fig. 4), an effect that occurred in the ionic strength range 0 to 100 mM. At higher concentrations, the activity decreased again to an extent that depended on the type of anion present. Glutamate, however, was exceptional in causing almost no such inhibition; the maximal activity remained essentially unaffected up to at least 0.9 M potassium glutamate. The stimulation of the glycerol-3-phosphate

185 dehydrogenase by high concentrations of potassium glutamate is of physiological interest, since K ÷ [13] and glutamate [14] are major intracellular ions in D. hansenii [13], which may attain concentrations close to 0.3 M and 0.1, respectively (calculated from Ref. 13 and 14, assuming a water content of 1.8 # l / m g dry wt.; Larsson, C., personal communication). Hence, these ions may reach even higher concentrations in an osmotically dehydrated cell that has to initiate osmoregulation by glycerol production. Potassium glutamate was recently shown to activate trehalose phosphate synthetase in Escherichia coli, a phenomenon suggested to play a role in the osmotically induced trehalose accumulation of this organism [25]. Glycerol, the main inducible osmolyte in D. hansenii [6], demonstrated as with enzymes from other sources [26-28] high compatibility with enzyme function. The dehydrogenase activity was only slightly influenced by molar concentrations of glycerol, at which levels the enzyme has to function in cells which have adapted to highly saline media [5]. It is likely that the biosynthesis of glycerol 3-phosphate by reduction of dihydroxyacetone phosphate is the main function of the D. hansenii dehydrogenase. This direction of the reaction is favoured by the equilibrium of the reaction [29] and by the low K m values for the substrates, being 10-50-fold lower than for the reverse reaction (Table II). The latter reaction, the oxidation of glycerol 3-phosphate, appears to be catalyzed in vivo by a mitochondrial glycerol-3-phosphate dehydrogenase. A mutant of D. hansenii with a defective mitochondrial enzyme was unable to grow on glycerol in spite of a functional cytoplasmic glycerol-3phosphate dehydrogenase [6]. This observation further supports a function of the cytoplasmic enzyme in the synthesis rather than in the degradation of glycerol 3-phosphate. The glycerol-3-phosphate dehydrogenase is located at a branch point in the metabolism, linking the glycolytic sequence to the glycerol pathway. In cells subjected to osmotic stress there is an increased flux towards glycerol [5] and the glycerol-3-phosphate dehydrogenase has some capacity for a regulatory role in the partitioning of the carbon flow. Whereas the low g m for the substrates (Table II) would give the enzyme little sensitivity to changes of the intracellular substrate concentration, the tolerance to high concentrations of potassium glutamate renders the enzyme capacity for remaining active after an osmotic dehydration of the cells. The consequent increase of the intracellular enzyme concentration may also serve to maintain the dehydrogenase in an active form by enhancing homologous interactions between the dehydrogenase molecules (cf. Fig. 1). It is also of interest that the D. hansenii enzyme is only weakly inhibited by its products (Table II), the inhibition by glycerol 3-phosphate being considerably less pronounced than for the E. coli enzyme [7]. Hence, the flux

through the enzyme would show little sensitivity to a transiently expanded glycerol 3-phosphate pool. Thus, the D. hansenii enzyme has characteristics which would sustain an increased flow through the dehydrogenase in a cell which is recovering from an osmotic shock.

Acknowledgments We are most grateful to Dr. Klaus S. Thomson for the help in initiating the purification of this enzyme. We also wish to thank Dr. Patricia Conway for linguistic advice. This work was supported by grants from Erna and Victor Hasselblads Stiftelse.

References 1 Norkrans, B. (1966) Arch. Mikorbiol. 54, 374-392. 2 0 n i s h i , H. (1963) Adv. Food Res. 12, 53-94. 3 Brown, A.D. (1976) Bacteriol. Rev. 40, 803-846. 4 Gould, G.W. and Measures, J.C. (1977) Phil. Trans. R. Soc. B 278, 151-166. 5 Andrr, L., Nilsson, A. and Adler, L. (1988).J. Gen. Microbiol. 134, 669-677. 6 Adler, L., Blomberg, A. and Nilsson, A. (1985) J. Bacteriol. 162, 300-306. 7 Kito, M. and Pizer, L.I. (1969) J. Biol. Chem. 244 (12), 3316-3323. 8 Haus, M. and Wegmann, K. (1984) Physiol. Plant. 60, 283-288. 9 Edgley, M. and Brown, A.D. (1983) J. Gen. Microbiol. 129, 3453-3463. 10 Merkel, J.R., Straume, M., Sajer, S.A. and Hopfer, R.L. (1982) Anal. Biochem. 122, 180-185. 11 Santora, G.T., Gee, R. and Tolbert, N.E. (1979) Arch. Biochem. Biophys. 196 (2), 403-411. 12 Ballas, R.A., Garavelli, J.S. and White, III, H.B. (1984) Evolution 38 (3), 658-664. 13 Norkrans, B. and Kylin, A. (1969) J. Bacteriol. 100, 836-845. 14 Adler, L. and Gustafsson, L. (1980) Arch. Microbiol. 124, 123-130. 15 Nilsson, A., Thomson, K.S. and Adler, L. (1989) Biochim. Biophys. Acta 991,296-302. 16 Sedmak, J.J. and Grossberg, S.E. (1977) Anal. Biochem. 79, 544552. 17 Boehme, H.J., Koppenschlaeger, G., Schulz, J. and Hofmann, E. (1972) J. Chromatogr. 69, 209-214. 18 Clonis, Y.D., Goldfinch, M.J. and Lowe, C.R. (1981) Biochem. J. 197, 203-211. 19 Cornish-Bowden, A. (1979) Fundamentals of Enzyme Kinjetics, Butterworths, London. 20 Chen, S.-M., Trumbore, M.W., Osinchak, J.E. and Merkel, J.R. (1987) Prep. Biochem. 17 (4), 435-446. 21 Batke, J., Asboth, G., Lakatos, S., Schmitt, B. and Cohen, R. (1980) Eur. J. Biochem. 107, 389-394. 22 Reinhart, G.D. and Hartleip, S.B. (1987) Arch. Biochem. Biophys. 258 (1), 65-76. 23 Bosma, H.J., Voordouw, G., De Kok, A. and Veeger, C. (1980) FEBS Lett. 120 (2), 179-182. 24 Miekka, S.I. and Ingham, K.C. (1980) Arch. Biochem. Biophys. 203 (2), 630-641. 25 Gi~iver, H.M., Styrvold, O.B., Kaasen, I. and Strom, A.R. (1988) J. Bacteriol. 170 (6), 2841-2849. 26 Adler, L. (1978) Biochem. Biophys. Acta 522, 113-121. 27 Bostian, K.A. and Betts, G.F. (1978) Biochem. J. 173, 787-798. 28 Jancsik, V. and Keleti, T. (1986) Biochem. Int. 13 (5), 819-826. 29 Bergmeyer, H.U. (1974) Methods of Enzymatic Analysis, 2nd Edn., Vol. 1, p. 468, Verlag Chemie/Academic Press, Weinheim/ New York.